Removal Of Random-valued Impulse Noise Research Articles

영상의 영역 특성을 이용한 랜덤 임펄스 잡음 제거 필터

영상의 획득 및 전송 과정에서 이미지 센서, 메모리 저장 장치 등 다양한 장비의 오류로 인해 일부 화소값이 부정확한 값을 나타낼 수 있다. 훼손된 화소는 에지 검출, 영역 분할, 객체 인식 등과 같은 영상 처리 작업의 신뢰도를 감소시킬 수 있어 영상 처리 과정 중 잡음 제거 기술을 필수적으로 사용해야 한다. 이러한 임펄스 잡음을 제거하기 위한 많은 비선형 필터가 제안되었지만, 화소값 범위가 넓은 이미지의 경우 원 영상의 표준편차가 크기 때문에 잡음 판단이 어렵다. 따라서 본 논문은 로컬 마스크의 영역 특성을 분석하여, 영역에 따라 랜덤 임펄스 잡음을 검출하는 필터를 제안하였다. 제안한 필터는 두 차례에 걸친 잡음 판단을 통해 영역을 구분하였으며, 표준편차를 사용하여 특수한 영역의 잡음을 판별하였다. 또한, 가중치 평균을 사용하여 잡음 영상을 복원하였다.

Fine-tuning convolutional neural network based on relaxed Bayesian-optimized support vector machine for random-valued impulse noise removal

Denoising convolutional neural network (DnCNN) has achieved competitive denoising performance for Gaussian noise using residual learning. The same idea can also be applied in the impulse noise (IN) removal by changing the training data. However, as a discriminative learning-based algorithm, DnCNN lacks flexibility in removing spatially varying noise, such as random-valued impulse noise (RVIN). To address this problem, we propose a fine-tuning CNN based on relaxed Bayesian-optimized support vector machine (BO-SVM) for RVIN removal using a relaxed BO-SVM as a detector and a fine-tuning DnCNN as a denoiser. The relaxed BO-SVM is developed to classify all pixels of the test image into two categories: noise and noise-free. A relaxed rule is provided to preserve the information of slightly corrupted pixels, and the system is trained with an optimal feature set, which comprises of newly introduced local ternary pattern with previously used gray-level difference, average background difference, and median value difference. The key hyperparameters of the relaxed SVM are obtained using the Bayesian optimization technique. Then the DnCNN works for image reconstruction through fine-tuning experiments by the use of the corrupted pixels identified by the detector. The experimental results show that the proposed method reports promising results on RVIN removal over other state-of-the-art denoising methods.

An efficient method to remove mixed Gaussian and random-valued impulse noise.

Mixed Gaussian and Random-valued impulse noise (RVIN) removal is still a big challenge in the field of image denoising. Existing denoising algorithms have defects in denoising performance and computational complexity. Based on the improved “detecting then filtering” strategy and the idea of inpainting, this paper proposes an efficient method to remove mixed Gaussian and RVIN. The proposed algorithm contains two phases: noise classification and noise removal. The noise classifier is based on Adaptive center-weighted median filter (ACWMF), three-sigma rule and extreme value processing. Different from the traditional “detecting then filtering” strategy, a preliminary RVIN removal step is added to the noise removal phase, which leads to three steps in this phase: preliminary RVIN removal, Gaussian noise removal and final RVIN removal. Firstly, RVIN is processed to obtain a noisy image approximately corrupted by Gaussian noise only. Subsequently, Gaussian noise is re-estimated and then denoised by Block Matching and 3D filtering method (BM3D). At last, the idea of inpainting is introduced to further remove RVIN. Extensive experimental results demonstrate that the proposed method outperforms quantitatively and visually to the state-of-the-art mixed Gaussian and RVIN removal methods. In addition, it greatly shortens the computation time.

A gray level indicator-based nonlinear diffusion equation for the removal of random-valued impulse noise

A gray level indicator-based nonlinear diffusion equation for the removal of random-valued impulse noise

Random-Valued Impulse Noise Detection and Removal based on Local Statistics of Images

Random-valued impulse noise removal from images is a challenging task in the field of image processing and computer vision. In this paper, an effective three-step noise removal method was proposed using local statistics of grayscale images. Unlike most existing denoising algorithms that assume the noise density is known, our method estimated the noise density in the first step. Based on the estimated noise density, a noise detector was implemented to detect corrupted pixels in the second step. Finally, a modified weighted mean filter was utilized to restore the detected noisy pixels while leaving the noise-free pixels unchanged. The noise removal performance of our method was compared with 10 well-known denoising algorithms. Experimental results demonstrated that our proposed method outperformed other denoising algorithms in terms of noise detection and image restoration in the vast majority of the cases.

Dual Mixture Model Based CNN for Image Denoising.

Non-Gaussian residual error and noise are common in the real applications, and they can be efficiently addressed by some non-quadratic fidelity terms in the classic variational method. However, they have not been well integrated into the architectures design in the convolutional neural networks (CNN) based image denoising method. In this paper, we propose a deep learning approach to handle non-Gaussian residual error. Our method is developed on an universal approximation property for the probability density functions of the non-Gaussian error/noise. By considering the duality of the maximum likelihood estimation for the non-Gaussian error, an adaptive weighting strategy can be derived for image fidelity. To get a good image prior, a learnable regularizer is adopted. Solving such a problem iteratively can be unrolled as a weighted residual CNN architecture. The main advantage of our method is that the weighted residual block can well handle the non-Gaussian residual, especially for the noise with non-uniformly spatial distribution. Numerical results show that it has better performance on non-Gaussian noise (e.g. Gaussian mixture, random-valued impulse noise) removal than the related existing methods.

Color Random Valued Impulse Noise Removal Based on Quaternion Convolutional Attention Denoising Network

A new quaternion convolutional attention denoising network known as DeQCANet for color random-valued impulse noise removal is proposed in this paper. First, a structural information extraction block based on a dilated convolution operation is designed to extract the structure and detail feature map. Subsequently, a quaternion convolutional neural network with a new quaternion map construction strategy is implemented to gain the color features across channels further. Finally, to integrate the global and local features, a feature enhancement block based on the attention mechanism is proposed to guide the network for impulse noise denoising. The experimental results demonstrate that the proposed denoising technique exhibits competitive performance compared to other well-known color image denoising methods.

A random-valued impulse noise removal algorithm via just noticeable difference threshold detector and weighted variation method

The paper presents a random-valued impulse noise (RVIN) removal algorithm by using distinct impulse noise detection and correction stages. In the noise detection phase, the local luminance adaption of an image is used to detect the obvious noise pixels firstly, and the just noticeable difference (JND) is introduced to detect the noise pixels hiding in image details afterward. In the noise candidate correction phase, the obvious noise pixels, almost distributing in the smooth region, are restored by a weighted mean filter. For the noise candidates hiding in image details, we adopt a weighted detail-preserving variation method to restore them. Since human visual perception and detail-preserving are taken into account during the course of the noises detection and restoration, our method outperforms some existing methods, both in visual and quantitative measurements.

A Two-Stage Algorithm for the Detection and Removal of Random-Valued Impulse Noise Based on Local Similarity

A two-stage denoising algorithm based on local similarity is proposed to process lowly and moderate corrupted images with random-valued impulse noise in this paper. In the noise detection stage, the pixel to be detected is centered and the local similarity between the pixel and each pixel in its neighborhood is calculated, which can be used as the probability that the pixel is noise. By obtaining the local similarity of each pixel in the image and setting an appropriate threshold, the noise pixels and clean pixels in the damaged image can be detected. In the image restoration stage, an improved bilateral filter based on local similarity and geometric distance is designed. The pixel detected as noise in the first stage is filtered and the new intensity value is the weighted average of all pixel intensities in its neighborhood. A large number of experiments have been conducted on different test images and the results show that compared with the mainstream denoising algorithms, the proposed method can detect and filter out the random-value impulse noise in the image more effectively and faster, while better retaining the edges and other details of the image.

Removal of random-valued impulse noise from Cerenkov luminescence images.

Cerenkov luminescence imaging(CLI) is an emerging molecular imaging technology able to optically visualize radioactive decay signals from medical isotopes and has found wide application in tumor diagnose, cancer therapy, drug development, intraoperative guidance, and so on. When Cerenkov luminescence data are collected, the high-energy particles from the radioactive nucleus will be detected by the sensitive CCD camera and lead to impulse noise. To suppress the impulse noise and improve the contrast of the useful signal to the background, the detection-based fuzzy switching median filtering framework is proposed in this paper. Several experiments were conducted respectively to investigate the statistical feature of the noise and to evaluate the performance of the proposed noise removal framework. The results show that the signal-to-noise ratio is improved after noise elimination. The proposed filtering framework outperforms the classical median filter in terms of root mean squared error and the structural similarity index. It also preserves the maximum value and the mean value in the regions of interest better than the median filter does. In addition, compared with the FLICMCDD algorithm, the proposed method works much faster while getting similar results. Graphical abstract.

Texture-preserving denoising method for the removal of random-valued impulse noise in gray-scale images

Image denoising has become one of the fundamental problems in the field of image processing. We present an iterative image denoising method named as multitexton noise identification and local dissimilarity-based noise removal that incorporates a noise restoration phase followed by a noise identification phase for gray-scale images corrupted by random-valued impulse noise (RVIN). Multiple textons of distinct orientations arranged on the basis of radial symmetry are proposed that give sharp edges of restored images. The noise identification phase works on an adaptive threshold range by employing the local statistics of textons. Moreover, the proposed method can better be utilized to deal with the uncertainty observed in identification of noisy and edge pixels. The dissimilarity of the identified corrupted pixel with its four-connected neighboring pixels is computed by considering the similarity among these four-connected pixels, to restore the gray-level intensities of identified corrupted pixels at the second phase. Standard gray-scale benchmark test images are used to assess the quantitative and visual performance of the proposed method by comparing it with state-of-the-art denoising methods. The extensive experimental results reveal that the presented method outperforms in terms of peak-signal-to-noise ratio, structural similarity index measurement, miss detection, false detection, and visual results for both lower and higher intensities of RVIN as compared to other existing techniques.

Detail-preserving switching algorithm for the removal of random-valued impulse noise

This paper presents a new algorithm for the denoising of images corrupted with random-valued impulse noise (RVIN). It employs a switching approach that identifies the noisy pixels in the first stage and then estimates their intensity values to restore them. Local statistics of the textons in distinct orientations of the sliding window are exploited to identify the corrupted pixels in an iterative manner; using an adaptive threshold range. Textons are formed by using an isometric grid of minimum local distance that preserves the texture and edge pixels of an image, effectively. At the noise filtering stage, fuzzy rules are used to obtain the noise-free pixels from the proposed tri-directional pixels to estimate the intensity values of identified corrupted pixels. The performance of the proposed denoising algorithm is evaluated on a variety of standard gray-scale images under various intensities of RVIN by comparing it with state-of-the-art denoising methods. The proposed denoising algorithm also has robust denoising and restoration power on biomedical images such as, MRI, X-Ray and CT-Scan. The extensive simulation results based on both quantitative measures and visual representations depict the superior performance of the proposed denoising algorithm for various noise intensities.

Region Adaptive Fuzzy Filter: An Approach for Removal of Random-Valued Impulse Noise

This paper proposes region adaptive fuzzy filter for removal of random-valued impulse noise (RVIN) from color images. It is observed from existing literature that the filter performance increases with improved accuracy of noise detection and better adaption of the filter parameters. Improved minimum mean value detection mechanism is proposed for better classification of noisy and nonnoisy pixels in context to the removal of RVIN from color images. The modified fuzzy filter considers the correlation among the color channels and recursively adapts itself in accordance with the local noise densities. This filter incorporates an adaption technique to determine the maximum allowable window size used during fuzzification and filtering. Rather than one efficient filtering step that has a potential for losing information, region selective second iteration of the filter is applied on highly corrupted regions so as to preserve more image details. Experimental results suggest that the proposed filter provides improved performance in terms of peak signal to noise ratio, normalized color difference, structural similarity index, feature similarity index for color images, and perceptual similarity than most of the state-of-the-art filters.

Texture-oriented image denoising technique for the removal of random-valued impulse noise

An iterative texture-oriented image denoising technique is proposed for the restoration of images corrupted with random-valued impulse noise (RVIN). The proposed technique opts a switching approach that first identifies the pixels corrupted by RVIN and then estimates their intensity values to restore the images. Textons of distinct orientations conforming to bilateral symmetry are proposed for the identification of corrupted pixels. To estimate the intensity values of the identified corrupted pixels, the textons having local similarity are used. As the textons are fundamental elements of texture perception, the proposed technique preserves the texture information of images, effectively. The performance of the proposed denoising technique is evaluated on standard benchmark test images under various intensities of RVIN by comparing it with state-of-the-art techniques. The simulation results depict the significant performance of the proposed denoising technique for low as well as higher intensities of RVIN.

Random Valued Impulse Noise Removal Using Region Based Detection Approach

Removal of random valued noisy pixel is extremely challenging when the noise density is above 50%. The existing filters are generally not capable of eliminating such noise when density is above 70%. In this paper a region wise density based detection algorithm for random valued impulse noise has been proposed. On the basis of the intensity values, the pixels of a particular window are sorted and then stored into four regions. The higher density based region is considered for stepwise detection of noisy pixels. As a result of this detection scheme a maximum of 75% of noisy pixels can be detected. For this purpose this paper proposes a unique noise removal algorithm. It was experimentally proved that the proposed algorithm not only performs exceptionally when it comes to visual qualitative judgment of standard images but also this filter combination outsmarts the existing algorithm in terms of MSE, PSNR and SSIM comparison even up to 70% noise density level.

Random Valued Impulse Noise Removal Using Region Based Detection Approach

Random Valued Impulse Noise Removal Using Region Based Detection Approach

RETRACTED ARTICLE: Cartesian vector-based directional nonparametric fuzzy filter for random-valued impulse noise removal

A new clustering technique based on most allied directional neighbors is proposed to suppress low and high-density impulse noise from digital images. Most allied neighbors exhibit a vital role in estimation as well restoration of appropriate gray level value of corrupted pixels. In first phase, most allied directional neighbors, i.e., pixels directly attached to central pixel and the directional pixels (horizontal, vertical and two diagonal directions) next to attached pixels in the processing window, are partitioned into two equal size clusters based on gradient values. Cluster with a minimum sum of gradient values (most similar neighbors) and the one with relatively large gradient values are passed to fuzzy inference system to infer the current pixel to be noisy-free, edge or a noisy. In second phase, a switching technique opts one of the three options depending upon fuzzy membership degrees and local information to restore the corrupted pixel value. A nonparametric approach based on local information for dynamic threshold setting using fuzzy logic makes the proposed filter computationally effective and adaptive to process a large number of images without user-defined parameters. Simulation results based on well-known quantitative measures indicate the supremacy of the proposed filter for random-valued impulse noise as well as salt and peppers noise.

Robust non-local median filter

This paper describes a novel image filter with superior performance on detail-preserving removal of random-valued impulse noise superimposed on natural gray-scale images. The non-local means filter is in the limelight as a way of Gaussian noise removal with superior performance on detail preservation. By referring the fundamental concept of the non-local means, we had proposed a non-local median filter as a specialized way for random-valued impulse noise removal so far. In the non-local processing, the output of a filter is calculated from pixels in blocks which are similar to the block centered at a pixel of interest. As a result, aggressive noise removal is conducted without destroying the detailed structures in an original image. However, the performance of non-local processing decreases enormously in the case of high noise occurrence probability. A cause of this problem is that the superimposed noise disturbs accurate calculation of the similarity between the blocks. To cope with this problem, we propose an improved non-local median filter which is robust to the high level of corruption by introducing a new similarity measure considering possibility of being the original signal. The effectiveness and validity of the proposed method are verified in a series of experiments using natural gray-scale images.

Adaptive impulse detection based selective window median filter for removal of random-valued impulse noise in digital images

Purpose One of the fundamental tasks in the field of image processing is image denoising. Images are often corrupted by different types of noise and the restoration of images degraded with random-valued impulse noise is still a challenging task. The paper aims to discuss these issues. Design/methodology/approach This paper presents an adaptive threshold-based impulse noise detection following by a novel selective window median filter for restoration of RVIN pixels. Findings The proposed method emphasis a local image statistics using an exponential nonlinear function with an adaptive threshold is derived from the rank-ordered trimmed median absolute difference (ROTMAD) are deliberated to detect the noisy pixels. In the filtering stage, a selective 3×3 moving window median filter is applied to restore the detected noisy pixel. Originality/value Experimental result shows that the proposed algorithm outperforms the existing state-of-art techniques in terms of noise removal and quantitative metrics such as peak signal to noise ratio (PSNR), mean absolute error (MAE), structural similarity index metric (SSIM) and miss and false detection rate.

Sparse regularization method for the detection and removal of random-valued impulse noise

In this paper, we propose a novel two-stage algorithm for the detection and removal of random-valued impulse noise using sparse representations. The main aim of the paper is to demonstrate the strength of image inpainting technique for the reconstruction of images corrupted by random-valued impulse noise at high noise densities. First, impulse locations are detected by applying the combination of sparse denoising and thresholding, based on sparse and overcomplete representations of the corrupted image. This stage optimally selects threshold values so that the sum of the number of false alarms and missed detections obtained at a particular noise level is the minimum. In the second stage, impulses, detected in the first stage, are considered as the missing pixels or holes and subsequently these holes are filled-up using an image inpainting method. Extensive simulation results on standard gray scale images show that the proposed method successfully removes random-valued impulse noise with better preservation of edges and other details compared to the existing techniques at high noise ratios.