Edge-preserving Image Research Articles

Region-Level SAR Image Segmentation Based on Edge Feature and Label Assistance

This paper proposes a novel segmentation algorithm for synthetic aperture radar (SAR) images. The algorithm performs region-level segmentation based on edge feature and label assistance (REFLA). It demonstrates improved performance in terms of segmentation accuracy while better preserving image edges. Firstly, an edge detection scheme is implemented, which fuses information from two advanced edge detection methods, thereby obtaining a more precise edge strength map (ESM). Secondly, a Canny algorithm is performed to divide the SAR image into edge regions and homogeneous regions, and different smoothing templates are selected according to pixel positions. Therefore, an anisotropic smoothing on the SAR image can be achieved, aiming at suppressing the noise within targets while also accurately maintaining the target boundaries. Thirdly, K-means clustering is applied on the smoothed result, to generate an initial set of labels. Using ESM and the initial labels as inputs, a watershed transformation and a majority voting strategy are employed to realize an initial segmentation at the region level. Finally, a label-aided region merging (LaRM) strategy is used to correctly segment the wrongly labeled regions, to give the final segmentation result. The LaRM, with merging rules based on label rather than gray characteristics, can avoid the need for calculating a large number of complex formulae, thus accelerating the region merging. Results are presented of experiments, on both simulated and real SAR images, in which the proposed REFLA method is compared against six state-of-the-art algorithms from the literature. REFLA achieves higher accuracy, while better retaining the image edges.

Two-Pass Bilateral Smooth Filtering for Remote Sensing Imagery

Bilateral filtering has been adopted for edge-preserving image smoothing and achieved the state-of-the-art performance. Most of the existing bilateral filters (BFs), however, focus on accelerating brute-force implementation but not on smoothing quality. In this letter, we propose a two-pass (TP) BF, TP-based BF, and an adaptive control scheme of range kernels for noise-invariant edge-preserving image smoothing. Specifically, the TP-based BF is composed of two bilateral filtering operations, which are, respectively, in charge of coarse context extraction and fine structure refinement. The control scheme of range kernels guides the TP bilateral mechanism to eliminate first high-frequency noisy pixels and then explore contribution between pixels from clean contexts. Experimental results on four aerial-imagery benchmark data sets show that our TP-based BF outperforms the existing BFs in terms of both feature- and gradient-aware measures.

Learning edge-preserved image stitching from multi-scale deep homography

Image stitching is a classical and challenging technique in computer vision, which aims to generate an image with a wide field of view. The traditional methods heavily depend on feature detection and require the feature points to be dense and evenly distributed in the image, leading to poor robustness in low-texture scenes. Learning methods are rarely studied due to the unavailability of ground truth stitched results, showing unreliable performance on real-world datasets. In this paper, we propose an image stitching learning framework, which consists of a multi-scale deep homography module and an edge-preserved deformation module. First, we design a multi-scale deep homography module to estimate the accurate homography progressively from coarse to fine. After that, an edge-preserved deformation module is designed to learn the deformation rules of image stitching from edge to content, generating the stitched image with artifacts eliminated. Besides, the proposed supervised learning framework can stitch images of arbitrary resolutions and demonstrate good generalization capability in real-world images. Experiments show that our superiority to the existing homography solutions and image stitching algorithms.

A non‐convex ternary variational decomposition and its application for image denoising

A non-convex ternary variational decomposition model is proposed in this study, which decomposes the image into three components including structure, texture and noise. In the model, a non-convex total variation (NTV) regulariser is utilised to model the structure component, and the weaker G and E spaces are used to model the texture and noise components, respectively. The proposed model provides a very sparse representation of the structure in total variation (TV) transform domain due to the use of non-convex regularisation and cleanly separates the texture and noise since two different weaker spaces are used to model these two components, respectively. In image denoising application, the proposed model can successfully remove noise while effectively preserving image edges and constructing textures. An alternating direction iteration algorithm combining with iteratively reweighted l1 algorithm, projection algorithm and wavelet soft threshold algorithm is introduced to effectively solve the proposed model. Numerical results validate the model and the algorithm for both synthetic and real images. Furthermore, compared with several state-of-the-art image variational restoration models, the proposed model yields the best performance in terms of the peak signal to noise ratio (PSNR) and the mean structural similarity index (SSIM).

Combined total variation of first and fractional orders for Poisson noise removal in digital images

Introduction: Many methods have been proposed to handle the image restoration problem with Poisson noise. A popular approach to Poissonian image reconstruction is the one based on Total Variation. This method can provide significantly sharp edges and visually fine images, but it results in piecewise-constant regions in the resulting images. Purpose: Developing an adaptive total variation-based model for the reconstruction of images contaminated by Poisson noise, and an algorithm for solving the optimization problem. Results: We proposed an effective way to restore images degraded by Poisson noise. Using the Bayesian framework, we proposed an adaptive model based on a combination of first-order total variation and fractional order total variation. The first-order total variation model is efficient for suppressing the noise and preserving the keen edges simultaneously. However, the first-order total variation method usually causes artifact problems in the obtained results. To avoid this drawback, we can use high-order total variation models, one of which is the fractional-order total variation-based model for image restoration. In the fractional-order total variation model, the derivatives have an order greater than or equal to one. It leads to the convenience of computation with a compact discrete form. However, methods based on the fractional-order total variation may cause image blurring. Thus, the proposed model incorporates the advantages of two total variation regularization models, having a significant effect on the edge-preserving image restoration. In order to solve the considered optimization problem, the Split Bregman method is used. Experimental results are provided, demonstrating the effectiveness of the proposed method. Practical relevance: The proposed method allows you to restore Poissonian images preserving their edges. The presented numerical simulation demonstrates the competitive performance of the model proposed for image reconstruction. Discussion: From the experimental results, we can see that the proposed algorithm is effective in suppressing noise and preserving the image edges. However, the weighted parameters in the proposed model were not automatically selected at each iteration of the proposed algorithm. This requires additional research.

Edge-Preserving Denoising of Image Sequences.

To monitor the Earth’s surface, the satellite of the NASA Landsat program provides us image sequences of any region on the Earth constantly over time. These image sequences give us a unique resource to study the Earth’s surface, changes of the Earth resource over time, and their implications in agriculture, geology, forestry, and more. Besides natural sciences, image sequences are also commonly used in functional magnetic resonance imaging (fMRI) of medical studies for understanding the functioning of brains and other organs. In practice, observed images almost always contain noise and other contaminations. For a reliable subsequent image analysis, it is important to remove such contaminations in advance. This paper focuses on image sequence denoising, which has not been well-discussed in the literature yet. To this end, an edge-preserving image denoising procedure is suggested. The suggested method is based on a jump-preserving local smoothing procedure, in which the bandwidths are chosen such that the possible spatio-temporal correlations in the observed image intensities are accommodated properly. Both theoretical arguments and numerical studies show that this method works well in the various cases considered.

A second-order nonlocal regularized variational model for multiframe image super-resolution

Abstract In this paper, we propose a new nonlocal super-resolution (SR) model which is a combination of the nonlocal total variation (TV) regularization and the nonlocal p-Laplacian term (with p = 2). This choice is motivated by the success of the nonlocal TV term in preserving image edges and the efficiency of the nonlocal p-Laplacian term in preserving the image texture. To ensure the convergence of the proposed optimization SR problem, we prove the existence and uniqueness of a solution in a well-posed framework. In addition, to resolve the encountered minimization problem, we proposed a modified primal-dual algorithm and numerical results are also given to show the performance of the proposed approach.

A Novel Iterative MLEM Image Reconstruction Algorithm Based on Beltrami Filter: Application to ECT Images.

The implementation of emission-computed tomography (ECT), including positron emission tomography and single-photon emission-computed tomography, has been an important research topic in recent years and is of significant and practical importance. However, the slow rate of convergence and the computational complexity have severely impeded the efficient implementation of iterative reconstruction. By combining the maximum-likelihood expectation maximization (MLEM) iteratively along with the Beltrami filter, this paper proposes a new approach to reformulate the MLEM algorithm. Beltrami filtering is applied to an image obtained using the MLEM algorithm for each iteration. The role of Beltrami filtering is to remove mainly out-of-focus slice blurs, which are artifacts present in most existing images. To improve the quality of an image reconstructed using MLEM, the Beltrami filter employs similar structures, which in turn reduce the number of errors in the reconstructed image. Numerical image reconstruction tomography experiments have demonstrated the performance capability of the proposed algorithm in terms of an increase in signal-to-noise ratio (SNR) and the recovery of fine details that can be hidden in the data. The SNR and visual inspections of the reconstructed images are significantly improved compared to those of a standard MLEM. We conclude that the proposed algorithm provides an edge-preserving image reconstruction and substantially suppress noise and edge artifacts.

A novel grid regression demodulation method for radiographic grid artifact correction.

In x-ray radiography, the commonly used antiscatter grid for enhancing image quality causes artifacts in the form of periodic noises, such as shadows, cutoff, and Moiré fringes. Software degridding is traditionally performed via linear or homomorphic filtering in the spectral domain. These methods inevitably result in image blurring, information loss, and distortion, thus hindering detection and assessment of diseases. We seek effective and practical solutions for grid artifact correction based on spatial-domain analysis toward high-quality imaging. By analyzing the physical process of grid artifact formation, we track down the root of the problem associated with spectral filtering. We propose the grid regression demodulation (GRD). The degridding cost is forged as a functional of the latent x-ray photon image and parametric grid model characterizing grid transmission property. Regularization on the grid spectra is incorporated. We devise optimization algorithms for artifact correction and grid pattern estimation. GRD decouples the partially overlapped spectra of the grid and anatomy, and removes the artifacts independently, thus restoring the underlying clinically relevant data. Method efficacy is demonstrated using simulated and real data. GRD effectively preserves image edges, textures, and patterns while removing grid artifacts. For the known ground truth setting, GRD gives a near-perfect correction. For real data, GRD is capable of correcting not only the primary grid artifacts, but also the higher grid harmonic artifacts while keeping image content unaltered, which is unachievable by the other methods. Our method has low residual errors and exhibits a successful demodulation effect without introducing additional artifacts, while ringing or cilia artifacts are present in the others. The proposed method outperforms the prevalent transform techniques for correcting grid artifacts in digital radiography. It is self-sustained and self-adaptive to a range of targets and beam quality. Our approach is advantageous in restoring the latent image while suppressing grid noises. It retrieves the true scale factor of the degridded data, which is unattainable via any spectral filtering techniques. This work unlocks a promising venue to improve and upgrade low-dose medical radiographic imaging technology.

Multi-modal medical image fusion framework using co-occurrence filter and local extrema in NSST domain

The fusion of vital imaging information has turned into a primary issue for biomedical applications. In this paper, a new multi-modality medical image fusion method is proposed in the shearlet domain. In the proposed algorithm, input images are decomposed using Non-subsampled shearlet transform (NSST) to get low and high frequencies components. A novel procedure to decompose and combine the base layers and detail layers using the local extrema (LE) approach is used and fused using a Co-occurrence filter (CoF) in low-frequency components. In high-frequency components, an edge-preserving image fusion strategy is performed using sum modified Laplacian (SML) to combine the high-frequency coefficients. The experimental outcomes and comparative analysis are performed over the Multi-modal medical image dataset using proposed and existing methods. It is exhibited through test outcomes and assessments that the proposed technique beats the cutting-edge fusion strategies concerning edge preservation in subjective and objective assessment criteria.

Edge-preserving smoothing algorithm based on <italic>l</italic><sub>0</sub> gradient minimization of image-patch

Image smoothing algorithm based on l₀ gradient minimization can smooth details and textures of image while preserving edges. Since the algorithm uses image gradient to determine the smoothed component, the region with smaller image gradients (weak edge) can be smoothed. However, the region with larger image gradients (strong edge) will be preserved. In order to overcome this drawback, we propose an image-patch based l₀ gradient minimization image smoothing algorithm (IP-<italic>l</italic>₀). Instead of globally smoothing the input image, our algorithm smoothed each image patch first and then combined the smoothed patches together to obtain the final smoothed image. The weight parameters in the objective function used to smooth each image patch are dynamically changed according to the local statistics of image patch, so that the patches containing strong textures will be smoothed with greater force, and vice versa. The experimental results show that compared to the original l₀ gradient minimization algorithm and several other state-of-the-art edge-preserving image smoothing algorithms including the local Laplacian filter based algorithm, the relative total variation based algorithm, the tree filter based algorithm, and two kinds of deep-learning based smoothing algorithms, the proposed algorithm can effectively smooth strong textures and well preserve weak edges or structures, and the ability of edge preservation and texture smoothing is better than other algorithms.

Two iterative methods for reverse image filtering

We consider the problem of recovering an original image {varvec{x}} from its filtered version {varvec{y}}={varvec{f}}({varvec{x}}), assuming that the internal structure of the filter {varvec{f}}(cdot ) is unknown to us (i.e., we can only query the filter as a black-box and, for example, cannot invert it). We present two new iterative methods to attack the problem, analyze, and evaluate them on various smoothing and edge-preserving image filters.

Image Reconstruction Using Variable Exponential Function Regularization for Wide-Field Polarization Modulation Imaging

Polarization modulation imaging technology plays an important role in microscopic super-resolution imaging. However, the specimen medium contains retardancy, while charge-coupled devices may provide discrete under-sampling, and the coupled wavefronts consisting of the polarization state of the light and the anisotropic distribution of the specimen can lead to vectorial phase fitting degradation. Considering that the point spread function (PSF) of the main degradation parts can be regarded as an asymmetric generalized Gaussian distribution with uncertain parameters, an adaptive image reconstruction method is proposed based on variable exponential function regularization. The proposed method concentrates on the diversity of the PSF and uses a variable exponent regularization to improve flexibility of the kernel. Moreover, it can balance image edge preservation and provide staircase artifact suppression, which reduces the over- and under-reconstruction of the microscopic images effectively. By optimizing the Split–Bregman algorithm, we create an efficient method that minimizes the iterative loss function under the premise of achieving high estimation accuracy. Compared with other methods, the experimental results reveal better effectiveness and robustness of the proposed method, with improvements of 18% in the peak signal-to-noise ratio, 21% in the structural similarity index measurement, and 337% in the mean structural similarity index measurement.

No-Reference Physics-Based Quality Assessment of Polarization Images and Its Application to Demosaicking.

Assessing the quality of polarization images is of significance for recovering reliable polarization information. Widely used quality assessment methods including peak signal-to-noise ratio and structural similarity index require reference data that is usually not available in practice. We introduce a simple and effective physics-based quality assessment method for polarization images that does not require any reference. This metric, based on the self-consistency of redundant linear polarization measurements, can thus be used to evaluate the quality of polarization images degraded by noise, misalignment, or demosaicking errors even in the absence of ground-truth. Based on this new metric, we propose a novel processing algorithm that significantly improves demosaicking of division-of-focal-plane polarization images by enabling efficient fusion between demosaicking algorithms and edge-preserving image filtering. Experimental results obtained on public databases and homemade polarization images show the effectiveness of the proposed method.

Dictionary learning based image-domain material decomposition for spectral CT

The potential huge advantage of spectral computed tomography (CT) is that it can provide accurate material identification and quantitative tissue information by material decomposition. However, material decomposition is a typical inverse problem, where the noise can be magnified. To address this issue, we develop a dictionary learning based image-domain material decomposition (DLIMD) method for spectral CT to achieve accurate material components with better image quality. Specifically, a set of image patches are extracted from the mode-1 unfolding of normalized material images decomposed by direct inversion to train a unified dictionary using the K-SVD technique. Then, the DLIMD model is established to explore the redundant similarities of the material images, where the split-Bregman is employed to optimize the model. Finally, more constraints (i.e. volume conservation and the bounds of each pixel within material maps) are integrated into the DLIMD model. Numerical phantom, physical phantom and preclinical experiments are performed to evaluate the performance of the proposed DLIMD in material decomposition accuracy, material image edge preservation and feature recovery.

Multi-focus image fusion based on L1 image transform

In this paper, a new multi-focus image fusion algorithm based on L1 image transform is proposed. A distinctive advantage of the proposed algorithm is that an edge-preserving image decomposition (EPID) framework is constructed by introducing a L1-norm based image transform, which can not only effectively preserve and sharpen salient edges and ridges while eliminating insignificant details in the smoothing subband, but also maintain the detail information in the detail subbands. Another advantage is that the fusion rules for the smoothing subband and detail subbands are designed respectively according to their own characteristics so that both the structure and detail information can be fully retained. The fusion process mainly consists of the following three steps. Firstly, each source image is decomposed into a smoothing subband and several detail subbands by utilizing the EPID framework. Then, the subbands are fused by different fusion rules respectively to obtain a fused smoothing subband and a series of fused detail subands. Finally, the final fused image is reconstructed with less distortions by synthesizing the fused smoothing subband and a series of fused detail subands. Experimental results demonstrate the superiority of the proposed algorithm in terms of the visual perception and objective assessments.

A Single Image Mosquito Streaks Removal Method of High-Speed Railway Traction Substation Using DCNN

With the proposal of intelligent high-speed railway, the research on remotely monitoring the intelligent traction substation becomes a key subject of high-speed railway safe operation. However, mosquitos adhered to the glass window or camera lens can severely hamper the visibility of a background scene, and degrade images considerably. Therefore, we propose a single image mosquito streaks removal method of high-speed railway traction substation based on the deep convolutional neural network. First, we employ guided filter to split input image into smoothing image and edge-preserving image. Then, the edge-preserving image is fed into our designed convolutional neural network to obtain learning map, which solves problems of background interference and focuses the model on the structure of rain streak in images, and the clean image is finally generated through adding the input image and learning map. The experiment results on Heishan traction substation real datasets show the effectiveness of our proposed method.

Real-time Image Smoothing via Iterative Least Squares

Edge-preserving image smoothing is a fundamental procedure for many computer vision and graphic applications. There is a tradeoff between the smoothing quality and the processing speed: the high smoothing quality usually requires a high computational cost, which leads to the low processing speed. In this article, we propose a new global optimization based method, named iterative least squares (ILS), for efficient edge-preserving image smoothing. Our approach can produce high-quality results but at a much lower computational cost. Comprehensive experiments demonstrate that the proposed method can produce results with little visible artifacts. Moreover, the computation of ILS can be highly parallel, which can be easily accelerated through either multi-thread computing or the GPU hardware. With the acceleration of a GTX 1080 GPU, it is able to process images of 1080p resolution (1920 × 1080) at the rate of 20fps for color images and 47fps for gray images. In addition, the ILS is flexible and can be modified to handle more applications that require different smoothing properties. Experimental results of several applications show the effectiveness and efficiency of the proposed method. The code is available at https://github.com/wliusjtu/Real-time-Image-Smoothing-via-Iterative-Least-Squares.

Image deblocking via shape-adaptive low-rank prior and sparsity-based detail enhancement

At low bit rates, visually annoying blocking artifacts are usually introduced in JPEG compressed images. In this paper, we proposed an image deblocking method combined with the shape-adaptive low-rank (SALR) prior, the quantization constraint (QC) prior and sparsity-based detail enhancement. We firstly design a deblocking model to obtain initial deblocked images under the maximum a posteriori (MAP) framework. More specifically, with the assumption of Gaussian quantization noise, the SALR prior is utilized to effectively separate signal from noise and preserve image edges. Compared with previous low rank priors, the SALR reconstructs a better result via shape adaptive blocks. The QC prior is also adopted to avoid over-smoothing and to enable a more accurate estimation. Finally, by extracting features of external images, the mapping matrix of sparse dictionary pairs is trained to enhance image details. Extensive experimental results demonstrate that the proposed deblocking method has superior performances in both subjective vision and objective quality.

Edge‐aware image filtering using a structure‐guided CNN

Image filtering is a fundamental preprocessing step for accurate, robust computer vision applications such as image segmentation, object classification, and reconstruction. However, many convolutional neural network (CNN)-based methods tend to lose significant edge information in the output layer, and generate undesired artefacts in the feature extraction layers. This study presents a deep CNN model for edge-aware image filtering. The proposed network model consists of three sub-networks: (i) feature extraction, (ii) convolution artefact removal, and (iii) structure extraction networks. The proposed network model has an end-to-end trainable architecture that does not need any post-processing steps. Especially, the structure extraction network can successfully preserve significant edges. The proposed filter outperforms state-of-the-art denoising filters in terms of both objective and subjective measures, and can be used for various image enhancement and restoration problems such as edge-preserving smoothing, image denoising, deblurring, and deblocking.