Image Decomposition Algorithm Research Articles

A dynamic alignment and illumination‐aware convolution for shadow removal

AbstractShadow removal is a challenging task because the variety of shadows is influenced by surface texture and lighting. This paper proposes a dynamic alignment and illumination‐aware convolution (DAIC), which consists of a Feature Alignment Module (FAM) and a Dynamic Weight Module (DWM). FAM aligns the downsampled deep features with the original features and helps to extract the optimal local information to ensure that the object texture features are not corrupted. DWM generates weights according to different lighting variations for a better shadow removal result. The shadow removal approach is based on an image decomposition algorithm using a multi‐exposure image fusion model. Here, the shadow removal network and refinement network use U‐Net framework, and the transposed convolution operations are replaced with DAIC in the decoder part of U‐Net to improve the performance of the network. The experiments are conducted on two large shadow removal datasets, ISTD+ and SRD. Compared to state‐of‐the‐art methods, this model achieves optimal performance in terms of Root Mean Square Error (RMSE) for the non‐shadow region. It also achieves performance comparable to the state‐of‐the‐art method in terms of RMSE for the shadow region and structural similarity index measurement for the entire image.

DeepVID: deep-learning accelerated variational image decomposition model tailored to fringe pattern filtration

The algorithms defined as variational image decomposition (VID) constitute the state-of-the-art in solving the image prefiltration problem. However, the discussion about the advantages and disadvantages of different VID models in the context of fringe pattern prefiltration is yet to be addressed and this work is the first one leaning into this issue. The unsupervised variational image decomposition (uVID) algorithm allows for automatic, accurate and robust preprocessing of diverse fringe patterns and introduces the parameters and stopping criterion for Chambolle’s iterative projection algorithm to separate the fringes and background. However, determining the stopping criterion in each iteration is a severely time-consuming process, which is particularly important given the fact that in many cases thousands of iterations must be calculated to obtain a satisfactory fringe pattern decomposition result. Therefore, the idea of using convolutional neural network to map the relationship between the fringe pattern spatial intensity distribution and the required number of Chambolle projection iterations has emerged. That way, it is no longer required to determine the value of the stopping criterion in every iteration, but the appropriate number of iterations is known in advance via machine learning process. We showed that the calculation time is reduced on average by 3–4 times by employing the deep learning-based acceleration (convolutional neural network called DeepVID) without jeopardizing the overall accuracy of the prefiltration. This way an important progress in developing uVID algorithm features towards real-time studies of dynamic phenomena is reported in this contribution. For the sake of metrological figure of merit, we employ deep learning based solution, for the first time to the best of our knowledge, to accelerate powerful and well-established VID approach, not to bypass it completely.

Medical image fusion based on enhanced three-layer image decomposition and Chameleon swarm algorithm

Medical image fusion has brought practical applications in clinical diagnosis. However, image fusion methods still face challenges because of problems with the quality of the input images. Phenomena such as noise or low contrast can appear in the original images, which significantly degrades the quality of the synthesized image. Most current image synthesis algorithms do not thoroughly focus on solving the image quality problem. Therefore, if the input images are noisy or low-contrast, it will significantly affect the resulting image synthesis. This study proposes a new image synthesis method that allows efficient operation even when the input image is noisy or has low contrast. Firstly, we propose a new image enhancement algorithm that focuses on solving the problem of noise or low contrast of the input image. This enhancement algorithm is built based on several methods, such as Contrast-limited adaptive histogram equalization (CLAHE), Block-matching and 3D filtering (BM3D), and Chameleon swarm algorithm (CSA). Next, we introduce a method to decompose the image into three enhanced layers based on the adaptive parameters obtained from the proposed image enhancement method. This image decomposition method is used to decompose the input medical images into high-frequency and low-frequency layers. Finally, high-frequency layers are synthesized based on the CSA method, and low-frequency layers are synthesized based on the sum of the local energy functions using the Prewitt compass operator (SLE_PCO). One hundred eighty medical images, various imaging enhancement methods, and medical image synthesis were used for comparison and evaluation. Experimental results show that our image enhancement method works well with noisy and low-contrast images. Furthermore, our image fusion method gives the best performance when compared with the latest image synthesis methods available today.

A new medical image enhancement algorithm using adaptive parameters

AbstractThe quality of medical images plays a vital role in many image processing applications such as image segmentation, feature extraction, image classification, image recognition, and image fusion. Some of the common problems with Medical images are noise, blur, or low contrast. According to our observations, current image enhancement algorithms only focus on solving individual problems such as gray level adjustment, noise reduction, or sharpness enhancement. This paper proposes a novel algorithm to solve problems on images simultaneously. First, we propose an image decomposition algorithm. This algorithm allows decomposing the image into three components: structure (IS), texture (IT), and noise (IN). Second, the structural component (IS) is enhanced by the contrast‐limited adaptive histogram equalization method to obtain ICLAHE. We use the structure tensor salient detection operator and the Laplace edge detection operator to add structural and texture features. These operators are applied to the ICLAHE and IT components to obtain the ISTS and ILED components, respectively. The IT and ILED components are used to generate the enhanced component (called ) by using the Max operator. Third, the Marine predators algorithm is used to find the optimal parameters β1, β2, β3, and β4 corresponding to ICLAHE, ISTS, , and IN. Finally, the enhanced image is made up of the sum of the ICLAHE, ISTS, , and IN images multiplied by the optimal parameters β1, β2, β3, and β4, respectively. Six state‐of‐the‐art image enhancement approaches, seven medical image fusion algorithms, and six image quality metrics have been utilized to verify the proposed approach's effectiveness. The experimental results show that the proposed method significantly improves the quality of the input medical images as well as significantly improves the efficiency of current medical image synthesis algorithms.

Image Decomposition Algorithm with a Structural Constraint of the Averaging Region

Image Decomposition Algorithm with a Structural Constraint of the Averaging Region

ShadingNet: Image Intrinsics by Fine-Grained Shading Decomposition

In general, intrinsic image decomposition algorithms interpret shading as one unified component including all photometric effects. As shading transitions are generally smoother than reflectance (albedo) changes, these methods may fail in distinguishing strong photometric effects from reflectance variations. Therefore, in this paper, we propose to decompose the shading component into direct (illumination) and indirect shading (ambient light and shadows) subcomponents. The aim is to distinguish strong photometric effects from reflectance variations. An end-to-end deep convolutional neural network (ShadingNet) is proposed that operates in a fine-to-coarse manner with a specialized fusion and refinement unit exploiting the fine-grained shading model. It is designed to learn specific reflectance cues separated from specific photometric effects to analyze the disentanglement capability. A large-scale dataset of scene-level synthetic images of outdoor natural environments is provided with fine-grained intrinsic image ground-truths. Large scale experiments show that our approach using fine-grained shading decompositions outperforms state-of-the-art algorithms utilizing unified shading on NED, MPI Sintel, GTA V, IIW, MIT Intrinsic Images, 3DRMS and SRD datasets.

A novel approach based on Three-scale image decomposition and Marine predators algorithm for multi-modal medical image fusion

Multi-modal medical image fusion not only creates an image that preserves important information from the input images but also significantly improves in quality. This work contributes significantly to improving the ability of the physician to diagnose. So far, there have been many proposed approaches to improve efficiency for medical image fusion. However, some existing approaches still have certain drawbacks. The first drawback is that some vital information such as edges may be lost in the output image because of the high-frequency component fusion rules’ inefficiency. The second drawback is that the fused images often have low contrast because they have applied an average rule for the low-frequency components. In this study, a novel approach is introduced to overcome the aforementioned drawbacks, and it includes the following main steps. Firstly, the three-scale decomposition (TSD) technique is introduced to obtain the base and detail components. Secondly, a rule base on local energy function using the Kirsch compass operator is applied to fusing detail layers, which helps the output image preserve important information. Thirdly, the Marine predators algorithm (MPA) is utilized to fuse base layers by optimal parameters, allowing the output image to have good quality. To verify the proposed approach's effectiveness, we have utilized five state-of-the-art medical image fusion approaches and six image quality metrics for comparison. Experimental results show that the proposed approach significantly enhanced the fused image’s quality and preserved edge information.

Scalable image decomposition

Image decomposition, which separates a given input image into structure and texture images, has been used for various applications in the fields of computer graphics and image processing. Most previous image decomposition algorithms and applications start with high-quality images, but it requires numerous steps to separate and edit the high-resolution structure and texture images from the low-resolution input counterparts. This paper proposes a simple but effective end-to-end deep neural image decomposition network, which is called ”scalable image decomposition”, by decomposing and upscaling structure and texture images from the degraded input image at the same time. We train the deep neural network to automatically estimate high-resolution structure and texture from the low-resolution input image by designing a shared feature extractor, and structure and texture upscaling networks which have a powerful capability to establish and distinguish a complex mapping between the low-resolution input image and high-quality structure and texture, while preserving more contextual information without any prior information regarding the low-resolution input image. Quantitative and qualitative analyses of the proposed scalable image decomposition network validate that the proposed method is stable and robust against blurring and staircase effects by separating texture and structure upscaling networks in real-time. The predicted upscaled structure and texture images can be used to a variety of applications, such as image abstraction, detail enhancement, and pencil sketching.

A comparative analysis of denoising algorithms for extragalactic imaging surveys

Aims.We present a comprehensive analysis of the performance of noise-reduction (denoising) algorithms to determine whether they provide advantages in source detection, mitigating noise on extragalactic survey images.Methods.The methods we analyze here are representative of different algorithmic families: Perona-Malik filtering, bilateral filter, total variation denoising, structure-texture image decomposition, non-local means, wavelets, and block-matching We tested the algorithms on simulated images of extragalactic fields with resolution and depth typical of theHubble,Spitzer, andEuclidSpace Telescopes, and of ground-based instruments. After choosing their best internal parameters configuration, we assessed their performance as a function of resolution, background level, and image type, in addition to testing their ability to preserve the objects fluxes and shapes. Finally, we analyze, in terms of completeness and purity, the catalogs that were extracted after applying denoising algorithms on a simulatedEuclidWide Survey VIS image and on real H160 andK-band (HAWK-I) observations of the CANDELS GOODS-South field.Results.Denoising algorithms often outperform the standard approach of filtering with the point spread function (PSF) of the image. Applying structure-texture image decomposition, Perona-Malik filtering, the total variation method by Chambolle, and bilateral filtering on theEuclid-VIS image, we obtain catalogs that are both more pure and complete by 0.2 magnitude than those based on the standard approach. The same result is achieved with the structure-texture image decomposition algorithm applied on the H160 image. The relative advantage of denoising techniques with respect to PSF filtering rises with increasing depth. Moreover, these techniques better preserve the shape of the detected objects with respect to PSF smoothing.Conclusions.Denoising algorithms provide significant improvements in the detection of faint objects and enhance the scientific return of current and future extragalactic surveys. We identify the most promising denoising algorithms among the 20 techniques considered in this study.

Image compression using adaptive multiresolution image decomposition algorithm

With the growth of modern digital technologies, demand for transmission multimedia and digital images, which require more storage space and transmission bandwidth, has been increased rapidly. Hence, developing new image compression techniques for reducing data size without degrading the quality of the image, has gained a lot of interest recently. In this study, an adaptive multiresolution image decomposition (AMID) algorithm is proposed and its application for image compression is explored. The developed algorithm is capable of decomposing an image along the vertical, horizontal, and diagonal directions using the pyramidal multiresolution scheme. Compared to the wavelet transform, the AMID can be used for decimation with the guarantees of perfect signal reconstruction. Furthermore, the application of the AMID for image compression is explored and its performance is compared with the state-of-the-art image compression techniques. The performance of compression method is evaluated using peak signal-to-noise ratio and compression ratio. Experimental results have shown promising performance compared with the results of using other image compression approaches.

Research on real-time data transmission and multi-scale video image decomposition of embedded optical sensor array based on machine learning

Aiming at the research of real-time data transmission and multi-scale image decomposition of embedded optical sensor array, the principle, method and fusion strategy of multi-sensor image fusion are studied comprehensively, thoroughly and systematically by combining the imaging characteristics of source image with multi-scale geometric analysis tools using machine learning algorithm. A new quality scalable video image coding framework is also proposed in this paper, which is implemented by a multi-scale online dictionary learning algorithm based on structured sparse video signals. For the purpose of different types of images and image fusion, a new high quality scalable video image coding framework based on machine learning algorithm is proposed on the basis of comprehensive analysis of prior information such as imaging mechanism of image sensor and imaging characteristics of source image. A multi-scale online dictionary learning algorithm based on machine learning for sparse video signal structure is proposed. Through the hierarchical structure of wavelet decomposition, the searching domain of online learning is optimized to a hierarchical sparse block, and its sparse representation coefficients are obtained by using machine learning sparse coding idea. The real-time data transmission of embedded optical sensor array based on machine learning and multi-scale image decomposition algorithm proposed in this paper have good fusion performance, which is of great significance for further research and engineering application of image fusion technology.

Robust spiking cortical model and total-variational decomposition for multimodal medical image fusion

In field of medical image, a large number of related works demonstrated that the decomposition theories (such as multi-scale transform (MST) techniques) have better performance in preserving details and structures information. However, the most of hierarchical structure-based methods fail to appropriately separate energy and detailed information, they even produce some unsatisfied effects such as low-contrast resolution and anamorphose. To address the above issues, we present a modified image decomposition algorithm to get a good trade-off between speed and performance. Specifically, it utilizes the total-variational transform into moving least squares method (TV-MLS), which can make the result more robust to noise and fully reserve the dominant structure. Firstly, we perform total-variational decomposition on the source images, and yield a series of detail layers and base layers. Next, the layers are fused using the robust adaptive dual-channel spiking cortical model (RA-DCSCM), CNNs and the fine-designed fusion rule. Eventually, we provide extensive qualitative analysis for the proposed fusion scheme in both subjective visual inspection and quality metrics to verify that our method is more competitive against other existing methods.

3D face mask presentation attack detection based on intrinsic image analysis

Face presentation attacks have become a major threat against face recognition systems and many countermeasures have been proposed over the past decade. However, most of them are devoted to 2D face presentation attack detection, rather than 3D face masks. Unlike the real face, the 3D face mask is usually made of resin materials and has a smooth surface, resulting in reflectance differences. Therefore, in this study, the authors propose a novel 3D face mask presentation attack detection method based on analysis of image reflectance. In the proposed method, the face image is first processed with intrinsic image decomposition algorithm to compute its reflectance image. Then, the intensity distribution histograms are extracted from three orthogonal planes to represent the intensity differences of reflectance images between the real face and 3D face mask. After that, given that the reflectance image of a smooth surface is more sensitive to illumination changes, 1D convolutional neural network is used to characterise how different materials or surfaces react differently to illumination changes. Extensive experiments with the public available 3DMAD database demonstrate the effectiveness of the proposed method for distinguishing a face mask from the real one and show that the detection performance outperforms other state-of-the-art methods.

Improved PSO Algorithm Based on Exponential Center Symmetric Inertia Weight Function and Its Application in Infrared Image Enhancement

In this paper, an improved PSO (Particle Swarm Optimization) algorithm is proposed and applied to the infrared image enhancement. The contrast of infrared image is enhanced while the image details are preserved. A new exponential center symmetry inertia weight function is constructed and the local optimal solution jumping mechanism is introduced to make the algorithm consider both global search and local search. A new image enhancement method is proposed based on the advantages of bi-histogram equalization algorithm and dual-domain image decomposition algorithm. The fitness function is constructed by using five kinds of image quality evaluation factors, and the parameters are optimized by the proposed PSO algorithm, so that the parameters are determined to enhance the image. Experiments showed that the proposed PSO algorithm has good performance, and the proposed image enhancement method can not only improve the contrast of the image, but also preserve the details of the image, which has a good visual effect.

Discrete Hermite Wavelet Filters with Prove Mathematical Aspects

This article describes a new image processing method in order to enhance the images under testing based on discrete Hermite wavelet filter. Discrete Hermite Wavelet Transform, which was used in image processing, including compression and noise removal was used after a number of theories proved to be mathematically ready for use in image processing. Through Discrete Hermite Wavelet Transform at different levels by finding a new filter and using it to find peak-to-noise ratio values (PSNR), compression ratio, mean square error (MSE) and bits per pixel found. Achieving a high compression ratio is acquired by using a new image decomposition algorithm. Bit reduction per pixel is obtained at the second level when increasing the level of decomposition to obtain compression ratio while PSNR is decreased with the basic wavelets, due to the features that characterize Discrete Hermite Wavelet Transform. A new filter was discovered more efficient and effective in reducing the error significantly in rebuilding, MSE and bits per pixel, the samples image is used show efficient intermittent wavelets that were built in this work. This method enabling to extract the integration matrices using Hermite wavelet operation matrix of integration that leads to improve the quality of images under testing. The obtained results for decreasing of and increasing of confirm the accuracy and effectiveness of the proposed method. These results can be used in many fields such as medicine, science treatment, compression, and noise removal images.

A New Self-Reference Image Decomposition Algorithm for Strip Steel Surface Defect Detection

Automatic vision-based defect detection on the steel surface is a challenging task due to miscellaneous patterns of defects, low contrast between the defect and the background, and so on. Image-decomposition-based method can analyze the structure and texture to inspect the defective objects. Currently, the state of the art of image decomposition-based defect detection methods is the one guided by a given fixed template. However, a fixed template cannot be suitable for all situations. In this article, a new self-reference template-guided image decomposition algorithm for strip steel surface defect detection is developed. Combined with the statistical characteristics of a large number of defect-free images, a specific template can be built for each test defect image. Then, a total variation (TV)-based image decomposition algorithm guided by the self-reference template is developed to decompose the test image into the structural component and textural component. Moreover, the decomposition is optimized by developing a new index-gradient similarity to measure the similarity between the self-reference template and decomposed textural component. Experimental results show that the proposed method can detect various types of defects on the homogeneously textured surface, including miscellaneous defects, even for tiny defects and under the low contrast condition, and the precision, recall, and F-measure of the proposed algorithm are better than the state-of-the-art algorithms.

Analysis-synthesis dictionary pair learning and patch saliency measure for image fusion

In image fusion, Sparse Representation (SR)-based method has received extensive attention because of its excellent performance. In such method, the quality of over-complete dictionary and the fusion rules of SR coefficients are the most important factors that affect the final fusion quality. However, the traditional dictionary learning methods are usually designed based on Synthesis Sparse Representation (SSR), and this strategy ignores the complementary between SSR and Analysis Sparse Representation (ASR). To construct two dictionaries with powerful representation capability and discriminative capability, we integrate the complementary representation mechanisms of analysis and synthesis SR into our dictionary learning and image decomposition. Then we develop a novel dictionary learning and image decomposition algorithm for image fusion. Moreover, the traditional SR-based fusion method often adopts the simple principle of maximum absolute value in the fusion of SR coefficients, which leads to a fused result with poor visual quality. To this end, we propose to fuse the coding coefficients of the major structure and edge detail components according to the saliency measure of the corresponding patches. Experimental results show that the proposed method can better preserve the image information and improve the image contrast.

Image Processing Algorithm Based on Bi-dimensional Local Mean Decomposition

Because images contain rich characteristic information, adaptive image decomposition algorithms are necessary to achieve multi-scale extraction of image information in multi-scale image decomposition processing. For this reason, based on local mean decomposition (LMD), which has good self-adaptive characteristics, this paper proposes a new adaptive image processing algorithm, bi-dimensional local mean decomposition (BLMD). BLMD can decompose the original image into multiple bi-dimensional product functions (BPFs). Aiming at the decomposition of BLMD, this paper proposes targeted solutions and designs for the extraction of extremum points, screening process interpolation methods, and decomposition and stop conditions involved in BLMD. After fully recognizing the self-adaptive and multi-scale characteristics of BLMD, this paper proposes a variable neighborhood window method to obtain the extreme points in the decomposition process and uses fractal theory to interpolate the image and obtain the corresponding mean surface and other information. Then, the number of non-coincident extreme points on the zero-valued plane projection between adjacent surfaces in the screening process is counted and analyzed, and a stop condition that matches the characteristics of the image is given to ensure the BPF component obtained by decomposition accurately reflects certain feature information of an image. Finally, the BLMD proposed in this paper is formed. Empirical analysis shows this method can quickly decompose and maintain the characteristics of data-drivenness, adaptability and scale consistency of LMD; it can also avoid the disadvantages of other adaptive processing algorithms, such as the bi-dimensional intrinsic mode function obtained by the decomposition of bi-dimensional empirical mode decomposition and the residual failing to completely contain the feature information of the original image.

Objective Evaluation of Image Decomposition Algorithms for Depth Map Upsampling

Depth map upsampling plays an essential role in various three-dimensional (3D) image and video applications such as multi-view rendering and 3D scene modeling. Most of existing depth map upsampling methods have suggested to use a color image as a guide. Recently, in our previous work, the use of a structure component obtained from image decomposition rather than the color image has proven to be very powerful in the task of depth map upsampling. In this paper, to determine how image decomposition algorithms can be best used for depth map upsampling, we conducted a comprehensive comparative study. More precisely, the purpose of this study is to present an “objective” evaluation of recent promising image decomposition methods in terms of the performance of depth map upsampling. This is, to the best of our knowledge, the first experimental comparative demonstration on the performance of depth map upsampling enhanced with several different image decomposition models. We investigated eight different promising recent image decomposition approaches under the same experimental setup. From our quantitative comparison, we can obtain novel and valuable insights into the image decomposition-based depth map upsampling: (1) the best image decomposition solution for depth map upsampling depends on the scaling factor of upsampling, (2) the guided filter-based image decomposition method gives rise to the best performance for lower scaling factors, whereas the tree filter-based image decomposition method leads to the best upsampling performance for higher scaling factors, and (3) the performance of image decomposition-based depth upsampling is not sensitive to image features. We believe that this comprehensive comparative study serves as a reference point and guide for developers and practitioners in choosing an appropriate image decomposition technique adopted for building depth map upsampling systems.

Compressed sensing image restoration algorithm based on improved SURF operator

Abstract Aiming at the excellent descriptive ability of SURF operator for local features of images, except for the shortcoming of global feature description ability, a compressed sensing image restoration algorithm based on improved SURF operator is proposed. The SURF feature vector set of the image is extracted, and the vector set data is reduced into a single high-dimensional feature vector by using a histogram algorithm, and then the image HSV color histogram is extracted.MSA image decomposition algorithm is used to obtain sparse representation of image feature vectors. Total variation curvature diffusion method and Bayesian weighting method perform image restoration for data smoothing feature and local similarity feature of texture part respectively. A compressed sensing image restoration model is obtained by using Schatten-p norm, and image color supplement is performed on the model. The compressed sensing image is iteratively solved by alternating optimization method, and the compressed sensing image is restored. The experimental results show that the proposed algorithm has good restoration performance, and the restored image has finer edge and texture structure and better visual effect.