Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileAcademic lineage
Possible advisorsa guess from early papers, not confirmed
- Jiangtao CuiSuggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works197 from public data
- An overview of multi-modal medical image fusion368
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.
- 231
- Union Laplacian pyramid with multiple features for medical image fusion229
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.
- 198
- 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.
- Heart sounds classification using a novel 1-D convolutional neural network with extremely low parameter consumption175
A block-stacked style architecture with clique blocks is employed, and in each clique block a bidirectional connection structure is introduced in the proposed CNN, which achieves both spatial and channel attention leading a promising classification performance.
- Image splicing forgery detection combining coarse to refined convolutional neural network and adaptive clustering161
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.
- TinyCLIP: CLIP Distillation via Affinity Mimicking and Weight Inheritance149
The method introduces two core techniques: affinity mimicking and weight inheritance, which enables student models to mimic teachers’ behavior of learning cross-modal feature alignment in a visual-linguistic affinity space during distillation.
- Computer aided Alzheimer's disease diagnosis by an unsupervised deep learning technology149
This paper tackled the problem of automatic prediction of Alzheimer's Disease based on Magnetic Resonance Imaging (MRI) images, and proposed a fully unsupervised deep learning technology for AD diagnosis, which first implement the un Supervised Convolutional Neural Networks (CNNs) for feature extraction, and then utilize the un supervised predictor to achieve the final diagnosis.
- Image analysis by Bessel–Fourier moments143
A new set of moments based on the Bessel function of the first kind, named Bessel-Fourier moments (BFMs), which are more suitable than orthogonal Fourier-Mellin and Zernike moments for image analysis and rotation invariant pattern recognition.
- 126
- 126
- 118
- 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.
- Fast Continual Multi-View Clustering With Incomplete Views93
Comprehensive experiments conducted on various datasets demonstrate the superiority of FCMVC-IV over the competing approaches, which maintains a scalable consensus coefficient matrix and updates its knowledge with the incoming incomplete view rather than storing and recomputing all the data matrices.
- 88
- Scaling and rotation invariant analysis approach to object recognition based on Radon and Fourier–Mellin transforms87
Results show the high classification accuracy of this approach as a result of using the rotation and scaling invariant function instead of image binarization and normalization, it is also shown that this method is relatively robust in the presence of white noise.
- 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.
- Recent Advances for Aerial Object Detection: A Survey82
This survey presents a comprehensive review of recent advances in aerial object detection and summarizes the five imbalance problems of aerial object detection, including scale imbalance, spatial imbalance, objective imbalance, semantic imbalance, and class imbalance.
- 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.
- 72
- PAM-DenseNet: A Deep Convolutional Neural Network for Computer-Aided COVID-19 Diagnosis72
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.
- Locally GAN-generated face detection based on an improved Xception70
Experimental results demonstrate the superiority of the proposed model which outperforms some existing models, especially for the faces having small generated regions, especially on the basis of FFHQ dataset.
- Radial shifted Legendre moments for image analysis and invariant image recognition66
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.
- 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.
- Generative AI model privacy: a survey64
This survey encompasses a wide array of generative AI models, including language models, Generative Adversarial Networks, diffusion models, and their multi-modal counterparts, and indicates the critical need for continued research and development in privacy-preserving techniques for generative AI models.
- 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.
- 55
- 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
- Combined blur, translation, scale and rotation invariant image recognition by Radon and pseudo-Fourier–Mellin transforms51
A new method to construct a set of combined blur, translation, scale and rotation invariants using Radon and pseudo-Fourier-Mellin Transform invariants (RPFMI) is proposed, which is robust to additive white noise as a result of summing pixel values to generate projections in the Radon transform step.
- 45
- 44
- 44
- 42
- 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.
- ARISE: Graph Anomaly Detection on Attributed Networks via Substructure Awareness39
This work proposes a new graph anomaly detection framework on attributed networks via substructure awareness (ARISE), which focuses on the substructures in the graph to discern abnormalities and introduces a graph contrastive learning scheme, which observes attribute anomalies in the meantime.
- 39
- 37
- 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
- Elucidating subcellular architecture and dynamics at isotropic 100-nm resolution with 4Pi-SIM36
The 4Pi-SIM microscope combines the benefits of structured illumination microscopy with 4Pi geometry for improved axial resolution, offering isotropic, multicolor super-resolution imaging of dynamic processes in living cells.
- 36
- Image splicing forgery detection by combining synthetic adversarial networks and hybrid dense U‐net based on multiple spaces34
This work proposes two novel tasks to conquer the above challenges: Synthetic Adversarial Networks (SANs) and Hybrid Dense U‐Net (HDU‐Net).
- 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.
- Radial Tchebichef moment invariants for image recognition32
Experimental results show the efficiency and the robustness to noise of the proposed method for recognition tasks and a set of scale and rotation invariants extracted from radial Tchebichef moment invariants (RTMI).
- 31
- 31
- Controlling Neural Learning Network with Multiple Scales for Image Splicing Forgery Detection31
A neural learning network controlled by multiple scales (MCNL-Net) based on U-Net is applied to identify whether an image has been tampered and to locate the tampered regions and a MaxBlurPool method is applied into image splicing forgery detection for the first time, preserving the shift-equivariance of a convolutional neural network.
- Revisiting Initializing Then Refining: An Incomplete and Missing Graph Imputation Network30
A novel graph imputation network termed revisiting initializing then refining (RITR), where both attribute-incomplete and attribute-missing samples are completed under the guidance of a novel initializing-then-refining imputation criterion, which is the first end-to-end unsupervised framework dedicated to handling hybrid-absent graphs.
- 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
- 28
- 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.
- SARF: Aliasing Relation–Assisted Self-Supervised Learning for Few-Shot Relation Reasoning27
A novel Self-supervised learning model by leveraging Aliasing Relations to assist FS-KGR, termed SARF is proposed, and extensive experiments demonstrate that SARF achieves state-of-the-art (SOTA) performance compared with other methods in most cases.
- 27
- 27
- 27
- 26
- Quaternion weighted spherical Bessel-Fourier moment and its invariant for color image reconstruction and object recognition26
Experimental results show that the proposed Quaternion weighted Spherical Bessel–Fourier Moment outperforms frequently-used quaternion moments in terms of the color image reconstruction capability and invariant recognition accuracy in noise-free, Salt & Pepper noise and the Gaussian noise conditions.
- 26
- 24
- 24
- 23
- 23
- 22
- HS-Vectors: Heart Sound Embeddings for Abnormal Heart Sound Detection Based on Time-Compressed and Frequency-Expanded TDNN With Dynamic Mask Encoder22
Experimental results demonstrate that the proposed HS-Vectors method excels the state-of-the-art models in abnormality detection.
- 22
- 22
- 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.
- 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.
- BMI-Net: A Brain-inspired Multimodal Interaction Network for Image Aesthetic Assessment19
A brain-inspired multimodal interaction network (BMI-Net) that simulates how the association area of the cerebral cortex processes sensory stimuli is proposed and the knowledge integration LSTM (KI-LSTM) is proposed to learn the image-text interaction relation.
- Recent advancement in haze removal approaches19
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.
- Image splicing localization using residual image and residual-based fully convolutional network19
Experimental results show that the proposed improved FCN-based method performs better than some existing methods in localization ability, generalization ability, and robustness against additional operations.
- Multi-modal Color Medical Image Fusion Using Quaternion Discrete Fourier Transform19
A novel algorithm for the fusion of multimodal color medical images by dividing source images into blocks, converts each RGB block into quaternion representation and transforms them from special domain to frequency domain by applying quaternions discrete Fourier transform.
- Pancreatic Segmentation via Ringed Residual U-Net18
A new loss function focuses not only on the area coincidence degree but also on the focus shape similarity in the proposed segmentation method, which can yield satisfactory results via deep convolution and can consolidate the characteristics of traditional deep learning networks.
- Efficient Dynamic Ensembling for Multiple LLM Experts17
This paper proposes an efficient Dynamic Ensemble Reasoning paradigm, called DER to integrate the strengths of multiple LLM experts conditioned on dynamic inputs, and develops a Knowledge Transfer Prompt that enables the subsequent LLM candidates to transfer complementary knowledge effectively.
- Jointly RS Image Deblurring and Super-Resolution With Adjustable-Kernel and Multi-Domain Attention17
A unified theoretical analysis of RS images’ spatial and blur degeneration processes is given and a dual-branch parallel network named adjustable-kernel and multi-domain network (AKMD-Net) is proposed for the JRSIDSR task.
- 17
- 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.
- Privacy-Preserving Krawtchouk Moment feature extraction over encrypted image data17
A Privacy-Preserving Krawtchouk Moment (PPKM) feature extraction framework over encrypted image data by utilizing the Paillier cryptosystem is proposed and the computation complexity of PPKM can be significantly reduced with the block-based parallel algorithm.
- Image analysis using modified exponent-Fourier moments17
The experimental results demonstrate that the MEFMs perform better than traditional EFMs and other classic orthogonal moments including the latest image moments in terms of the image reconstruction capability and the invariant recognition accuracy of smoothing filters, in both noise-free and noisy conditions.
- Generic radial orthogonal moment invariants for invariant image recognition17
A generic approach based on Jacobi-Fourier moments for scale and rotation invariant analysis of radial orthogonal Moments invariants (JFMIs) is presented and the superiority of the proposed method and its robustness to noise in comparison with some exist methods is shown.
- 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.
- 15
- 15
- 15
- Quaternion fractional-order color orthogonal moment-based image representation and recognition15
Both theoretical analysis and experimental results demonstrate the value of the proposed quaternion fractional-order generalized Laguerre orthogonal moments and their geometric invariants in the representation and recognition of color images.
- 15
- Rotation invariant analysis and orientation estimation method for texture classification based on Radon transform and correlation analysis15
Results show the high classification accuracy of this approach as a result of using the Radon projection correlation distance instead of repetitious usage of discrete transforms, and it is shown that the proposed method presents high noise tolerance and yields high accuracy in orientation estimation in comparison with Khouzani's method.
- 14
- 14
- 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
- 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.
- 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
- 11
- 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
- 9
- VSSI-GGD: A Variation Sparse EEG Source Imaging Approach Based on Generalized Gaussian Distribution9
A new ESI method called Variation Sparse Source Imaging based on Generalized Gaussian Distribution (VSSI-GGD), which uses the generalized Gaussian prior as a sparse constraint on the spatial variation domain and embeds it into the Bayesian framework for source estimation.
- 9
- 9
- 9
- Compact and discriminative representation of Bag-of-Features9
This paper proposes a novel unsupervised dimensionality reduction framework for the histogram vectors and shows that a compact representation of BOFs can improve the classification accuracy compared with the original BOF.
- 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
- 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.
- A 3D polar-radius-moment invariant as a shape circularity measure8
The experimental results show that the proposed circularity measure can accord with human visual perception and consistently performs best in terms of retrieval efficiency in test datasets than other compared methods.
- Shape circularity measure method based on radial moments8
Compared with the classical circularity measure and Hu moment invariants based method, the new measure is robust to noise and performs better in the case of shapes with boundary defects.
- 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.
- 6
- 6
- 6
- 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
- Rotation, scaling and translation invariant texture recognition by Bessel-Fourier moments6
Experimental results validate the conclusion of theoretical derivation: BFM performs better in recognition capability and noise robustness in terms of RST texture recognition under both noise-free and noisy condition when compared with orthogonal Fourier-Mellin moments and Zernike moments based methods.
- Robust Image Stitching With Optimal Plane5
It is proposed to incorporate the universal prior of content perception into the image stitching model by a dual-branch architecture, which separately captures coarse and fine features and integrates them to achieve highly generalizable performance across diverse unseen real-world scenes.
- 5
- RSID-CR: Remote Sensing Image Denoising Based on Contrastive Learning5
This article proposes a remote sensing image denoising framework based on contrastive learning, named RSID-CR, which constructs positive and negative sample pairs between clean, noisy, and denoised images, and constructs a joint loss function consisting of reconstruction loss and contrastive regularization as a guide signal to train the denoising network.
- 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.
- 5
- 5
- 5
- 5
- 5
- 4
- 4
- 4
- Rethinking Detection Based Table Structure Recognition for Visually Rich Document Images4
The revisiting of existing detection-based models comprehensively and the underlying reasons hindering these models' performance are explored are explored, demonstrating that the findings can be a guideline for improving detection-based TSR models and that a purely detection-based solution is competitive with other types of solutions, such as graph-based and image-to-sequence solutions.
- 4
- 4
- LOMIA: Label-Only Membership Inference Attacks against Pre-trained Large Vision-Language Models3
This work proposes a novel framework: label-only membership inference attacks (LOMIA) targeting pre-trained VLLMs where only the model’s top-1 prediction is available, and proposes three effective attack methods that exploit the intuition that training samples are more likely to be memorized by the VLLMs, resulting in outputs that exhibit higher semantic alignment and lower perplexity.
- 3
- 3
- 3
- 3
- Secondary Labeling: A Novel Labeling Strategy for Image Manipulation Detection3
SeLa generates a query table containing multiple tampered categories and randomly reassigns these tampered classes to different types of tampered data, effectively improving the detection performance of models by refocusing the differences among the various data.
- 3
- 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
- ZeroPur: Succinct Training-Free Adversarial Purification2
A simple adversarial purification method without further training to purify adversarial images, called ZeroPur, which is independent of external models and requires no retraining of victim classifiers or auxiliary functions, relying solely on victim classifiers themselves to achieve purification.
- 2
- 2
- 2
- 2
- Sniffer: A Novel Model Type Detection System against Machine-Learning-as-a-Service Platforms2
This paper demonstrates a system, named Sniffer, that is capable of making model-type-sensitive attacks "great again" in real-world applications and shows how the audience can interact with SnIFFer in a web-based interface against five mainstream MLaaS platforms.
- 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.
- 2
- 2
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- Image Enhancement Based on Quotient Space1
Experimental results show that IEQS can enhance the contrast of original image while preserving the brightness and an image enhancement method based on quotient space (IEQS) is proposed in this paper.
- ClinAlign: Clinical Workflow Aligned Memory Retrieval for Radiology Report Generation–
ClinAlign—a memory-based retrieval framework aligned with clinical workflow, drawing inspiration from clinical diagnostic workflows is proposed, which constructs a disease-aware visual memory bank and introduces Classification-Guided Prompt Augmentation (CGPA), where disease state predictions are converted into structured diagnostic prompts to provide explicit semantic guidance for textual memory retrieval.
- Timestep-Aware Optimization Dynamics Analysis for Diffusion Training–
The corrected measurement shows that gradient similarity decays rapidly with timestep distance, and gradients from distant timesteps become nearly orthogonal as training progresses, suggesting that diffusion training consists of weakly coupled timestep-wise objectives.
- –
- –
- –
- –
- –
- –
- –
- 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.
- –
- Transfer morphological features for segmentation with few labels on fluorescent mitochondria images–
By reconstructing mitochondria in the masked regions, the model learns features more effectively on unlabeled samples, and improves segmentation performance even with limited labeled data.
- –
- –
- Novel Color Orthogonal Moments-Based Image Representation and Recognition–
Both theoretical analysis and experimental results demonstrate the value of the proposed quaternion fractional-order generalized Laguerre orthogonal moments and their geometric invariants in the representation and recognition of color images.
- –
- Blue noise sampling method based on mixture distance–
A new distance metric called mixture distance for blue noise sampling is presented, which is a combination of geodesic and feature distances, and an adaptive adjustment for parameters is also proposed to achieve a balance between the preservation of features and spatial properties.
- Three feature functions based on Radon transform for image retrieval and object recognition–
Theoretical and experimental results verify that the proposed method has strong capability and robustness in handling a variety of shapes and is invariant to rotation.
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 2026-10-12. 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.