Showing papers on "Median filter published in 2023"

Speckle Noise Reduction in Ultrasound Images

Abstract: Abstract: Ultrasound Imaging is used to examine various organs in the Human Body. However, in the process of obtaining a Ultrasound Image, Speckle noise is generated due to backscattered echo signals. So, a noise reduction method is required. To improve the speckle noise reduction, we propose a novel algorithm using characteristics of speckle noise and filtering methods based on Speckle Reducing Anisotropic Diffusion (SRAD) filtering, Discrete Wavelet Transform (DWT), Weighted Guided Image Filtering (WGIF) and Gradient Domain Guided Image Filtering (GDGIF). The SRAD filter is exploited as a preprocessing filter because it can be directly applied to a medical US image containing speckle noise without a log-compression. The wavelet domain has the advantage of suppressing the additive noise. Therefore, a homomorphic transformation is utilized to convert the multiplicative noise into additive noise. After two-level DWT decomposition is applied, to suppress the residual noise of an SRAD filtered image, GDGIF and WGIF are exploited to reduce noise from seven high-frequency subband images and one low-frequency sub-band image, respectively. Finally, a noise-free image is attained through inverse DWT and an exponential transform

Low-energy Pipelined Hardware Design for Approximate Medium Filter

Abstract: Image and video processing algorithms are currently crucial for many applications. Hardware implementation of these algorithms provides higher speed for large computation applications. Removing noise is often a typical pre-processing step to enhance the results of later analysis and processing. Median filter is a typical nonlinear filter that is very commonly used for impulse noise elimination in digital image processing. This article suggests a low-energy median filter hardware design for battery-based hardware applications. An approximate solution with high accuracy is investigated to speed up the filtering operation, reduce the area, and consume less power/energy. Pipelining and parallelism are used to optimize the speed and power of this technique. Non-pipelined, two different pipelined structures, and two parallel architectures versions are designed. The design versions are implemented first with a Virtex-5 LX110T FPGA and then using the UMC 130nm standard cell ASIC technology. The selection and the even-odd sorting-based median filters are also implemented for an equitable comparison with the standard median filtering techniques. The suggested non-pipelined median filter design enhances the throughput 35% more than the highest investigated state of the art. The pipelining enhances the throughput to more than twice its value. Additionally, the parallel architecture decreases the area and the consumed power by around 40%. The simulation results reveal that one of the suggested designs significantly decreases the area, with the same speed as the fastest design in the literature, without noticeably degrading the accuracy, and a significant decrease in energy consumption by about 60%.

Neural Network System for Recognizing Images Affected by Random-Valued Impulse Noise

Abstract: Images taken with different sensors and transmitted through different channels can be noisy. In such conditions, the image most often suffers from random-valued impulse noise. Denoising an image is an important part of image preprocessing before recognition by a neural network. The accuracy of image recognition by a neural network directly depends on the intensity of image noise. This paper presents a three-stage image cleaning and recognition system, which includes a developed detector of pulsed noisy pixels, a filter for cleaning found noisy pixels based on an adaptive median, and a neural network program for recognizing cleaned images. It was noted that at low noise intensities, cleaning is practically not required, but noise with an intensity of more than 10% can seriously damage the image and reduce recognition accuracy. As a training base for noise, cleaning, and recognition, the CIFAR10 digital image database was used, consisting of 60,000 images belonging to 10 classes. The results show that the proposed neural network recognition system for images affected by to random-valued impulse noise effectively finds and corrects damaged pixels. This helped to increase the accuracy of image recognition compared to existing methods for cleaning random-valued impulse noise.

High-density salt & pepper noise removal using machine learning

Abstract: This paper proposes an additional ‘Median Layer’ to the existing Convolutional Neural Network (CNN).The work primarily focuses on removing ‘High Density Impulse Noise’ present in an image. The proposed median layer performs median filtering on all feature channels. The developed network is an end-to-end network that does not require any non-trivial preprocessing tasks. Based on the experiments performed, results show that, by inserting a median layer to a CNN, the proposed architecture has given convincing results, when compared to the existing approaches.

Three Stage Weighted-Mean Filter Using Euclidean Distance

Abstract: This paper presents a three-stage weighted-mean (TSWM) filter to restore digital images corrupted by high noise density impulsive noise. The proposed filter evaluates the weighted mean of non-noisy pixels (NFP) in $$3\times 3$$ or $$5\times 5$$ grid where the respective weights taken are proportional to their euclidean distance from the center pixel. The first step functions as a pre-processing step where the noisy pixels with the highest information or least uncertainty are denoised first using the two adjacent pixels in a straight line. And in the third stage, edge cases such as corners and boundaries are handled. The performance comparison of this algorithm is done using a $$512\times 512$$ grey-scale Lena image. The experiment results show on an average increment of 2.54 dB and 0.15 dB of Peak signal to noise ratio for high (10% to 90%) and extremely high (90% to 98%) noise densities over popular denoising filters.

Research on medical image denoising based on wavelet pyramid fusion

Abstract: The noise in medical images will affect the subsequent image processing. In order to reduce the noise in the prostate capsule, we proposed a new noise reduction technology using principal component analysis, wavelet pyramid fusion and adaptive median filtering. Compared with other commonly used denoising algorithms, this method is simple and efficient, and can better maintain image details while reducing noise.

Accuracy improvement of edge image recognition AI in dark light environment by image enhancement

Abstract: In order to solve the problem of poor image recognition accuracy of edge devices in the dark environment, the author simulate the change of brightness and contrast in the dark environment by using matlab's internal function on the image through the point operation method in matlab, and add Salt & pepper and Gaussian noise to simulate the noise generated by the image in the dark environment. The modified images are imported into the image recognition system trained by migration learning to compare the changes in recognition accuracy and to investigate the main factors affecting the image recognition accuracy in low light conditions. Meanwhile, the main factors affecting the image recognition accuracy are improved by median filtering and Wiener filtering to find the image enhancement method that is most beneficial to improve the image recognition accuracy of edge devices in the dark light environment. The experimental results show that the main factor affecting the image recognition accuracy in the dark environment is the Salt & pepper noise, and the median filtering can remove the Salt & pepper noise well and improve the recognition accuracy of a single image up to 70%.

Salt-and-pepper Noise Reduction for Medical Images Based on Image Fusion.

Abstract: BACKGROUND During the collection process, the prostate capsule is prone to introduce salt and pepper noise due to gastrointestinal peristalsis, which will affect the precision of subsequent object detection. OBJECTIVE A cascade optimization scheme for image denoising based on image fusion was proposed to improve the peak signal-to-noise ratio(PSNR) and contour protection performance of heterogeneous medical images after image denoising. METHOD Anisotropic diffusion fusion (ADF) was used to decompose the images denoised by adaptive median filter, non-local adaptive median filter and artificial neural network to generate the base layer and detail layer, which were fused by weighted average and Karhunen-Loeve Transform respectively. Finally, the image was reconstructed by linear superposition. RESULTS Compared with the traditional denoising method, the image denoised by this method has a higher PSNR while maintaining the image edge contour. CONCLUSION Using the denoised dataset for object detection, the detection precision of the obtained model is higher.

Comparison of Edge Detection of Images with Various Types of Noise and Denoising Methods

Abstract: In this paper, we show comparison results of edge detection of images that have additive gaussian noise or salt and pepper noise by using various techniques of noise removal such as filtering, morphology and deep learning based ones. In particular, this present work provides comparison results of noise removal by using gaussian filter, open and close operations of morphology and auto-encoder model followed by carrying out edge detection. Robert cross, Sobel, Prewitt and Canny detectors are used for edge detection of the images with noise removal. Experimental results show that noise removal results are different with characteristics of noise and techniques applied for noise removal. In addition, deep learning based technique, auto-encoder does not always shows superior results of noise removal, particularly in the case of existence of salt-pepper noise. In the experiments, gaussian noise or salt-pepper noise is used and peak signal noise ratio (PSNR) is used for quantitative comparison and the results of edge detection is qualitatively compared from visual perspective.

Lung Carcinoma Detection from CT Images Using Image Segmentation

Abstract: Nowadays, researchers experience a severe challenge in detecting lung infection automatically. Lung carcinoma is the most common cancer in which aberrant cells develop and form a malignant tumor that causes death worldwide. During the primary stages of cancer detection and therapy, image processing technologies are often used to increase image quality. Because noise signals are mixed in with creative signals during the image capture process, the quality of the image can be distorted, resulting in poor performance. The pre-processing stage for lung cancer has become crucial, and image denoising is a crucial step that reduces noise. This research work focuses on improving the quality of lung images and diagnosing lung cancer by eliminating misdiagnosis. The images are taken from the Cancer Imaging Archive (CIA) dataset, and noise is removed by utilizing the median filter technique, which successfully eliminates noise from the image while also improving the image quality. The impacted region yields a variety of spectral characteristics. These are examined by using an enhanced watershed and then CNN is applied for the proper classification of the tumor as Malignant or Benign. .

Counting Bacterial Colony and Reducing noise on Low-Quality Image Using Modified Perona-Malik Diffusion Filter with Sobel Mask Fractional Order

Abstract: In the field of microbiology, the counting of bacterial colonies is fundamental and mandatory. This is done to estimate the number of bacterial cells in every 1 milliliter or gram of sample. The counting takes a long time and is tedious, so it requires an accurate and fast counting method. The image quality used is very low and contains noise. Therefore, a preprocessing method is needed to reduce the noise. The Perona-Malik filter method is known to be able to remove noise well. However, it is difficult to determine the appropriate gradient threshold parameter ( ) for each different image. To find the appropriate value of , the original Sobel Mask method and Sobel Mask Fractional-Order are used to estimate the value of . The experimental results show the results of noise reduction using PMD with a value of from the original Sobel Mask and Sobel Mask Fractional-Order. The results of the accuracy of determining the value of k with the Sobel Mask Fractional-Order (α=1.0) show higher results based on the F-Measure values for samples 1, 2, and 3 respectively 97%, 98%, and 90%.

New Real-Time High-Density Impulsive Noise Removal Method Applied to Medical Images

Abstract: This paper introduces a new method for real-time high-density impulsive noise elimination applied to medical images. A double process aimed at the enhancement of local data composed of Nested Filtering followed by a Morphological Operation (NFMO) is proposed. The major problem with heavily noisy images is the lack of color information around corrupted pixels. We show that the classic replacement techniques all come up against this problem, resulting in average restoration quality. We only focus on the corrupt pixel replacement phase. For the detection itself, we use the Modified Laplacian Vector Median Filter (MLVMF). To perform pixel replacement, two-window nested filtering is suggested. All noise pixels in the neighborhood scanned by the first window are investigated using the second window. This investigation phase increases the amount of useful information within the first window. The remaining useful information that the second window failed to produce in the case of a very strong connex noise concentration is then estimated using a morphological operation of dilatation. To validate the proposed method, NFMO is first evaluated on the standard image Lena with a range of 10% to 90% impulsive noise. Using the Peak Signal-to-Noise Ratio metric (PSNR), the image denoising quality obtained is compared to the performance of a wide variety of existing approaches. Several noisy medical images are subjected to a second test. In this test, the computation time and image-restoring quality of NFMO are assessed using the PSNR and the Normalized Color Difference (NCD) criteria. Finally, an optimized design for a field-programmable gate array (FPGA) is suggested to implement the proposed method for real-time processing. The proposed solution performs excellent quality restoration for images with high-density impulsive noise. When the proposed NFMO is used on the standard Lena image with 90% impulsive noise, the PSNR reaches 29.99 dB. Under the same noise conditions, NFMO completely restores medical images in an average time of 23 milliseconds with an average PSNR of 31.62 dB and an average NCD of 0.10.

Comparison of algorithms for the removal of impulsive noise from an image

Abstract: Image pre-processing is an important operation that is used to redefine an image to improve human visual perception and information extraction. To de-noise an image tainted with impulsive noise, several state-of-the-art methods have been presented. This work examines and compares several de-noising methods based on the median filter and its advanced non-linear techniques. The research also focuses on the recommended implementation methodologies for using deep learning to de-noise impulsive noise. The paper focuses on one approach's limitations and possible remedies, as well as other techniques that have been offered. The study also identifies several other difficulties that have yet to be resolved.

A New Hybrid Model for Segmentation of the Skin Lesion Based on Residual Attention U-Net

Abstract: Skin segmentation participates significantly in various biomedical applications, such as skin cancer identification and skin lesion detection. This paper presents a novel framework for segmenting the skin. The framework contains two main stages: The first stage is for removing different types of noises from the dermoscopic images, such as hair, speckle, and impulse noise, and the second stage is for segmentation of the dermoscopic images using an attention residual U-shaped Network (U-Net). The framework uses variational Autoencoders (VAEs) for removing the hair noises, the Generative Adversarial Denoising Network (DGAN-Net), the Denoising U-shaped U-Net (D-U-NET), and Batch Renormalization U-Net (Br-U-NET) for removing the speckle noise, and the Laplacian Vector Median Filter (MLVMF) for removing the impulse noise. In the second main stage, the residual attention u-net was used for segmentation. The framework achieves (35.11, 31.26, 27.01, and 26.16), (36.34, 33.23, 31.32, and 28.65), and (36.33, 32.21, 28.54, and 27.11) for removing hair, speckle, and impulse noise, respectively, based on Peak Signal Noise Ratio (PSNR) at the level of (0.1, 0.25, 0.5, and 0.75) of noise. The framework also achieves an accuracy of nearly 94.26 in the dice score in the process of segmentation before removing noise and 95.22 after removing different types of noise. The experiments have shown the efficiency of the used model in removing noise according to the structural similarity index measure (SSIM) and PSNR and in the segmentation process as well.

Hardware oriented bilateral filter algorithm and architecture survey

Abstract: Images and videos are often interfered with and affected by various noises. To filter out noise, researchers have proposed many filtering algorithms. bilateral filtering is a nonlinear filtering technique that can retain detailed information such as image edges well while denoising. However, it is a difficult task to implement fast and high-precision bilateral filtering algorithms in hardware platforms. In this paper, for the optimization of bilateral filtering algorithms, we analyze the gradient bilateral filtering and the piecewise approximate bilateral filtering algorithm that can be implemented in hardware, respectively. Then, we analyze and compare the implementation and optimization of bilateral filtering algorithms on FPGA and VLSI hardware architectures, and finally, make suggestions for choosing the appropriate bilateral filtering algorithms and hardware architectures for different purposes.

Character Recognition Technique Implementation for Complicated Deteriorated Scene

Abstract: There has been an increase in interest in digitizing and preserving old books and papers in the last few years. The quick advancement of data innovation and the Internet’s quick spread has also contributed to the enormous volume of image and video data. The texts that are included in the image and video assist us in analyzing them and are also utilized for indexing, archiving, and retrieval. Different noises, such as Gaussian noise, salt and pepper noise, speckle noise, etc., can readily damage an image. Several image filtering algorithms, including the Gaussian filter, mean filter, median filter, etc., are employed to eliminate these various noises from images. This article analyses the impact of several pre-processing approaches, such as thresholding, morphology, and blurring procedures, to maximise text extraction strategies. The experiment’s findings demonstrate that pre-processing approaches unquestionably improve the document’s structural and visual quality.

Optimized Algorithms and Hardware Implementation of Median Filter for Image Processing

Abstract: Abstract Image processing algorithms are essential for clarifying the image and improving the ability to recognize distinct characteristics of the image. The field of digital image processing is widespread in several research and technology applications. In many of these applications, the existence of impulsive noise in the obtained images is one of the most frequent problems. The median filter is a strong method to remove the impulsive noise; it effectively eliminates salt and pepper noise from the image. The main target of this paper is to investigate efficient median filter units to be connected to a general-purpose processor (GPP) for FPGA-based embedded systems. The paper exposes three novel techniques, two of them specially for median filtering techniques and the third one is used to get the maximum number of any 9 elements array. The proposed algorithms are inspired by the Median Of Median (MOM) algorithm. The first two techniques are tested for filtering $$3 \times 3$$ 3 × 3 image windows and optimized for producing the expected result in high accuracy, short time, and reduced number of comparisons. The last technique is tested for a 9 elements array for extracting the maximum number in same high efficiency manner. Furthermore, the three proposed techniques are implemented leveraging the advantage of the parallel processing and the FPGA flexible resources to satisfy the real-time processing constraints. A comparison between the first two proposed filtering units and their counterparts in the literature is included. The comparison reveals the superiority of the first technique in terms of accuracy with fewer comparators than previously published techniques. Besides, the paper illustrates how the concept beyond the proposed techniques can be used to perform the maximum pooling for convolution neural networks.

Removal of Salt and Pepper Noise Using Hybrid Adaptive Switching Median Filter with Ant Colony Optimization Technique in Nano Electronic Applications

Abstract: Background: Restoration of noisy images from the salt and pepper noise is an interesting area in the field of image processing. The restoration process can be done using various filtering algorithms. The restoration process should not affect the pixels of the original image. The problem of the existing work persists as the increase in the error rate while the dimensions as well as the image format changes. The proposed work consists of Hybrid Adaptive Switching median filtering (HASMF). The hybrid technique corrupted images’ high-density salt and pepper noise removal using Ant colony Optimization technique. This hybrid technique would remove the high-density salt and pepper noise from the corrupted images. The noisy pixel value from the corrupted images is identified and selected using the Ant Colony Optimization technique (ACO). The identified corrupted value can be replaced using the Adaptive Switching Median Filter. The switching process is carried out using the pixel by pixel with the normalized median values. The noisy pixels are identified and selected using Ant colony Optimization. The optimized values are subjected to the filtering process. The proposed method decreases the salt and pepper noise within the original image. The hybrid design approach was used in the proposed study, which used 45 nm technology combined with a Verilog-A model-based circuit that was implemented using Spintronic. It was discovered that the suggested changed task had less latency, used less space, and dissipated less power than the original. Furthermore, it was discovered that designed memory arrays were both energy and space-efficient. It does not affect the normal pixels within the original image. The comparison process has been made with the various existing algorithms such as Median Filter (MF), Modified Decision Based Unsymmetrical Trimmed Median Filter (MDBUTMF). The proposed method has overcome the various performance metrics such as Peak Signal Noise Ratio (PSNR), Mean Square Error (MSE), and Structural Similarity Index (SSIM). The results obtained have shown the significant results in terms of object measures as well as visual perception of the denoised image.

DTMF: Decision Based Trimmed Multimode Approach Filter for Denoising MRI Images

Abstract: The brain MRI image denoising is a challenging and attracting field for young researchers because it enhances the quality of medical images. Salt and pepper noise is the most dangerous noise which reduces the accuracy of brain diagnosis, and it damages the brain medical images severely, that leads to neurologists to fix incorrect treatments or surgery. The pitfalls raised in the existing denoising methods are less Peak signal to noise ratio, high time consumption and incapable for enormous level of noise range. Hence, this research proposes a novel denoising filter which is entitled as ‘Decision based Trimmed Multimode approach oriented Filter (DTMF)’ for salt and pepper noise removal. Herein, the noise removal section is branched into six steps which efficiently reduce noises based on multimode of majority strength. The main concepts used in this research are viz. decision based approach, trimming process, majority of intensity, median, mean, dynamic windows and Square shaped Exemplar Modeled Patch Mechanism (SEMPM). The essential contributions of this approach are i) designing rule set for majority strength structured multimode denoising, ii) computation of majority property oriented parameters like majority-instance, majority strength and majority value, iii) novel SEMPM mechanism to predict noise-free data. The novel SEMPM mechanism grants a solution for the prediction of noise-free pixel in account of the noisy pixels whose surrounding window is completely packed by noisy data. The proposed decision-based approach removes the salt and pepper noise with high peak signal to noise ratio even for huge noise range, with reasonable time consumption.

A Noise reduction in the medical images using hybrid combination of filters with nature-inspired Black Widow Optimization Algorithm

Abstract: This paper proposes an image filtering method to remove the noises in medical images in a controlled manner. To achieve this goal, the optimal parameters of the conventional filters are determined using the nature-inspired black widow (BWO) optimization algorithm to remove the noise efficiently. The BWO algorithm is chosen over other optimization algorithms because it quickly explores the optimal parameter values due to its procreate and cannibalism steps. The procreate step explores new solutions, whereas the cannibalism steps remove the inappropriate solutions while exploring the optimal solution. In the proposed method, speckle and sharpening filters are considered. In the proposed method, initially, medical images are read. After that, they are enhanced using the power law method because images are either low or high contrast. In the power law method, the gamma value plays an important role. Therefore, the optimal gamma value is determined using the BWO algorithm as done for the filter values. After that, noise addition is performed on them and removed them using the speckle filter. Further, the edges of the image are filtered using the sharpening filter. The proposed method is validated on the standard dataset images downloaded from Kaggle. It is found that the proposed method enhances the image and removes the noise in a controlled manner. Besides that, it achieves better Mean Square Error (MSE) and Peak Signal to Noise Ratio (PSNR) in the output.

Image denoising using adaptive bi-dimensional stochastic resonance system

Abstract: Abstract Using stochastic resonance (SR) mechanism, the output signal can be enhanced by adding noise to the nonlinear system. Therefore, an image denoising algorithm based on adaptive bi-dimensional stochastic resonance (ABSR) is proposed in this paper. Firstly, the image is sampled as a bi-dimensional signal, and an adaptive bi-dimensional dynamic nonlinear system model is constructed. The peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) of the output image are used as the double evaluation model of the adaptive system, and the optimal parameters of the model are automatically obtained by adjusting the parameters of the dynamic nonlinear system using the reverse positioning method. Compared with the traditional mean filter, median filter and one-dimensional stochastic resonance, the image restoration effect of dynamic adaptive bi-dimensional stochastic resonance is more closer to the original image, and the histogram, PSNR and SSIM of the output image are also significantly better than the other three methods. The results show that dynamic adaptive bi-dimensional stochastic resonance has better denoising effect and better robustness to the change of noise intensity in image processing.

Ultrasound Image Noise Reduction and Enhancement Model based on Yellow Saddle Goatfish Optimization Algorithm

Abstract: In the modern-day diagnostics, ultrasound play an important role in different applications such as vascular, gynecological, cardiac, and obstetrical for diagnosis the various diseases. The main benefit of the ultrasound is that it is non-invasive method and inexpensive. However, in the real-scenario, ultrasound images contain speckle noise which negatively impact the image quality in terms of edges, texture information, and boundaries. In order to eliminate noise, various filters are deployed by researchers in the literature. The limitations of their method are that a fixed level of noise is removed using conventional filters in which parameter values of the filters are fixed. However, in the real-time situation, the noise is random and adaptive filters are required which eliminate any level of noise. To achieve this goal, this paper proposes an adaptive filtering model for eliminate speckle noise based on yellow saddle goatfish optimization (YSGO) algorithm. The YSGO algorithm is based on the hunting behaviour of the fishes. In the proposed model, bilateral filter and speckle-reducing anisotropic diffusion filtering methods and enhancement power law method are taken under consideration. Further, the parameter values of the filtering method and enhancement methods are determined using the nature-inspired YSGO algorithm. The YSGO algorithm minimize the noise and enhances the image brightness and edge information based on the objective function. In our model, mean square error (MSE) and entropy is taken as the objective function. Further, the proposed model is applied on the standard ultrasound images. The visual analysis of the images is done based on the subjective analysis whereas various performance metrics are measured for measure the image quality in the objective analysis. The results reveals that the proposed model outperforms over the existing models in terms of PSNR.

Detection of Colon Cancer Using Image Processing

Abstract: Image processing is one of the most widely used techniques in the medical field for the early diagnosis of diseases. The most common application of image processing in medicine is the detection of cancer. Colon cancer is one of the most typical cancers worldwide. Early disease detection using image processing, which is more accurate. Early cancer detection can save lives because it enables more efficient medical care for those who are affected. The suggested approach involves gathering a set of histopathological images of the colon and using extraction, filtering, and classification to distinguish between normal and abnormal images. A grayscale version of the photo is created. Filters are used to reduce image noise. Procedure for feature extraction Using K-Mean, FCM, and FFT, the values are recorded. For storage of values, it is normalized. For each calculation, the mean, variance, standard deviation, kurtosis, and deviation are determined. Through the use of a classifier, nonlinear data is transformed into linear data.

VLSI Implementation for Noise Suppression Using Parallel Median Filtering Technique

Abstract: This paper proposes a computationally efficient median filtering approach to reduce impulsive noise in a 2D signal. The median filter is a nonlinear method for reducing noise from an image. In this paper, a preprocessing step of identifying the corrupted pixels is included. This step ensures that the uncorrupted pixel preserves its value unlike the traditional median filter. A comparative analysis is done between the median filter and modified median filter. And, the results show that the modified median filter has improved RMSE value. Further, in the Verilog implementation, parallelism is incorporated to improve latency. The proposed work is implemented using ModelSim and Xilinx.