Edge Details Research Articles

DTAUE: A multi-input multi-branch decoupled semi-supervised 3D network based on threshold adaptive uncertainty estimation for echocardiography segmentation

Echocardiography is a non-invasive examination method based on sound waves to generate images of the heart. Segmentation of the left ventricle (LV) and right ventricle (RV) is crucial for evaluating cardiac function. However, due to the limitations in ultrasound imaging angles and the high prevalence of noise in ultrasound data, automatic and accurate segmentation of the left and right ventricles is a challenging task. In this paper, we present a decoupled semi-supervised 3D echocardiography segmentation network based on threshold adaptive uncertainty estimation. To minimize the impact of missing targets caused by blind zones on segmentation accuracy, pixel-wise segmentation maps and target skeleton maps are jointly predicted by decoupled semi-supervised 3D network (DTAUE) so that our network is more effective to perceive the morphological features of segmented targets. And quantifying the uncertainty inherent to a model’s prediction is a promising endeavor to address the lack of reliability and overconfidence in the model. Previous uncertainty estimation methods set a uniform threshold for all training samples. However, this leads to the loss of edge detail information. Different from the existing methods, in our work, we propose a method of adaptive threshold uncertainty estimation to set different confidence threshold for different targets. Experiments on our self-collected 3D echocardiographic dataset and two publicly available datasets show that our method outperforms state-of-the-art semi-supervised learning methods. DTAUE would potentially serve as a new tool for diagnosing and monitoring various heart conditions.

Mangrove Extraction Algorithm Based on Orthogonal Matching Filter-Weighted Least Squares.

High-precision extraction of mangrove areas is a crucial prerequisite for estimating mangrove area as well as for regional planning and ecological protection. However, mangroves typically grow in coastal and near-shore areas with complex water colors, where traditional mangrove extraction algorithms face challenges such as unclear region segmentation and insufficient accuracy. To address this issue, in this paper we propose a new algorithm for mangrove identification and extraction based on Orthogonal Matching Filter-Weighted Least Squares (OMF-WLS) target spectral information. This method first selects GF-6 remote sensing images with less cloud cover, then enhances mangrove feature information through preprocessing and band extension, combining whitened orthogonal subspace projection with the whitened matching filter algorithm. Notably, this paper innovatively introduces Weighted Least Squares (WLS) filtering technology. WLS filtering precisely processes high-frequency noise and edge details in images using an adaptive weighting matrix, significantly improving the edge clarity and overall quality of mangrove images. This innovative approach overcomes the bottleneck of traditional methods in effectively extracting edge information against complex water color backgrounds. Finally, Otsu's method is used for adaptive threshold segmentation of GF-6 remote sensing images to achieve target extraction of mangrove areas. Our experimental results show that OMF-WLS improves extraction accuracy compared to traditional methods, with overall precision increasing from 0.95702 to 0.99366 and the Kappa coefficient rising from 0.88436 to 0.98233. In addition, our proposed method provides significant improvements in other metrics, demonstrating better overall performance. These findings can provide more reliable technical support for the monitoring and protection of mangrove resources.

RGB-T tracking with frequency hybrid awareness

Recently, impressive progress has been made with transformer-based RGB-T trackers due to the transformer’s effectiveness in capturing low-frequency information (i.e., high-level semantic information). However, some studies have revealed that the transformer exhibits limitations in capturing high-frequency information (i.e., low-level texture and edge details), thereby restricting the tracker’s capacity to precisely match target details within the search area. To address this issue, we propose a frequency hybrid awareness modeling RGB-T tracker, abbreviated as FHAT. Specifically, FHAT combines the advantages of convolution and maximum pooling in capturing high-frequency information on the architecture of transformer. In this way, it strengthens the high-frequency features and enhances the model’s perception of detailed information. Additionally, to enhance the complementary effect between the two modalities, the tracker utilizes low-frequency information from both modalities for modality interaction, which can avoid interaction errors caused by inconsistent local details of the multimodality. Furthermore, these high-frequency information and interaction low-frequency information will then be fused, allowing the model to adaptively enhance the frequency features of the modal expression. Through extensive experiments on two mainstream RGB-T tracking benchmarks, our method demonstrates competitive performance.

Gradient-Guided Network With Fourier Enhancement for Glioma Segmentation in Multimodal 3D MRI.

Glioma segmentation is a crucial task in computer-aided diagnosis, requiring precise discrimination between lesions and normal tissue at the pixel level. Popular methods neglect crucial edge information, leading to inaccurate contour delineation. Moreover, global information has been proven beneficial for segmentation. The feature representations extracted by convolution neural networks often struggle with local-related information owing to the limited receptive fields. To address these issues, we propose a novel edge-aware segmentation network that incorporates a dual-path gradient-guided training strategy with Fourier edge-enhancement for precise glioma segmentation, a.k.a. GFNet. First, we introduce a Dual-path Gradient-guided Training strategy (DGT) based on a Siamese network guiding the optimizing direction of one path by the gradient from the other path. DGT pays attention to the indistinguishable pixels with large weight-updating gradient, such as the pixels near the boundary, to guide the network training, addressing hard samples. Second, to further perceive the edge information, we derive a Fourier Edge-enhancement Module (FEM) to augment feature edges with high-frequency representations from the spectral domain, providing global information and edge details. Extensive experiments on public glioma segmentation datasets, BraTS2020 and Medical Segmentation Decathlon (MSD) glioma and prostate segmentation, demonstrate that GFNet achieves competitive performance compared to other state-of-the-art methods, both qualitatively and quantitatively.

A multi-exposure image fusion using adaptive color dissimilarity and dynamic equalization techniques

In the domain of image processing, Multi-Exposure Image Fusion (MEF) emerges as a crucial technique for developing high dynamic range (HDR) representations from fusing sequences of low dynamic range images. Conventional fusion methods often suffer from shortcomings such as detail loss, edge artifacts, and color inconsistencies, thereby compromising the quality of the fused output which is further diminished with extremely exposed and limited inputs. While there have been a few efforts to conduct fusion on limited and impaired static input images, there has been no exploration into the fusion of dynamic image sets. This paper proposes an effective MEF approach that operates on a minimum of two extremely exposed, limited datasets of both static and dynamic scenes. The approach initiates with categorizing input images into under-exposed and over-exposed categories based on lighting levels, subsequently applying tailored exposure correction strategies. Through iterative refinement and selection of optimally exposed variant, we construct an advanced intermediate stack, upon which fusion is performed by a pyramidal fusion technique. The method relies on adaptive well-exposedness and color gradient to develop weight maps for pyramidal fusion. The initial weights are refined using a Gaussian filter and this results in the creation of a seamlessly fused image with expanded dynamic range. Additionally, for dynamic imagery, we propose an adaptive color dissimilarity and dynamic equalization to reduce ghosting artifacts. Comparative assessments against existing methodologies, both visually and empirically confirms the superior performance of the proposed model.

Super-Resolution Reconstruction of Remote Sensing Images Using Chaotic Mapping to Optimize Sparse Representation.

Current super-resolution algorithms exhibit limitations when processing noisy remote sensing images rich in surface information, as they tend to amplify noise during the recovery of high-frequency signals. To mitigate this issue, this paper presents a novel approach that incorporates the concept of compressed sensing and explores the super-resolution problem of remote sensing images for space cameras, particularly for high-speed imaging systems. The proposed algorithm employs K-singular value decomposition (K-SVD) to jointly train high- and low-resolution image blocks, updating them column by column to obtain overcomplete dictionary pairs. This approach compensates for the deficiency of fixed dictionaries in the original algorithm. In the process of dictionary updating, we innovatively integrate the circle chaotic mapping into the solution process of the dictionary sequence, replacing pseudorandom numbers. This integration facilitates balanced traversal and simplifies the search for global optimal solutions. For the optimization problem of sparse coefficients, we utilize the orthogonal matching pursuit method (OMP) instead of the L1 norm convex optimization method used in most reconstruction techniques, thereby complementing the K-SVD dictionary update algorithm. After upscaling and denoising the image using the dictionary pair mapping relationship, we further emphasize image edge details with local gradients as constraints. When compared with various representative super-resolution algorithms, our algorithm effectively filters out noise and stains in low-resolution images. It not only performs well visually but also stands out in objective evaluation indicators such as the peak signal-to-noise ratio and information entropy. The experimental results validate the effectiveness of the proposed method in super-resolution remote sensing images, yielding high-quality remote sensing image data.

A Novel Panorama Depth Estimation Framework for Autonomous Driving Scenarios Based on a Vision Transformer.

An accurate panorama depth estimation result is crucial to risk perception in autonomous driving practice. In this paper, an innovative framework is presented to address the challenges of imperfect observation and projection fusion in panorama depth estimation, enabling the accurate capture of distances from surrounding images in driving scenarios. First, the Patch Filling method is proposed to alleviate the imperfect observation of panoramic depth in autonomous driving scenarios, which constructs a panoramic depth map based on the sparse distance data provided by the 3D point cloud. Then, in order to tackle the distortion challenge faced by outdoor panoramic images, a method for image context learning, ViT-Fuse, is proposed and specifically designed for equirectangular panoramic views. The experimental results show that the proposed ViT-Fuse reduces the estimation error by 9.15% on average in driving scenarios compared with the basic method and exhibits more robust and smoother results on the edge details of the depth estimation maps.

A Feature-Driven Inception Dilated Network for Infrared Image Super-Resolution Reconstruction

Image super-resolution (SR) algorithms based on deep learning yield good visual performances on visible images. Due to the blurred edges and low contrast of infrared (IR) images, methods transferred directly from visible images to IR images have a poor performance and ignore the demands of downstream detection tasks. Therefore, an Inception Dilated Super-Resolution (IDSR) network with multiple branches is proposed. A dilated convolutional branch captures high-frequency information to reconstruct edge details, while a non-local operation branch captures long-range dependencies between any two positions to maintain the global structure. Furthermore, deformable convolution is utilized to fuse features extracted from different branches, enabling adaptation to targets of various shapes. To enhance the detection performance of low-resolution (LR) images, we crop the images into patches based on target labels before feeding them to the network. This allows the network to focus on learning the reconstruction of the target areas only, reducing the interference of background areas in the target areas’ reconstruction. Additionally, a feature-driven module is cascaded at the end of the IDSR network to guide the high-resolution (HR) image reconstruction with feature prior information from a detection backbone. This method has been tested on the FLIR Thermal Dataset and the M3FD Dataset and compared with five mainstream SR algorithms. The final results demonstrate that our method effectively maintains image texture details. More importantly, our method achieves 80.55% mAP, outperforming other methods on FLIR Dataset detection accuracy, and with 74.7% mAP outperforms other methods on M3FD Dataset detection accuracy.

Edge enhancement and feature modulation based network for light field depth estimation

Estimating depth from light field images is a critical issue in light field applications. While learning-based methods have made significant strides in light field depth estimation, achieving high accuracy and speed simultaneously remains a major challenge. This paper proposes a light field depth estimation network based on edge enhancement and feature modulation, which significantly improves depth estimation results by emphasizing inter-view correlations while preserving image edge features. Specifically, to prioritize edge details, we introduce an Edge-Enhanced Cost Constructor (EECC) that integrates edge information with existing cost constructors to improve depth estimation performance in complex areas. Furthermore, most light field depth estimation networks utilize only sub-aperture images (SAIs) without considering the inherent angular information in macro-pixel image (MacPI). To address this limitation, we propose the MacPI-Guided Feature Modulation (MGFM) module, which leverages angular information between different views in MacPI to modulate features at each view. Experimental results show that our method not only performs excellently on synthetic datasets but also demonstrates outstanding generalization on real-world datasets, achieving a better balance between accuracy and computation speed.

Capsule Endoscopy Image Enhancement for Small Intestinal Villi Clarity

Wireless capsule endoscopy (WCE) has become an important tool for gastrointestinal examination due to its non-invasive nature and minimal patient discomfort. However, the quality of WCE images is often limited by built-in lighting and the complex gastrointestinal environment, particularly in the region filled with small intestinal villi. Additionally, the morphology of these villi usually serves as a crucial indicator for related diseases. To address this, we propose a novel method to enhance the clarity of small intestinal villi in WCE images. Our method uses a guided filter to separate the low- and high-frequency components of WCE images. Illumination gain factors are calculated from the low-frequency components, while gradient gain factors are derived from Laplacian convolutions on different regions. These factors enhance the high-frequency components, combined with the original image. This approach improves edge detail while suppressing noise and avoiding edge overshoot, providing clearer images for diagnosis. Experimental results show that our proposed method achieved a 45.47% increase in PSNR compared to classical enhancement algorithms, a 12.63% improvement in IRMLE relative to the original images, and a 31.84% reduction in NIQE with respect to the original images.

Research on coastline extraction and dynamic change from remote sensing images based on deep learning

Accurate coastline extraction is crucial for the scientific management and protection of coastal zones. Due to the diversity of ground object details and the complexity of terrain in remote sensing images, the segmentation of sea and land faces challenges such as unclear segmentation boundaries and discontinuous coastline contours. To address these issues, this study improve the accuracy and efficiency of coastline extraction by improving the DeepLabv3+ model. Specifically, this study constructs a sea-land segmentation network, DeepSA-Net, based on strip pooling and coordinate attention mechanisms. By introducing dynamic feature connections and strip pooling, the connection between different branches is enhanced, capturing a broader context. The introduction of coordinate attention allows the model to integrate coordinate information during feature extraction, thereby allowing the model to capture longer-distance spatial dependencies. Experimental results has shown that the model can achieves a land-sea segmentation mean intersection over union (mIoU) ration and Recall of over 99% on all datasets. Visual assessment results show more complete edge details of sea-land segmentation, confirming the model’s effectiveness in complex coastal environments. Finally, using remote sensing data from a coastal area in China as an application instance, coastline extraction and dynamic change analysis were implemented, providing new methods for the scientific management and protection of coastal zones.

A Novel Fractional-Order Non-Convex TVα,p Model in Image Deblurring

In this paper, we propose a non-convex model with fractional-order applied to image deblurring problems. In the new model, fractional-order gradients have been introduced to preserve detailed features, and a source term with a blurry kernel is used for deblurring. This aspect of the model ensures that it can handle various blurring scenarios. Additionally, we devise an algorithm that maintains the non-expansiveness of the support set for image gradients, serving as a critical component in our approach to address image deblurring issues. After approximate linearization, the algorithm can be easily implemented. Some standard image processing techniques similar to fast Fourier transform can be utilized. Global convergence has likewise been confirmed and established. Moreover, we have also demonstrated that the proposed deblurring algorithm exhibits edge preservation properties. Compared with several existing classic models, the proposed method maintains a good balance between detail preservation, edge preservation, and deblurring. In addition, compared with several classic methods, the proposed method improved PSNR and SSIM by 0.9733 and 0.0111, respectively.

High dynamic range image tone mapping based on variational image decomposition and color correction

Tone mapping aims to reproduce high dynamic range (HDR) images on low dynamic range (LDR) display devices. The layer-decomposition-based tone mapping algorithms usually decompose HDR images into base and detail layers and process them accordingly. Since the detail layer is not refined again, these algorithms may suffer from halo artifacts, relatively dull color, and over-enhancement problems. To solve these problems, we introduce the variational image decomposition model and color correction into the field of tone mapping, and propose a tone mapping algorithm that combines the three-part variational image decomposition model with a color correction scheme based on minimum color loss. In contrast to traditional layer-decomposition-based tone mapping algorithms, we employ the three-part decomposition model that decomposes the HDR image into three components: a cartoon component, a texture component, and a high-frequency detail component. The cartoon component is compressed using a gamma function, while the texture and high-frequency detail components are each enhanced using the nonlinear stretching functions. This approach allows for better detail enhancement while preserving the smoothness of the image. In addition, we further use the color correction scheme based on minimum color loss to solve the color problem caused by the layer-decomposition-based tone mapping algorithms. Experimental results show that our proposed tone mapping algorithm can reduce the halo artifacts and preserve the edge details while outperforming some state-of-the-art tone mapping algorithms.

Multiobjective optimization guided by image quality index for limited-angle CT image reconstruction.

Due to the incomplete projection data collected by limited-angle computed tomography (CT), severe artifacts are present in the reconstructed image. Classical regularization methods such as total variation (TV) minimization, ℓ0 minimization, are unable to suppress artifacts at the edges perfectly. Most existing regularization methods are single-objective optimization approaches, stemming from scalarization methods for multiobjective optimization problems (MOP). To further suppress the artifacts and effectively preserve the edge structures of the reconstructed image. This study presents a multiobjective optimization model incorporates both data fidelity term and ℓ0-norm of the image gradient as objective functions. It employs an iterative approach different from traditional scalarization methods, using the maximization of structural similarity (SSIM) values to guide optimization rather than minimizing the objective function.The iterative method involves two steps, firstly, simultaneous algebraic reconstruction technique (SART) optimizes the data fidelity term using SSIM and the Simulated Annealing (SA) algorithm for guidance. The degradation solution is accepted in the form of probability, and guided image filtering (GIF) is introduced to further preserve the image edge when the degradation solution is rejected. Secondly, the result from the first step is integrated into the second objective function as a constraint, we use ℓ0 minimization to optimize ℓ0-norm of the image gradient, and the SSIM, SA algorithm and GIF are introduced to guide optimization process by improving SSIM value like the first step. With visual inspection, the peak signal-to-noise ratio (PSNR), root mean square error (RMSE), and SSIM values indicate that our approach outperforms other traditional methods. The experiments demonstrate the effectiveness of our method and its superiority over other classical methods in artifact suppression and edge detail restoration.

Adaptive orthogonal directional total variation with kernel regression for CT image denoising.

Low-dose computed tomography (CT) has been successful in reducing radiation exposure for patients. However, the use of reconstructions from sparse angle sampling in low-dose CT often leads to severe streak artifacts in the reconstructed images. In order to address this issue and preserve image edge details, this study proposes an adaptive orthogonal directional total variation method with kernel regression. The CT reconstructed images are initially processed through kernel regression to obtain the N-term Taylor series, which serves as a local representation of the regression function. By expanding the series to the second order, we obtain the desired estimate of the regression function and localized information on the first and second derivatives. To mitigate the noise impact on these derivatives, kernel regression is performed again to update the first and second derivatives. Subsequently, the original reconstructed image, its local approximation, and the updated derivatives are summed using a weighting scheme to derive the image used for calculating orientation information. For further removal of stripe artifacts, the study introduces the adaptive orthogonal directional total variation (AODTV) method, which denoises along both the edge direction and the normal direction, guided by the previously obtained orientation. Both simulation and real experiments have obtained good results. The results of two real experiments show that the proposed method has obtained PSNR values of 34.5408 dB and 29.4634 dB, which are 1.2392-5.9333 dB and 2.828-6.7995 dB higher than the contrast denoising algorithm, respectively, indicating that the proposed method has good denoising performance. The study demonstrates the effectiveness of the method in eliminating strip artifacts and preserving the fine details of the images.

Multidirectional Attention Fusion Network for SAR Change Detection

Synthetic Aperture Radar (SAR) imaging is essential for monitoring geomorphic changes, urban transformations, and natural disasters. However, the inherent complexities of SAR, particularly pronounced speckle noise, often lead to numerous false detections. To address these challenges, we propose the Multidirectional Attention Fusion Network (MDAF-Net), an advanced framework that significantly enhances image quality and detection accuracy. Firstly, we introduce the Multidirectional Filter (MF), which employs side-window filtering techniques and eight directional filters. This approach supports multidirectional image processing, effectively suppressing speckle noise and precisely preserving edge details. By utilizing deep neural network components, such as average pooling, the MF dynamically adapts to different noise patterns and textures, thereby enhancing image clarity and contrast. Building on this innovation, MDAF-Net integrates multidirectional feature learning with a multiscale self-attention mechanism. This design utilizes local edge information for robust noise suppression and combines global and local contextual data, enhancing the model’s contextual understanding and adaptability across various scenarios. Rigorous testing on six SAR datasets demonstrated that MDAF-Net achieves superior detection accuracy compared with other methods. On average, the Kappa coefficient improved by approximately 1.14%, substantially reducing errors and enhancing change detection precision.

An Edge-Enhanced Network for Polyp Segmentation.

Colorectal cancer remains a leading cause of cancer-related deaths worldwide, with early detection and removal of polyps being critical in preventing disease progression. Automated polyp segmentation, particularly in colonoscopy images, is a challenging task due to the variability in polyp appearance and the low contrast between polyps and surrounding tissues. In this work, we propose an edge-enhanced network (EENet) designed to address these challenges by integrating two novel modules: the covariance edge-enhanced attention (CEEA) and cross-scale edge enhancement (CSEE) modules. The CEEA module leverages covariance-based attention to enhance boundary detection, while the CSEE module bridges multi-scale features to preserve fine-grained edge details. To further improve the accuracy of polyp segmentation, we introduce a hybrid loss function that combines cross-entropy loss with edge-aware loss. Extensive experiments show that the EENet achieves a Dice score of 0.9208 and an IoU of 0.8664 on the Kvasir-SEG dataset, surpassing state-of-the-art models such as Polyp-PVT and PraNet. Furthermore, it records a Dice score of 0.9316 and an IoU of 0.8817 on the CVC-ClinicDB dataset, demonstrating its strong potential for clinical application in polyp segmentation. Ablation studies further validate the contribution of the CEEA and CSEE modules.

ECT image reconstruction algorithm based on Res-SEUnet

A Res-SEUnet algorithm for ECT image reconstruction is proposed to address the problem of artifacts and blurring of media boundaries in ECT (capacitance tomography) reconstructed images of high-density and small-size media. A convolutional neural network is used to extract the detail features of the LBP image and recover the edge details of the ECT inverted image. Simulation experiments are performed on various complex multi-media datasets built in Comsol and Matlab. The results of the test set show that the Res-SEUnet based ECT reconstruction algorithm can by better reconstruction results on various complex multimedia datasets. [Display omitted]

Tone mapping algorithm based on BL-Hilbert-L 2 decomposition model for HDR image

Abstract Tone mapping algorithms are mainly used to produce standard dynamic range (SDR) images from high dynamic range (HDR) images. To address the issues of halo artifacts and over-enhancement in current tone mapping algorithms based on layer decomposition model, we propose a tone mapping algorithm based on the BL-Hilbert-L2 decomposition model. In our proposed algorithm, we first convert HDR image from RGB space to HSV space. Subsequently, utilizing the BL-Hilbert-L2 decomposition model, we decompose the V channel of the HDR image into cartoon, texture and high-frequency detail components. Next, we separately compress the cartoon component and stretch the texture and high-frequency detail components. The processed cartoon, texture and high-frequency detail components are then recombined to form a new V channel. Finally, by converting the H and S channels of the HDR image, along with the newly formed V channel, back into the RGB color space, we obtain a tone mapped SDR image. Experimental results demonstrate that our proposed algorithm can effectively eliminate the halo artifacts and avoid the issue of over-enhancement while retaining the edge details.

MEEAFusion: Multi-Scale Edge Enhancement and Joint Attention Mechanism Based Infrared and Visible Image Fusion.

Infrared and visible image fusion can integrate rich edge details and salient infrared targets, resulting in high-quality images suitable for advanced tasks. However, most available algorithms struggle to fully extract detailed features and overlook the interaction of complementary features across different modal images during the feature fusion process. To address this gap, this study presents a novel fusion method based on multi-scale edge enhancement and a joint attention mechanism (MEEAFusion). Initially, convolution kernels of varying scales were utilized to obtain shallow features with multiple receptive fields unique to the source image. Subsequently, a multi-scale gradient residual block (MGRB) was developed to capture the high-level semantic information and low-level edge texture information of the image, enhancing the representation of fine-grained features. Then, the complementary feature between infrared and visible images was defined, and a cross-transfer attention fusion block (CAFB) was devised with joint spatial attention and channel attention to refine the critical supplemental information. This allowed the network to obtain fused features that were rich in both common and complementary information, thus realizing feature interaction and pre-fusion. Lastly, the features were reconstructed to obtain the fused image. Extensive experiments on three benchmark datasets demonstrated that the MEEAFusion proposed in this research has considerable strengths in terms of rich texture details, significant infrared targets, and distinct edge contours, and it achieves superior fusion performance.