Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks512 from public data
- An overview of multi-modal medical image fusion366
In this review, methods in the field of medical image fusion are characterized by image decomposition and image reconstruction, image fusion rules, image quality assessments, and experiments on the benchmark dataset.
- Pixel convolutional neural network for multi-focus image fusion298
Experimental results demonstrate that the proposed p-CNN is competitive with or even outperforms the state-of-the-art methods in terms of both subjective visual perception and objective evaluation metrics.
- 229
- Union Laplacian pyramid with multiple features for medical image fusion228
Visual and statistical analyses show that the quality of fused image can be significantly improved over that of typical image quality assessment metrics in terms of structural similarity, peak-signal-to-noise ratio, standard deviation, and tone mapped image quality index metrics.
- 196
- 194
- Detecting Generated Images by Real Images178
This work observed that the noise pattern of real images exhibits similar characteristics in the frequency domain, while the generated images are far different, and can perform image authentication by checking whether an image follows the patterns of authentic images.
- Image splicing forgery detection combining coarse to refined convolutional neural network and adaptive clustering160
The proposed C2RNet cascades a coarse convolutional neural network and a refined CNN and extracts the differences in the image properties between un-tampered and tampered regions from image patches with different scales to guarantee a stable detection performance.
- Metabolic profiling studies on the toxicological effects of realgar in rats by 1H NMR spectroscopy140
The toxicological effects of realgar after intragastrical administration were investigated over a 21 day period in male Wistar rats using metabonomic analysis of (1)H NMR spectra of urine, serum and liver tissue aqueous extracts to illustrated the high reliability of NMR-based metabonomic approach on the study of the biochemical effects induced by traditional Chinese medicine.
- 126
- 126
- 125
- 122
- 116
- Image analysis by fractional-order orthogonal moments115
Fractional-order orthogonal moments are not only capable of region-of-interest (ROI) feature extraction but also have potential for image reconstruction and face recognition and have high noise robustness in invariant image recognition.
- Fractional discrete Tchebyshev moments and their applications in image encryption and watermarking104
A novel framework for deriving fractional order DTMs (FrDTMs) by the eigen-decomposition of kernel matrices is proposed in this paper, and some properties of the proposed FrDTMs are analyzed.
- DSAGAN: A generative adversarial network based on dual-stream attention mechanism for anatomical and functional image fusion92
A generative adversarial network based on dual-stream attention mechanism (DSAGAN) for anatomical and functional image fusion, which consumes less fusion time and achieves better objective metrics in terms of Q AG, QEN and QNIQE.
- 88
- 87
- 87
- Brightness and contrast controllable image enhancement based on histogram specification84
Two novel histogram based image enhancement algorithms are proposed that provide the way to control the brightness and contrast of enhanced image by adjusting two parameters.
- 82
- Lossless image compression based on integer Discrete Tchebichef Transform77
Results show that the proposed iDTT algorithm not only has higher compression ratio than iDCT method, but also is compatible with the widely used JPEG standard, and a framework of lossless image compression based on integer DTT is proposed.
- 76
- 72
- MAD‐UNet: A deep U‐shaped network combined with an attention mechanism for pancreas segmentation in CT images72
An improved segmentation model called the multiscale attention dense residual U-shaped network (MAD-UNet) based on a convolutional network and an attention mechanism and multiscales convolution to enrich the contextual information and suppress learning in unrelated areas is proposed.
- PAM-DenseNet: A Deep Convolutional Neural Network for Computer-Aided COVID-19 Diagnosis71
A CNN called dense connectivity network with parallel attention module (PAM-DenseNet), which can perform well on coarse labels without manually delineated infection regions without any manual delineation is proposed.
- Principal Component Analysis based on Nuclear norm Minimization70
A novel robust method, called nuclear norm-based on PCA (N-PCA) is presented, developed under a novel unified framework of PCA to remedy the bias of computing data mean and the low-dimensional representation of a sample both of which are treated as unknown variables in a single model together with projection matrix.
- MyoPS: A benchmark of myocardial pathology segmentation combining three-sequence cardiac magnetic resonance images67
A new task of medical image analysis to perform myocardial pathology segmentation combining three-sequence cardiac magnetic resonance (CMR) images, which was first proposed in the MyoPS challenge, in conjunction with MICCAI 2020 is defined.
- Multimodal medical image fusion via laplacian pyramid and convolutional neural network reconstruction with local gradient energy strategy65
In the fusion results of Alzheimer and Glioma, the disease details are much clearer than those of compared algorithms, which can provide a reliable basis for doctors to analyze disease and make pathological diagnoses.
- Radial shifted Legendre moments for image analysis and invariant image recognition64
A mathematical framework for obtaining the rotation, scaling and translation invariants of these two types of radial shifted Legendre moments is provided and the superiority of the proposed methods in terms of image reconstruction capability and invariant recognition accuracy under both noisy and noise-free conditions is shown.
- 63
- Low-dose chest X-ray image super-resolution using generative adversarial nets with spectral normalization62
Results from quantitative and qualitative evaluations demonstrate that the proposed supervised generative adversarial nets approach delivers more authentic improvement for CXR super-resolution (SR) compared to recent state-of-the-art methods.
- 59
- 59
- D-Net: A dual-encoder network for image splicing forgery detection and localization58
A novel network called dual-encoder U-Net (D-Unet) is proposed for image splicing forgery detection, which employs an unfixed encoder and a fixed encoder that outperformed the other methods in image-level and pixel-level detection, without requiring pre-training or training on a large number of forgery images.
- MSNet: A Multi-Stream Fusion Network for Remote Sensing Spatiotemporal Fusion Based on Transformer and Convolution58
A multi-stream fusion network for remote sensing spatiotemporal fusion based on Transformer and convolution, called MSNet is proposed, which aims to learn the global temporal correlation of the image and uses a convolutional neural network to establish the relationship between input and output.
- 57
- Medical image fusion by combining parallel features on multi-scale local extrema scheme53
Experimental results demonstrate that the proposed image fusion algorithms show the best performances among the other fusion methods in the domain of MRI-CT and MRI-PET fusion.
- Fusion of anatomical and functional images using parallel saliency features52
Experimental results show that the proposed method for fusion of MRI-CBV and SPECT-Tc images and fusion ofMRI-T1 and PET-FDG images successfully presents the pleasing fused medical images with high-spatial-resolution anatomical structural boundaries and high-intensity color detail.
- 51
- 50
- 49
- Spatiotemporal Fusion of Remote Sensing Images using a Convolutional Neural Network with Attention and Multiscale Mechanisms49
The results of the proposed spatiotemporal fusion method based on a convolutional neural network to which the added attention and multiscale mechanisms (AMNet) have richer spatial details and more accurate prediction of temporal changes.
- 49
- 49
- Three-layer medical image fusion with tensor-based features48
A new three-layer (intensity, detail, and base layers) medical image fusion method with differential features for gray and pseudo-color images is proposed and the superiority of the proposed method is demonstrated by subjective and objective evaluation on experimental results.
- 47
- BPGAN: Bidirectional CT-to-MRI prediction using multi-generative multi-adversarial nets with spectral normalization and localization46
A bidirectional prediction is proposed by using multi-generative multi-adversarial nets (BPGAN) for the prediction of any modal from another modal image in paired and unpaired fashion and Experimental results show that BPGAN generates better predictions than recently state-of-the-art methods.
- Single image haze removal based on haze physical characteristics and adaptive sky region detection46
Quantitative comparisons with seven state-of-art algorithms on a variety of real-world haze images demonstrate that the proposed novel single image haze removal approach can remove haze effectively and keep color fidelity better.
- 45
- A multiscale double-branch residual attention network for anatomical–functional medical image fusion45
A new convolutional neural network, which is called the multiscale double-branch residual attention (MSDRA) network, for fusing anatomical-functional medical images and the Feature L1-Norm fusion strategy to fuse the features obtained from the input images is proposed.
- 44
- Integrated fusion framework based on semicoupled sparse tensor factorization for spatio-temporal–spectral fusion of remote sensing images43
A novel integrated spatio-temporal–spectral fusion framework based on semicoupled sparse tensor factorization to generate synthesized frequent high-spectral and high-spatial resolution images by blending multisource observations is proposed.
- Multi-Modal Sensor Medical Image Fusion Based on Multiple Salient Features With Guided Image Filter43
An efficient image fusion algorithm using multiple salient features with guided image filter to prevent the problem of low contrast detail is proposed and can achieve better performance than other fusion methods in the domains of MRI-PET and MRI-SPECT fusion.
- 43
- 42
- 41
- Multiscale receptive field based on residual network for pancreas segmentation in CT images41
This method is not only superior to the other established methods in terms of accuracy and robustness but can also effectively restrain pancreas over-segmentation, under-se segmentation and shape inconsistency between the predicted result and ground truth.
- Efficiency at maximum power of a quantum heat engine based on two coupled oscillators40
The semigroup approach is adopted to model the thermal relaxation dynamics along the two isochoric processes, and the upper bound of efficiency at maximum power (EMP) η* is found to be a function of the Carnot efficiency.
- 39
- 38
- Explicit Krawtchouk moment invariants for invariant image recognition37
Experimental results show that the proposed direct method to derive RST invariants from Krawtchouk moment invariants can significantly improve the performance in terms of recognition accuracy and noise robustness.
- 36
- 36
- Intrinsic Image Decomposition-Based Grey and Pseudo-Color Medical Image Fusion34
Two algorithms based on intrinsic image decomposition to decompose MRI and PET images into its two separate components in the spatial domain are proposed and demonstrate that the proposed fusion methods could improve the values of mutual information by the metrics on the disease database.
- Pose-robust face recognition with Huffman-LBP enhanced by Divide-and-Rule strategy34
A new LBP-like feature extraction is proposed which modifies the code rule by Huffman, and a Divide-and-Rule strategy is applied to both face representation and classification, which aims to improve recognition performance across pose.
- 34
- 32
- Object recognition based on the Region of Interest and optimal Bag of Words model32
The results show that the proposed method for object recognition based on Region of Interest (ROI) and Optimal Bag of Words model can be more accurately recognize images.
- 31
- A Spatiotemporal Fusion Method Based on Multiscale Feature Extraction and Spatial Channel Attention Mechanism30
Experimental results obtained from two datasets show that the proposed spatiotemporal fusion method outperforms existing fusion methods in both subjective and objective evaluations.
- Scale-free heterogeneous cycleGAN for defogging from a single image for autonomous driving in fog30
A Scale-free Heterogeneous CycleGAN (SH-CycleGAN) to utilize unpaired real-word images for boosting image defogging and outperforms state-of-the-art performance and obtains more pleasing visually defogging results.
- Moments and moment invariants in the Radon space30
It is proved that the proposed moments in the Radon space can be represented by a linear combination of classical geometric moments, which means the implementation time of the moments can be significantly reduced and the recognition accuracy can also be increased.
- Detecting Generated Images by Real Images Only29
This paper approaches the generated image detection problem from a new perspective: Start from real images and mapping them to a dense subspace in feature space, so that generated images, regardless of their generative model, are then projected outside the subspace.
- 29
- Remote sensing image super-resolution using cascade generative adversarial nets29
This work proposes a novel SR method for remote sensing images using generative adversarial nets (CGAN) with introduction of content fidelity and scene constraint, which can achieve arbitrary high-time high-quality SR image.
- Say ‘YES’ to Positivity: Detecting Toxic Language in Workplace Communications29
A general and computationally viable taxonomy to study toxic language at workplace based on the taxonomy and analysis on why offensive language and hate-speech datasets are not suitable to detect workplace toxicity are provided.
- Medical image segmentation based on active fusion-transduction of multi-stream features28
A novel medical image segmentation framework, namely AFT-Net, is introduced, in which an attention-based data fusion model is proposed to effectively cooperate with the authors' multi-stream encoder, and an Inception Res-Atrous Convolution block is proposing to collect correlated contextual information in the decoding stage.
- LVC-Net: Medical Image Segmentation with Noisy Label Based on Local Visual Cues28
This work proposes a novel medical image segmentation algorithm based on automatic label error correction that can effectively suppress the influence of label noise by utilizing potential visual guidance during the learning process, thereby generate better semantic segmentation results.
- Studies on Acute Toxicity of Model Toxins by Proton Magnetic Resonance Spectroscopy of Urine Combined with Two-step Cluster Analysis28
The combination of NMR and TSCA can be applied to the detection for the NMR markers of novel toxicities and the prediction for the toxicities of other compounds.
- 27
- RPS‐Net: An effective retinal image projection segmentation network for retinal vessels and foveal avascular zone based on OCTA data27
An effective retinal image projection segmentation network (RPS-Net) that can make full use of the information of 3D data and 2D data to generate segmented images with higher continuity and accuracy is proposed.
- 27
- Face recognition method based on dynamic threshold local binary pattern27
The DTLBP method, proposed by this paper, uses dynamic threshold to encode, and computes the statistic histogram referring to the work in L BP method, which shows that the method is superior to the traditional LBP and LTP method.
- 26
- CIFG-Net: Cross-level information fusion and guidance network for Polyp Segmentation26
This work proposes a polyp segmentation network based on cross-level information fusion and guidance, which has better robustness and more accurate segmentation effect, and proposes an information guidance module (IGM) to integrate the processed features of EFPM and CIPM to obtain accurate positioning and segmentation of polyps.
- Knowledge-driven multi-graph convolutional network for brain network analysis and potential biomarker discovery26
The biomarkers identified by the KMGCN model align well with recent neuroscience research, indicating the effectiveness of the model in brain network analysis and potential biomarker discovery.
- Neural ordinary differential grey model and its applications26
The comparison results from predicting China’s annual crude oil consumption and forecasting oilfield production in northern China show that the prediction accuracies achieved by NODGM model are 28% and 8% higher, respectively, than those achieved by the state-of-the-art grey forecasting models.
- 26
- 25
- 25
- 24
- 24
- 24
- Multimodal biomedical image fusion method via rolling guidance filter and deep convolutional neural networks24
A multimodal biomedical image fusion method via rolling guidance filter and deep convolutional neural network is proposed in this paper that has better fusion results and objective metrics than the reference fusion methods.
- LRFNet: A real-time medical image fusion method guided by detail information23
A lightweight residual unit is designed to replace the conventional residual convolution in the model, thereby improving the fusion efficiency and reducing the overall model size by approximately 5 times.
- 23
- 23
- A Multi-Cooperative Deep Convolutional Neural Network for Spatiotemporal Satellite Image Fusion23
Results prove that MCDNet can effectively reconstruct higher-quality prediction images and the new compound loss function can help multinetwork models better complete cooperative training.
- Hyperspectral Image Superresolution Using Global Gradient Sparse and Nonlocal Low-Rank Tensor Decomposition With Hyper-Laplacian Prior23
A novel global gradient sparse and nonlocal low-rank tensor decomposition model with a hyper-Laplacian prior for hyperspectral image (HSI) superresolution is presented to produce a high-resolution HSI (HR- HSI) by fusing a low-resolution LSI (LR-HSI), and an alternating direction method of multipliers-based algorithm is designed to efficiently solve the optimization problem.
- 22
- 22
- A Pseudo-Siamese Deep Convolutional Neural Network for Spatiotemporal Satellite Image Fusion22
A pseudo-Siamese deep convolutional neural network (PDCNN) for spatiotemporal fusion is proposed and the subjective and objective results demonstrate that the PDCNN can effectively reconstruct the fusion image with higher quality.
- 22
- 22
- Medical image fusion based on hybrid three-layer decomposition model and nuclear norm22
Subjective and objective experiments confirm that the proposed algorithm has a distinct advantage compared with other state-of-the-art algorithms.
- 22
- 22
- 21
- HS-Vectors: Heart Sound Embeddings for Abnormal Heart Sound Detection Based on Time-Compressed and Frequency-Expanded TDNN With Dynamic Mask Encoder21
Experimental results demonstrate that the proposed HS-Vectors method excels the state-of-the-art models in abnormality detection.
- 21
- A focus measure in discrete cosine transform domain for multi-focus image fast fusion21
The experimental results demonstrate that the proposed method takes at least 1 % less fusion time than the existing fastest methods by neural network while ensuring the fusion effect in both visual perception and objective evaluations.
- Algebraic Multi-Grid Based Multi-Focus Image Fusion Using Watershed Algorithm21
Experimental results demonstrate that the AMG-based method outperforms the state-of-the-art fusion approaches in terms of selected objective image quality assessments.
- Geometry Structure Preserving Based GAN for Multi-Pose Face Frontalization and Recognition20
The proposed GSP-GAN can generate high-quality frontal images under arbitrary pose, and get satisfactory recognition performance, compared with the state-of-the-art models on datasets of Multi-PIE, LFW and CFP.
- Adaptive remote-sensing image fusion based on dynamic gradient sparse and average gradient difference20
The measured value of the proposed model is used in the iterations of the fusion algorithm as a feedback to adaptively adjust the algorithm to improve the quality of the fused results.
- 19
- 19
- New conception for the development of hypertension in preeclampsia19
The placental endothelium contributed very poorly to vasodilatation, and altered vascular functions in preeclampsia mainly occurred in VSMCs instead of endothelial cells, indicating that endothelial-derived nitric oxide (NO) was extremely weak in placental vessels.
- PGNet: Positioning Guidance Network for Semantic Segmentation of Very-High-Resolution Remote Sensing Images18
A novel semantic segmentation framework for VHR remote sensing images, called PGNet, which consists of the feature extractor, a positioning guiding module (PGM), and a self-multiscale collection module (SMCM), which can effectively extract multi-scale information and generate high-resolution feature maps with high-level semantic information.
- Recent advancement in haze removal approaches18
This review paper explores the various methods for efficiently removing the haze and short comings of the earlier presented techniques used in the revolutionary era of image processing applications to provide a step-by-step approach to the complex solution of the ill-posed inverse problem.
- Multi-granularity generative adversarial nets with reconstructive sampling for image inpainting18
This paper presents an image inpainting method based on GANs with reconstructive sampling and multi-granularity generative adversarial strategy, which is adopted to make inpainted image more continuous and realistic in texture structure and vision, respectively.
- Adaptive illumination-invariant face recognition via local nonlinear multi-layer contrast feature18
This paper first determines a dynamic thresholds strategy including the information of illumination variation to extract nonlinear multi-layer contrast features to improve the discriminative ability of the face image by enlarging the margin between different classes relative to the same class.
- Two‐scale image decomposition based image fusion using structure tensor18
A two‐step model for PET/SPECT image decomposition is proposed to achieve contrast enhancement that inherently captures gradient to distinguish individual edge information from detail information and makes a comparison of fusion result by two‐scale image fusion method in terms of the full‐ reference objective metric structural similarity and no‐reference objective metric natural image quality evaluator.
- The Recognition of the Point Symbols in the Scanned Topographic Maps18
A new algorithm, which is named shear line segment GHT (SLS-GHT), to recognize the point symbols directly in the scanned topographic maps, based on the framework of generalized Hough transform (GHT).
- Myocardial Pathology Segmentation of Multi-modal Cardiac MR Images with a Simple but Efficient Siamese U-shaped Network17
A simple but efficient U-shaped network, named Siamese U-Net, is proposed to solve the problems of segmentation of multi-modal myocardial pathology images, and is able to fully leverage complementary information between different modalities.
- 17
- 17
- Dual-modality visual feature flow for medical report generation16
Dual-Modality Visual Feature Flow (DMVF) is proposed, which introduces region-level features based on grid-level features to enhance the method's ability to identify lesions and key areas and enhances two types of feature flows based on their attributes to prevent the loss of key information.
- 16
- 16
- Multiple herbicide resistance in a Cyperus difformis population in rice field from China16
The GX-35 population has indeed evolved multiple herbicide resistance in China, and a diverse range of weed control tactics should be implemented in rice field.
- 16
- 16
- 16
- 16
- Multimodal medical image fusion by cloud model theory16
A novel multimodal medical image fusion method by cloud model theory that fits the histograms of input images using the high-order spline function firstly and then divides intervals in line with the valley point of the fitted curve to achieve the fused image.
- 16
- An edge detection algorithm based on the multi-direction shear transform16
A new edge detection method that works efficiently on the images, and the continuity of the edge is getting better, and it is able to distinguish the real edges from the noise.
- 15
- 15
- 15
- AEP-GAN: Aesthetic Enhanced Perception Generative Adversarial Network for Asian facial beauty synthesis15
A new method called aesthetic enhanced perception generative adversarial network (AEP-GAN) is proposed, which addresses the over-beautification problem and achieves excellent results.
- 15
- IoU-guided Siamese region proposal network for real-time visual tracking15
An IoU-guided Siamese RPN (SiamIG) that predicts the intersection-over-union (IoU) with the ground truth for each regressed anchor by using an IoU predictor to address the low correlation between the classification score and localization accuracy in tracking.
- 15
- 15
- Synergistic integration of graph-cut and cloud model strategies for image segmentation15
Experiments on different data sets including natural images from Berkeley database, synthetic data, and medical images suggest that the proposed method based on cloud model and graph cuts outperforms other state of theart approaches.
- Self-Supervised Image Local Forgery Detection by JPEG Compression Trace14
This paper firstly analyzed the JPEG compression traces which are mainly caused by different JPEG compression chains, and designed a trace extractor to learn such traces, and utilized the trace Extractor as the backbone and trained self-supervised to strengthen the discrimination ability of learned traces.
- 14
- 13
- 13
- Errata and comments on “Orthogonal moments based on exponent functions: Exponent-Fourier moments”13
The IEFMs avoid the imprecisions of EFMs and performs much better than EFMs, and experimental results show the superiority of the proposed methods.
- Osiris: A Malware Behavior Capturing System Implemented at Virtual Machine Monitor Layer13
This paper study and design a system called Osiris, which makes use of virtual machine technique to capture malware behavior and monitors Windows API calls invoked by the process under analysis (or target program) to rebuild its behaviors.
- DM-FNet: Unified Multimodal Medical Image Fusion via Diffusion Process-Trained Encoder-Decoder12
A two-stage diffusion model–based fusion network (DM-FNet) to achieve unified MMIF is proposed, demonstrating that the proposed method performs exceptionally well regarding objective evaluation metrics and preserves appropriate brightness, a comprehensive distribution of radioactive tracers, rich textures, and clear edges.
- 12
- Gesture-ProxylessNAS: A Lightweight Network for Mid-Air Gesture Recognition Based on UWB Radar12
The Gesture-ProxylessNAS (GPNAS) is presented to enhance the adaptability of model search and overcome the challenge of the network's computational complexity, and revisits the multilayer ShuffleNet's design spaces using an optimization problem, greatly reducing thenetwork's parameters and computational complexity.
- 12
- 12
- YOLOSA: Object detection based on 2D local feature superimposed self-attention12
A novel self-attention module, called 2D local feature superimposed self-Attention, is proposed for the feature concatenation stage of the neck network, which reflects global features through local features and local receptive fields and achieves SOTA results.
- 12
- 12
- 12
- 12
- Hyperspectral Image Super-Resolution via Sparsity Regularization-Based Spatial-Spectral Tensor Subspace Representation11
This article proposes a sparsity regularization-based spatial-spectral tensor subspace representation (SR-SSTSR) approach for HSI-SR, which accurately depicts the spectral correlations and spatial self-similarities, as well as their distinctions, and confirms the superiority of the proposed method.
- 11
- Enhanced Multi-Stream Remote Sensing Spatiotemporal Fusion Network Based on Transformer and Dilated Convolution11
The existing multi-stream remote sensing spatiotemporal fusion network MSNet is improved using dilated convolution and an improved transformer encoder to develop an enhanced version called EMSNet, and a new weight strategy is used for fusion that substantially improves the prediction accuracy of the model, image quality, and fusion effect.
- Multi-Focus Image Fusion Method Based on Multi-Scale Decomposition of Information Complementary11
The proposed multi-focus image fusion method can more accurately distinguish the focused and non-focused areas in the case of image pre-registration and unregistration, and the subjective and objective evaluation indicators are slightly better than those of the existing methods.
- Identification of a novel mutation in the KITLG gene in a Chinese family with familial progressive hyper- and hypopigmentation11
A novel mutation in KITLG, c.104A > T, that was first reported in FPHH within the conserved 35 N motif is identified and strengthened the understanding of F PHH and expand the mutational spectrum of the KIT LG gene.
- MDECNN: A Multiscale Perception Dense Encoding Convolutional Neural Network for Multispectral Pan-Sharpening11
To produce high-quality pan-sharpened images, a multiscale perception dense coding convolutional neural network (MDECNN) is proposed, based on dual-stream input, which is superior to the existing methods in both subjective and objective evaluations.
- A Lightweight SE-YOLOv3 Network for Multi-Scale Object Detection in Remote Sensing Imagery11
A lightweight one-stage approach based on YOLOv3 is proposed in this paper, which is named Squeeze-and-Excitation YOLov3 (SE-YOLO v3), which achieves more competitive detection effect with less computational consumption.
- 11
- A level set image segmentation method based on a cloud model as the priori contour11
A novel image segmentation method combining a cloud model and a level set (CM-LS) is proposed, which uses the rough segmentation result of cloud model as the initial contour of the level set and then obtains the final result by the contour evolution.
- CustomTTT: Motion and Appearance Customized Video Generation via Test-Time Training10
This work first analyzes the prompt influence in the current video diffusion model and finds the LoRAs are only needed for the specific layers for appearance and motion customization, and proposes CustomTTT, where they can joint custom the appearance and the motion of the given video easily.
- 10
- Deep Lifelong Cross-Modal Hashing10
This paper proposes a novel deep lifelong cross-modal hashing to achieve lifelong hashing retrieval instead of re-training hash function repeatedly when new data arrive, and introduces enhanced-semantic similarity to supervise hash learning and proves that the similarity improves performance with detailed analysis.
- YTHDF3 modulates the Cbln1 level by recruiting BTG2 and is implicated in the impaired cognition of prenatal hypoxia offspring10
The results suggest that abnormal cognition in PH offspring may be related to the dysregulation of the m6A pathway, specifically increased levels of YTHDF3 in the hippocampus.
- 10
- 10
- 10
- A Fast Low-Cost Shortest Path Tree Algorithm10
The algorithm is based on the low-cost shortest path tree algorithm DDSP and through improving on the search procedure, a Fast Low-cost Shortest Path Tree (FLSPT) algorithm is presented in this paper.
- MCU-Net: A multi-prior collaborative deep unfolding network with gates-controlled spatial attention for accelerated MRI reconstruction9
This study designs a gates-controlled spatial attention module (GSAM), evaluating the relative confidence and overall confidence maps for intermediate reconstructions produced by different subnetworks, and designs correction modules to enhance the effectiveness in regions where both subnetworks exhibit limited performance.
- 9
- 9
- MTSNet: Joint Feature Adaptation and Enhancement for Text-Guided Multi-view Martian Terrain Segmentation9
A novel multi-view Martian Terrain Segmentation framework (MTSNet) is proposed by developing an efficient Martian Terrain text-Guided Segment Anything Model (MTG-SAM) and combining it with a tailored Local Terrain Feature Enhancement Network (LTEN) to capture intricate terrain details.
- 9
- 9
- 9
- 9
- 9
- MEDL‐Net: A model‐based neural network for MRI reconstruction with enhanced deep learned regularizers9
To improve the MRI reconstruction performance of model‐based networks and to alleviate their large demand for GPU memory, a new approach is proposed to combine supervised and unsupervised approaches.
- Convolution neural network with edge structure loss for spatiotemporal remote sensing image fusion9
This paper proposes an edge structure loss, which is added to a spatiotemporal fusion network without pre training model, which enables the network to perform spatiotmporal fusion with better generalizability and robustness and achieves a higher accuracy in terms of quality evaluation indicators and better visual effects.
- 9
- MDCNN: multispectral pansharpening based on a multiscale dilated convolutional neural network9
Experimental results using three satellite image datasets show that the proposed pansharpening algorithm generates high-quality fusion images, and it outperforms most advanced panshARPening methods in subjective visual and objective evaluation indexes.
- Methylation-reprogrammed CHRM3 results in vascular dysfunction in the human umbilical vein following IVF-ET9
It is revealed that the hypermethylation of CHRM3, leading to a reduction inCHRM3 expression and downregulation of the CaV1.2/PKC-MLC20 phosphorylation pathway, was responsible for the decreased sensitivity to ACh observed in the umbilical vein under IVF-ET conditions.
- A contour-aware feature-merged network for liver segmentation based on shape prior knowledge9
A bidirectional convolutional long short term memory (BiConvLSTM) is proposed to explore contextual information and slightly outperforms other newly published methods and achieves good performance for liver segmentation.
- Automatic Segmentation of Liver from CT Scans with CCP–TSPM Algorithm9
With the increase in the morbidity of liver cancer and its high mortality rate, liver segmentation in abdominal computed tomography (CT) scan images has received extensive attention.
- Gesture recognition based on Global Template DTW for Chinese Sign Language9
This paper proposes global template dynamic time warping (GTDTW) algorithm for gesture recognition with the wearable gloves and a gesture segmentation system based on GTDTW, which has high segmentation rate and recognition rate and has a better real-time performance.
- 9
- Three novel invariant moments based on radon and polar harmonic transforms9
Three novel invariant moments on the basis of radon and Polar Harmonic Transforms are presented, which indicate that the classification accuracy of the method is higher and with lower time cost than that of other moments.
- 9
- 8
- ShiftMorph: A Fast and Robust Convolutional Neural Network for 3D Deformable Medical Image Registration8
This work proposes losslessly downsampling by shifting the strided convolution into a lower-dimensional space, significantly reducing decoding overhead and improving the deformation quality and enables halving the proposed network after training.
- 8
- 8
- Hyperspectral Image Super-Resolution via Adaptive Factor Group Sparsity Regularization-Based Subspace Representation8
A novel factor group sparsity regularized subspace representation (FGSSR)-based method for HSI super-resolution is proposed that can simultaneously address these issues encountered in previous methods and outperforms several state-of-the-art fusion methods with significant improvements.
- 8
- 8
- 8
- Multi-Stage Liver Segmentation in CT Scans Using Gaussian Pseudo Variance Level Set8
A Multi-stage framework for location and segmentation for liver segmentation is presented and the Gaussian mixture model-based signed distance function is proposed to increase the flexibility of shape prior models.
- Model-based variational pansharpening method with fast generalized intensity–hue–saturation8
A method is proposed, which aims to combine the fast generalized IHS (FGIHS) fusion technique based on spectral adjustment with a variational optimization model to estimate a more accurate intensity image, which improves the spatial quality of the fusion results in the view of the subjective and objective evaluations.
- 8
- An Optimized Quantization Technique for Image Compression Using Discrete Tchebichef Transform8
An efficient lossy compression based DTT to produce better quality reconstructed image for the desired compression ratio is proposed and soft decision quantization (SDQ) is combined to design optimal quantization table and to approximate the rate-distortion for the purpose of the reconstruction quality.
- Image Fusion Based on Shearlets8
Wavelet transforms are more useful than Fourier transforms, and it is efficient in dealing with one-dimensional point-wise smooth signal, but the limitations of the direction make it not perform well for multidimension‐ al data.
- An algorithm on fairness verification of mobile sink routing in wireless sensor network8
This paper focuses on verifying the fairness of mobile sink routing based on both state and action, which is realized mainly by composing Labeled Kripke Transition Systems (LKTS).
- 8
- Anisotropic Filter Based Modified Canny Algorithm8
The comparison between the old method and the proposed method proves that the proposed algorithm has advantage in the ability to detect edges and suppress noises, and the change of parameters has a slight effect on the proposed algorithms.
- 7
- Semi-Supervised FMCW Radar Hand Gesture Recognition via Pseudo-Label Consistency Learning7
This work proposes a semi-supervised learning framework guided by pseudo-label consistency, which utilizes a dual-branch structure with a mean-teacher network to alleviate accuracy degradation resulting from individual differences and interference from other body parts, thereby bolstering the network’s robustness.
- 7
- DPT‐tracker: Dual pooling transformer for efficient visual tracking7
A dual pooling transformer tracking framework, dubbed as DPT, which consists of a simple yet efficient spatiotemporal attention model (SAM), a mutual correlation pooling Transformer (MCPT), a mutual correlation pooling Transformer (MCPT) and a multiscale aggregation pooling Transformer (MAPT).
- 7
- Proteomic analysis reveals the adaptation of Vibrio splendidus to an iron deprivation condition7
Basic information on the global protein profiles of V. splendidus grown under iron deprivation condition is given and it is suggested that theIron deprivation condition cause the cell growth enter a state of higher cell density earlier.
- 7
- 7
- Multi-Focus Color Image Fusion Based on Quaternion Multi-Scale Singular Value Decomposition7
Simulation results show that image fusion using the quaternion multi-scale singular value decomposition method achieves great overall visual effects, with high resolution images, rich colors, and low information loss.
- Multilevel dense neural network for pan-sharpening7
A novel neural network for pan-sharpening that uses a multilevel shortcut link and dense block to improve the depth of the neural network and is superior to the state-of-the-art methods in both subjective visual and objective assessments is proposed.
- Anatomical‐functional image fusion based on deep convolution neural networks in local Laplacian pyramid domain7
The experimental results prove that the fused image via the proposed method has the characteristics of high contrast, high brightness, and high color saturation and great advantages in retaining the color and structural information.
- Evaluation method and reduction measures for the flicker effect in road lighting using fixed Low Mounting Height Luminaires7
The reduction in the brightness of the light source is investigated, hence reducing the energy of the visual stimulation signal to the human eye in order to judge the degree of fatigue in human vision.
- Object recognition based on the region of interest and optical bag of words model7
A method for object recognition based on Region of Interest (ROI) and optimal Bag of Words model (BOW) and the results show that the recognition accuracy of the proposed method is better than the traditional object recognition methods.
- 6
- 6
- 6
- An elastic competitive and discriminative collaborative representation method for image classification6
This work proposes a novel CR approach for image classification called the elastic competitive and discriminative collaborative representation-based classifier (ECDCRC) to simultaneously strengthen representation contribution and discrimination of the correct class.
- 6
- 6
- Multi-scale adversarial learning with difficult region supervision learning models for primary tumor segmentation6
A cascade approach that combines multi-scale adversarial learning and difficult supervision to achieve precise tumor segmentation is introduced and confirms that the combination can improve the segmentation performance, especially for small objects.
- Vibration Control with Reinforcement Learning Based on Multi-Reward Lightweight Networks6
A reinforcement learning method using a deep residual shrinkage network based on multi-reward priority experience playback for high-frequency and high-dimensional continuous vibration control based on multi-reward priority experience playback for high-frequency and high-dimensional continuous vibration control.
- 6
- 6
- MFGAN: Multi-modal Feature-fusion for CT Metal Artifact Reduction Using GANs6
This work proposes a novel method based on generative adversarial nets (GANs) to reduce metal artifacts by encoding interactive information and imaging CT to yield multi-modal feature-fusion representation, which overcomes representative ability limitation of single- modal CT images.
- SCRNet: an efficient spatial channel attention residual network for spatiotemporal fusion6
This work designs an efficient and flexible spatial channel attention mechanism that not only focuses on spatial information but also takes channel information into account and designs an optimized residual network that can further enhance the ability of the network to extract feature information.
- 6
- 6
- 6
- Investigation on the acute biochemical effects of light rare earths on rat serum by NMR-based metabonomic approaches6
1H-NMR spectroscopy and pattern recognition method were used to assess the acute biochemical effects of light rare earths and exhibited a higher toxicity than La(NO3)3 at the same doses, which may mean that high-dosage of La and Ce impair a specific region of liver.
- 5
- 5
- 5
- CS-CoLBP: Cross-Scale Co-occurrence Local Binary Pattern for Image Classification5
Empirical evaluations demonstrate that CS-CoLBP outperforms the existing state-of-the-art LBP-based methods even in the presence of geometric transformations and image manipulations.
- DeHier: decoupled and hierarchical graph neural networks for multi-interest session-based recommendation5
A multi-interest session-based recommender system named DeHier, which consists of three parts: decoupled item global embedding, multi-interest item embedding, and next-clicked item prediction, which achieves up to 4.12% and 7.43% improvements on the LastFM dataset.
- 5
- 5
- Hierarchical neural network with efficient selection inference5
This paper proposes to use category hierarchies to integrate ResNet-style modules to form a hierarchical network model in a top-down manner, and adopts residual block selection based on coarse categories to allocate different computation paths.
- 5
- 5
- 5
- Facial Expression Recognition Based on Deep Binary Convolutional Network5
A lightweight convolution network (BRNet) with parallel operation of binary convolution and traditional convolution is designed to reduce the complexity of network and improve the speed of recognition.
- Medical image fusion based on local Laplacian decomposition and iterative joint filter5
Experimental results indicated that the proposed method for integrating the information contained in functional and anatomical medical images has a distinct advantage over existing state‐of‐the‐art algorithms in terms of vision quality and objective metrics.
- Dual-Branch Remote Sensing Spatiotemporal Fusion Network Based on Selection Kernel Mechanism5
A dual-branch remote sensing spatiotemporal fusion network based on a selection kernel mechanism that combines the reconstructed fine-difference image with the known fine image is proposed and objective evaluations validate the superiority of the method.
- 5
- A Pansharpening Generative Adversarial Network with Multilevel Structure Enhancement and a Multistream Fusion Architecture5
Experiments show that the proposed pansharpening generative adversarial network with multilevel structure enhancement and a multistream fusion architecture is superior to state-of-the-art methods in both the subjective visual and objective assessments of fused images, especially in road and building areas.
- A generative adversarial network with structural enhancement and spectral supplement for pan-sharpening5
A novel generative adversarial network is proposed for pan-sharpening, which utilizes the supplemented spectral information from low-resolution multi-spectral images and the enhanced structural information from panchromatic images to generate high- resolution multi-Spectral images.
- 5
- Pan-sharpening with a spatial-enhanced variational model5
The proposed variational pan-sharpening method is proven to enhance the spatial quality of the fused result and demonstrate stable performance in preserving the spectral information.
- 5
- 4
- 4
- 4
- Semi-supervised medical image segmentation method using multi-scale consistency adversarial learning4
- 4
- 4
- 4D foetal cardiac ultrasound image detection based on deep learning with weakly supervised localisation for rapid diagnosis of evolving hypoplastic left heart syndrome4
A cost‐effective deep learning framework for rapid diagnosis of eHLHS (RDeH), which is named RDeH‐Net and demonstrates visually fine detection performance with weakly supervised localisation that could be used to accelerate the diagnosis of eHLHS.
- 4
- A Swin Transformer with Dynamic High-Pass Preservation for Remote Sensing Image Pansharpening4
A novel Swin-transformer-based pansharpening model named SwinPAN is proposed, and a detail reconstruction network (DRNet) is developed in an image difference and residual learning framework to reconstruct the high-resolution detailed information from the original images.
- 4
- A location-aware siamese network for high-speed visual tracking4
A novel context enhancement module (CEM) is proposed, which contributes to capturing distinguished object information from both the local and the global levels and Distance-IoU loss is employed to guide the tracker to obtain a more accurate estimation of the target position.
- 4
- 4
- MDCwFB: A Multilevel Dense Connection Network with Feedback Connections for Pansharpening4
This work proposes a multilevel dense connection network with a feedback connection that achieves significant gains on the multiple objective index values used to measure the spectral quality and spatial details of the generated image, namely spectral angle mapper (SAM), relative global dimensional synthesis error (ERGAS), and structural similarity (SSIM).
- Academic Paper Recommendation Method Combining Heterogeneous Network and Temporal Attributes4
The method HNTA for academic paper recommendation based on the combination of heterogeneous network and temporal attributes, which can not only comprehensively utilize both relationships of scholars and the content information of papers, but also it considers the impact of the temporal weight of scholars' research interests.
- Video saliency detection based on low-level saliency fusion and saliency-aware geodesic4
A strategy-based low-level saliency fusion that guarantees to complementarily employ the smoothed color saliency, spatial edges, and temporal motion boundaries clues toward producing high-accuracy low- level saliency is proposed.
- 4
- 4
- Color perception of diffusion tensor images using hierarchical manifold learning4
The experimental results demonstrate that the proposed algorithm can reduce the computation complexity from O ( N 3 ) (based on algorithms in the literature (Ghassan et al., 2011 [9])) to O (N 2 ) and highlight the different regional structures of the brain via color perception of variation.
- 4
- Visual saliency detection based on gradient contrast and color complexity4
The result shows that the proposed algorithm can accurately extract the salient regions, yielding higher precision and better recall rates, and it does better in highlighting the internal details of salient regions.
- Human action recognition based on dense of spatio-temporal interest points and HOG-3D descriptor4
The spatio-temporal interest points and HOG-3D descriptors have been used for human action recognition and combination of both has not been studied in earlier research.
- Online Burning Material Pile Detection on Color Clustering and Quaternion based Edge Detection in Boiler4
This paper presents an online machine vision system for automatically measuring and detecting the burning material amount inside a burner or a boiler and shows that the proposed algorithm works satisfactorily.
- Intrusion detection method based on rough set and genetic reduction algorithm4
An approach to intrusion detection based on rough set and genetic reduction algorithm that employs the genetic algorithm and improved greedy algorithm to find a possibly short reduction and can obtain outstanding results on intrusion detection.
- 4
- 4
- 3
- 3
- 3
- DVAP-Reg: Dual-view anatomical prior-driven cross-dimensional registration for spinal surgery navigation3
A dual-view anatomical prior-driven cross-dimensional registration method for spinal surgery navigation that achieves real-time 2D-3D image-assisted navigation with rotational and translational accuracy, demonstrating its ability to keep pace with the time-sensitive, ever-changing nature of intraoperative workflows.
- 3
- 3
- 3
- Multi-scale Consistency Deep Lifelong Cross-modal Hashing3
A regularization-based lifelong learning module is introduced to balance plasticity for learning new knowledge and stability for maintaining old knowledge and update incremental hash codes without retraining cumulative data, and multi-scale consistency network which employs multi-scale feature fusion module is introduced to learn multi-level semantic representations with consistency.
- 3
- High-Performance Fine Defect Detection in Artificial Leather Using Dual Feature Pool Object Detection3
YOLOD demonstrated outstanding performance on the artificial leather defect dataset, achieving an impressive increase of 11.7% - 13.5% in AP_50 compared to YOLOv5, along with a significant reduction of 5.2% in the error detection rate.
- 3
- 3
- A Novel Mode Selection-Based Fast Intra Prediction Algorithm for Spatial SHVC3
A novel Mode Selection-Based Fast Intra Prediction algorithm for SSHVC that can significantly improve the coding speed with negligible coding efficiency loss and the classic Gaussian Mixture Model and Expectation Maximization in machine learning is proposed.
- Autism spectrum disorder diagnosis based on deep unrolling-based spatial constraint representation3
A new approach to FBNs construction is proposed called the deep unrolling-based spatial constraint representation (DUSCR) model and it is integrated with a convolutional classifier to create an end-to-end framework for ASD recognition.
- A deep unrolling pansharpening method based on spectral consistency and double spatial priors3
This work proposes a neural network pansharpening method with interpretable deep spatial detail injection based on the assumption of spectral consistency and double spatial detail priors in HRMS images, and achieves an optimal balance between spectral and spatial quality.
- 3
- 3
- BH2I-GAN: Bidirectional Hash_code-to-Image Translation using Multi-Generative Multi-Adversarial Nets3
- Extraction‐and‐excitation deep neural network for pansharpening3
The neural network is modified with domain knowledge in pansharpening to fully extract spectral and spatial structures, and the proposed method outperforms traditional methods.
- 3
- Multi-focus image fusion method based on discrete Tchebichef transform and focus measure3
The proposed multi-focus image fusion method based on the discrete Tchebichef transform and a focus measure can reduce the fusion time while ensuring the fusion effect, and exhibit high noise robustness during image fusion.
- 3
- 3
- Multimodal sensor medical image fusion based on mutual‐structure for joint filtering using sparse representation3
The proposed image fusion method based on mutual‐structure for joint filtering and sparse representation shows the best performance on preserving detail information and contrast information in the views of subjective and objective evaluations.
- AN ADAPTIVE FUZZY FUSION FRAMEWORK FOR FACE RECOGNITION UNDER ILLUMINATION VARIATION BASED ON LOCAL MULTIPLE PATTERNS3
It is found that the LMP features of different layers can be used to recognize face independently so that it would preserve more abundant recognition information, and the contribution of the L MP features from different layers is blurry under varying illumination.
- Video-based real-time measurement for human body height3
A novel algorithm for real-time human height measurement without knowing any camera parameter, just having a vertical reference height is proposed and the optimal number and position of the human body in a camera field is shown.
- 3
- Face detection method based on multi-skin color space and AdaBoost algorithm3
The proposed human face detection method based on multil-skin color space detection and AdaBoost algorithm can handle the problem of false and missing better on face detection of color image under the complex background, and improve efficiency of face detection.
- Image matching method based on LTP features3
Due to solve the problem that image matching was very difficult when illumination condition was changeable, proposed an image matching method based on LTP(local ternary patterns) features of image that is more efficient and robust than SIFT.
- Design and Implementation of SSL VPN3
Introduces three kinds of realization of the SSL VPN and gives out a new kind of the framework design and the realization of an SSL VPN server which uses the tunnel mode and brings the computer into the Internet without changing any applications.
- Target classification method for situation assessment based on fuzzy sets3
A method based on fuzzy equivalence relation is applied to implement target classification and a synthetic algorithm is presented to fulfill multi-layer structure among groups by using the nearest-neighbor method and field knowledge.
- 2
- 2
- 2
- A spatiotemporal fusion method based on two-stream high temporal sensitive convolutional neural network2
The Two-Stream High Temporal Sensitive Convolutional Neural Network (TSCNN) is proposed, which introduces the Long-Term Dynamic Capture module (LTDC), which incorporates the temporal variation information extracted from coarse-resolution images into fine-resolution images, enhancing the weight of post-prediction images and preserving the integrity of temporal information, thus reducing the uncertainty of image mutations.
- 2
- Retrieval-and-alignment based large-scale indoor point cloud semantic segmentation2
A novel architecture for direct and accurate large-scale point cloud segmentation based on point cloud retrieval and alignment that outperforms the state-of-the-art approaches in terms of both accuracy and robustness against noise and outliers.
- 2
- 2
- Siam2C: Siamese visual segmentation and tracking with classification-rank loss and classification-aware2
A classification-rank loss algorithm is designed to enlarge the margin between positive and negative samples, ensuring that positive samples are ranked higher than negative ones, enhancing the network’s ability to learn from positive and negative samples, allowing the tracker to accurately select the object for segmentation and tracking rather than being misled by interfering targets.
- 2
- Pansharpening via Detail Guided and Global Scale Convolution2
A pansharpening network based on detail guided and global scale convolution, which can balance the parameter quantity of the model and its accuracy is proposed, which alleviates issues such as insufficient receptive fields and excessive compression of long-distance information found in traditional convolutional neural networks.
- 2
- Focus Stacking with High Fidelity and Superior Visual Effects2
The results show that the proposed method can generate high-quality all-in-focus images by achieving two goals simultaneously, especially can successfully solve the TR problem and eliminate the visual effect degradation of synthesized images caused by the TR problem.
- Association of short-term nitrogen dioxide exposure with hospitalization for urolithiasis in Xinxiang, China: a time series study2
Stratified analysis showed that NO_2 had a more sensitive impact on urolithiasis hospitalizations in women and in those aged ≥65 years, and men and those younger than 65 years of age (exclude people aged 65) incurred more costs for urolithiasis hospitalizations.
- Establishing a prognostic model with ferroptosis-related long non-coding RNAs in bladder cancer2
Thirty-six lncRNAs were found to be linked to ferroptosis for the prognosis of patients with BLCA, and the risk assessment model exhibited superior performance in predicting prognosis for B LCA compared to conventional clinicopathological features.
- 2
- 2
- 2
- 2
- Novel Pathogenic Mutations of FERMT1 in two Chinese Kindler Syndrome Families2
Two novel pathogenic mutations in FERMT1 that caused Kindler syndrome are revealed that caused KNDLRS and all pathogenic mutants in F ERMT1 are summarized via the PubMed.
- 2
- 2
- Binary Hashing CNN Features for Action Recognition2
Experimental results on the JHMDB and MPII-Cooking datasets show that, for these new local features, binary hashing mapping on the sparsely sampled features led to significant performance improvements.
- Dual linear structured support vector machine tracking method via scale correlation filter2
The collaborative tracker comprised of a DLSSVM model and a scale correlation filter obtains good results in tracking target position and scale estimation and outperforms many popular top-ranking trackers.
- Fusion of Multimodal Color Medical Images Using Quaternion Principal Component Analysis2
An image fusion algorithm based on Quaternion Principal Component Analysis (QPCA), to fuse multimodal colour medical images, and comparison results show that the proposed algorithm performed better than existing algorithms in fusing colourMedical images.
- Adaptive Randomized Ensemble Tracking Using Appearance Variation and Occlusion Estimation2
A novel method to model the time-varying appearance of an object that takes appearance variation and occlusion of local patches into consideration is proposed and the results demonstrate that the proposed method is more robust against appearanceVariation and Occlusion than those state-of-the-art approaches.
- 2
- Object Recognition Method Based on Fusing Multi-features of Interest Points2
Experimental results show that this method effectively improves the object recognition accuracy.
- 2
- 1
- 1
- 1
- 1
- MSSDIBNet: Multiple Spatial–Spectral Dual-Injection Balance Network for Pansharpening1
A multiple spatial–spectral dual-injection balance network that alternately refines sharpness and fidelity without mutual interference, ensuring balanced optimization and substantial performance gains is proposed.
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- A Spatiotemporal Fusion Network for Remote Sensing Based on Global Context Attention Mechanism1
A spatiotemporal fusion network for remote sensing based on a global context (GC) attention mechanism that improves the fusion quality in the temporal, spatial, and visual domains is proposed.
- 1
- Robust and adaptive subspace learning for fast hyperspectral image denoising1
This paper introduces a semisequentially truncated higher-order singular value decomposition scheme for jointly performing basic estimation and learning a robust subspace basis, and the subspace dimensionality is adaptively detected by designing an appropriate information-theoretic criterion.
- 1
- Cardiocerebrovascular benefits of early rhythm control in patients with atrial fibrillation detected after stroke: a systematic review and meta-analysis1
Patients with AFDAS who underwent ERC treatment exhibited a decreased risk of cardiocerebrovascular events compared to those receiving usual care and ERC was associated with superior outcomes in terms of dementia.
- 1
- PriFU: Capturing Task-Relevant Information Without Adversarial Learning1
A method for privacy preservation in the inference phase by removing task-irrelevant information, which requires no knowledge of the privacy attacks nor introduction of adversarial learning, and introduces a metric to distinguish task-irrelevant information from task-relevant information.
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- Multi-layer Features Fusion Strategy with GIoU Loss for Siamese Network Tracking1
This work verifies the range of neuron receptive field which is suitable for the virtual tracking, and fuse three output layers of the network to improve the generalization ability, and adds GIoU (Generalized Intersection over Union) loss as a part of the target regression loss function to improve accuracy.
- 1
- Experimental Study on Visual Recovery Time of Drivers after Super Glare1
In the restoration process, the time of dark adaptation was significantly longer than that of light adaptation, and the change of the brightness of the fill light has great impact on the longest visual recovery time of the driver.
- A New Energy Saving Method for Street Lamp under the Equivalent Brightness Perception1
This pupil size aims to explore the specific influence of different glare on the brightness perception, and further understand how to reduce the glare of street lamp and the luminance of road surface while maintaining the same brightness perception.
- 1
- Human Action Recognition by Fusion of Convolutional Neural Networks and spatial-temporal Information1
This paper modify the existing network structure for action recognition, and then develop a different 3D CNN models in order to fuse the information of spatial and temporal dimensions.
- 1
- Facial Feature Extraction Based on Robust PCA and Histogram1
Compared with the related state-of-the-art methods, experimental results on Extended Yale B database verify the advancement of the proposed method for partially corrupted and occluded face images.
- 1
- 1
- 1
- A Planning Heuristic Based on Subgoal Ordering and Helpful Value1
A heuristic algorithm combined subgoal ordering with helpful value is proposed and the experimental results show the efficiency of the proposed heuristic search algorithm on the benchmarks of International Planning Competitions (IPC) 2008.
- 1
- 1
- Improved algorithm for iris location based on active contour1
Aim at the problem that traditional snake model is sensitive to the initial position and needs to rely on other mechanisms to locate snake points of the active contour in the vicinity of the target image, and show that this method is effective for iris localization.
- Plan recognition method based on intelligent planning1
Based on the basic ideas of robots'action planning in the fields of intelligent planning, a plan recognition solving process by using intelligent planning was proposed, which shows the feasibility of using the planning method to solve the problems of plan recognition.
- 1
- A New Delay Constraint Multicast Routing Algorithm1
A new algorithm for delay-constrained minimum-cost multicast routing directed at ensuring real time requirement and resource usage of multicast tree is proposed, which has nice creating speed and nice cost performance.
- Determination of the Th235 β-decay end point energy1
Sources of /sup 235/Th were prepared by 14-MeV neutron irradiation of uranium followed by radiochemical separation and the ..beta.. and ..gamma.. spectra were measured with a semiconductor telescope and a HpGe detector, respectively.
- –
- –
- Big data-driven fault diagnosis for railway turnout equipment–
A railway switch equipment fault diagnosis support system based on big data technology significantly enhances the accuracy and efficiency of fault diagnosis while reducing fault resolution time, demonstrating significant application value and promising prospects for widespread adoption.
- –
- –
- –
- –
- –
- HAG-YOLO: a tuberculosis bacilli detection algorithm based on hierarchical attention guidance mechanism–
This paper proposes HAG-YOLO, a hierarchical attention guidance framework for microscopic detection of tuberculosis bacilli, and proposes an inter stage transmission block to selectively regulate feature transmission, suppress irrelevant response accumulation, and enhance the expression of key features.
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- Pathology-Aware Reconstruction with Discriminative Knowledge Boosting Alignment for Che-Xray Vision-Language Pre-training–
A novel pre-training framework that integrates two key components: Pathology-Aware Reconstruction (PAR) and Discriminative Knowledge-Boosted Alignment (DKBA) that effectively combines the strengths of both paradigms while mitigating their inherent limitations.
- Denoising smooth‐detail representation of MRI‐SPECT fusion–
A method for calculating fused image from an anatomical image and a functional image, in which detail saliency information is preserved while suppressing noise, is presented, based on smooth‐detail representation used as a decomposition model scheme and based on denoising fusion rules to predict a fused version of the desired output.
- –
- –
- –
- –
- –
- Proteomic analysis of the initial wake up of vibrio splendidus persister cells–
This is the first time to study the initial wake up of persister cells with l -glutamic acid based on proteomic analysis, and the results revealed the main pathways involved in the early resuscitation of V. splendidus persisters cells.
- –
- Feature Selector: an effective module for robust high-voltage switchgear detection–
Experimental results on testing sets in three different environments demonstrate that the feature extraction module effectively improves the generalization performance of the YOLOv5-nano (You Only Look Once) network.
- A spatiotemporal fusion method based on interpretable deep networks–
The proposed UNSTF network method outperforms existing fusion methods in terms of both subjective and objective metrics and proposes a new network iteration loss function, where the predicted images of each iteration stage of the network are constrained by the real images, effectively improving the accuracy and reliability of model prediction.
- Double-Stack Aggregation Network Using a Feature-Travel Strategy for Pansharpening–
A double-stack feature aggregation module that can efficiently retain useful feature information by aggregating features extracted at different levels is proposed and achieves significant improvements in both spatial and spectral quality measurements compared to state-of-the-art methods.
- Parallel interactive delayed attention network for pansharpening–
A parallel interactive delayed attention network for panchromatic sharpening (PIDAN), which consists of multiple resolution subnetworks with parallel inputs to improve the representation of high-resolution feature maps by multiscale low-resolution features of the same depth and similar level and a delayed channel attention module.
- Multi-label Learning with Missing Labels Based on Instance-Wise and Label-Wise Correlations for Image Classification–
A multi-label learning algorithm with missing labels for image classification, which makes effective use of instance-wise correlation and label-wise correlations and the Laplacian manifold regularization is proposed.
- High-Low Spatial Frequency Feature for Siamese Network Visual Tracking–
A new ResNet framework with the high-frequency sampling convolution and the low-frequency sampled convolution for Siamese visual tracking, which is to replace AlexNet extractors ofSiamese trackers, is proposed.
- Attenuated Function of Angiotensin-(1-7) in Placental Artery During Preeclampsia not through MAS1 Receptor–
The data indicated that Ang-(1-7) antagonizes the function of Ang II in placenta vessels of normal pregnant women, but this effect is obviously eliminated in preeclampsia.
- –
- –
- –
- Global-Local Mutual Guided Learning for Person Re-identification–
Comprehensive experiments conducted on the public datasets of Market-1501 and DukeMTMC-ReID indicate that the proposed end-to-end feature learning strategy outperforms state-of-the-art approaches in several cases.
- Liver of CT Slice Localization and Segmentation Based on GPV-levelset–
The Gaussian pseudo-variance based on local patch is utilized to improve the level set segmentation technique, which named as GPV-levelset and considers the liver scale, the spatial texture, and the changes of the energy functional over iterations to improve segmentation of challenging liver cases.
- Collaborative correlation filters for real-time tracking with spatial constraint–
An attempt is made to evaluate the accuracy of the methods used to estimate the target size by employing an exhaustive scale search, as well as their efficiency and time efficiency.
- –
- –
- –
- –
- –
- –
- –
- Moving Object Detection Method Based on Spatio-Temporal Background Model–
The experimental results indicate that moving objects can be detected effectively by the proposed background modelling method in real time and can weaken the effects caused by dynamic background and illumination changes.
- Facial Feature Localization Using Robust Active Shape Model and POEM Descriptors–
An adaptive parameter estimation method are proposed by defining a rotation factor that significantly outperforms the original ASM and ASM plus LBP method under exterior variations and can provide more robust and accurate guidance for searching than intensity profiles.
- A histogram equalization method based on frequency division of mean filtering–
Experiments show that this histogram equalization method based on frequency division can not only enhance the contrast of images, but also effectively stretch the dynamic range of the image gray levels.
- Translation Algorithm for Negative Literals in Conformant Planning–
A translation algorithm for negative literals is proposed to reduce the state spaces in a conformant planning task based on the one-of relaxation technique in domain language and shows the efficiency of the proposed algorithm in pruning the state space in conformant plans tasks.
- –
- Concise Representations for State Spaces in Conformant Planning Tasks–
The paper proposed following methods: aborting void literal, translating the negative literal, and utilize domain transition graph to remove literals in the finite-domain representations.
- –
- –
- –
- –
- –
- –
- –
- New face detection method combined with image enhencement and skin segment–
A new method which combines image enhancement and skin segment with AdaBoost algorithm for face detection is put forward, which has a better improvement to the problem of leak detection and error detection.
- –
- –
- Multi-class Object Recognition Method Based on Affinity Propagation Clustering–
A new approach for learning a discriminative model of object classes combined affinity propagation clustering by using Affinity Propagation (AP), which shows that it is superior to the recognition rate of k-means algorithms.
- –
- A study and implementation of self-adaptive allocation algorithm for parallel program–
From the result of tests, the algorithm can largely raise the efficiency of parallel algorithms and encapsulate this algorithm into MPI API for engineering applications, and as long as software developers substitute this API for message send and receive functions, adaptive allocation can be achieved.
- –
- –
- –
- Automatic Extraction Algorithm for Chatbot Knowledge Based on Rough Set and Ensemble Learning–
The experimental results show that the extracted 〈thread-title,reply〉 pairs have high quality and could be used as chat knowledge and could help improve the recognition results.
- A Study and Accomplishment of the Huffman Algorithm Based on Condensed Huffman Table–
Compared with traditional Huffman coding table and other improved tables, the best advantage of new condensed Huffman table is that the space requirement is reduced significantly.
- Two implementation solutions of Bluetooth advanced audio distribution profile and its application–
It was given a description of two implementing solutions of Bluetooth Advanced Audio Distribution Profile (A2DP), focused on A2DP source side, compared with two solutions in applicable scenario.
- A New Dynamic QoS Multicast Routing Heuristic Algorithm–
A new algorithm for dynamic delay-constrained multicast routing directed at ensuring real time requirement of multicast tree and for efficiency of network resource using is proposed.
- Research on Risk Evaluation TSP-Based–
Based on the feathers of TSP and the analysis of present methods, this work presents a new probability-based classified weighted sum method that has been used in several medium/small scale software development and provides effective risk data.
- –
- Implementation of MPI-based parallel wavecluster algorithm on Shuguang TC1700–
A MPI-based parallel wavecluster algorithm was used to achieve parallel clustering by message-passing model on TC1700 to explain the main problem involved in programming, finding and labling the connected components, especially on the skill of programme making.
- Study and Implementation of Web Services-Based Network Management–
A new model based on Web service and the interface between SOAP and SNMP is brought forward and analyzed in software engineering and design patterns aspects to provide users with flexible interface and software reuse.
- Relation-based XML storage techniques–
The actualities of the recent researches on XML storage are discussed, and two issues are proposed, including multi-dimensional processing on node paths and efficient modification-validating support on XML data.
- –
- Constructing Effective Model for Web Application Development with Design Patterns–
This paper discusses and analyzes how to construct an effective model for Web application development from the view of data stream using design patterns and Java technology.
- A Survey of Blind Signature Studies–
This paper surveys the studies in the field of blind signature and points out three main research directions as well as several problems in the existing schemes.
- Application of Connection Pool in Personalized Information Service System–
Connection pool makes the database accessing efficient and safe and it is important to realize personalized information service and the great spending of database accessing can be saved and the connection can be reused.
- –
- Research and Implementation of Information BusArchitecture Based on CORBA–
A Information Bus Architecture and its model are introduced based on CORBA architecture and test shows that a large of time for RPC is saved on and efficiency is improved highly since this architecture is adopted.
- A XML Blind Signature Scheme Based on RSA Public Key System–
A XML blind signature scheme based on RSA public key system is proposed, which has the merits of RSA blind signature and XML, and has high security as well as many valuable applications.
- –
- The Shortest Path Subgraph–
The conception of Shortest Path Subgraph is presented, and it is able to save the storage spending by using this structure to store all possible shortest paths between two nodes.
- –
- –
- –
- SPT full dynamic algorithm based on BS model–
A full dynamic SPT algorithm is worked out which can update a SPT after topology changes happened, and it is legible and practical.
- –
- –
- Research on Evaluation Standard of Predicting Flight Ability by Means of P.E. Selecting Qualified Personnel–
By means of selecting personnel, pilots' a bility is researched on multi-quota comprehensive evaluation and relation is inquired between six exam quotas and pilots' flight ability, which is direct ratio and is highly effective.
- Object-oriented modeling for map generalisation–
The model for designing a map generalisation system is fully described, and an example is developed which shows that the proposed method is feasible, and demonstrates the efficiency of the object-oriented paradigm.
- –
- –
- –
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-15. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.