Showing papers on "Median filter published in 2022"

Using Drones for Thermal Imaging Photography and Building 3D Images to Analyze the Defects of Solar Modules

Abstract: In this research, drones were used to capture thermal images and detect different types of failure of solar modules, and MATLAB® image analysis was also conducted to evaluate the health of the solar modules. The processes included image acquisition and transmission by drone, grayscale conversion, filtering, 3D image construction, and analysis. The analyzed targets were the solar modules installed on buildings. The results showed that the employment of drones to monitor solar module farms could significantly improve inspection efficiency. Moreover, by combining the mean and median filtering techniques, an innovative box filtering method was successfully created. Additionally, this study compared the differences between the mean, median, and box filtering techniques, and proved that the 3D image improved by box filtering is a more convenient and accurate way to check the health of solar modules than the mean and median filtering methods. In addition, this new method can simplify the maintenance process, as it helps maintenance personnel to determine whether to replace the solar modules on site, achieving the goal of power generation efficiency enhancement. It is worth noting that 3D image recognition technology can enhance the clarity of thermal images, thereby providing maintenance personnel with better defect diagnosis capability. It is also able to provide the temperature value of the defect zone, and to indicate the scale of defects through the cumulative temperature chart, so the 3D image is qualified as a quantitative and qualitative indicator. The analysis of the transmitted image is innovative that it not only can locate the defect area of the module, but also can display the temperature of the module, providing more information for maintenance personnel.

A Median Filtering Forensics CNN Approach Based on Local Binary Pattern

Abstract: Median filtering forensics has gradually become a research hotspot because of the wide application of the traditional median filtering (MF) in image tampering and anti-forensics. The difficulty of traditional median filtering based on machine learning forensics is feature extraction which is a manual selection process, and the Convolutional Neural Network (CNN) cannot also well perceive the traces left by median filtering straightly. By taking the Local Binary Pattern (LBP) data of an image as a presentation of the streaking artifact that is a very strong indication for median filtering, a median filtering forensics CNN approach with LBP is proposed, which can automatically learn and get median filtering features from image. Different from traditional CNN, an LBP perception layer is added before the following deep learning layers. Then, five convolution layers constitute the feature extraction group. Finally, the classifier group is composed of three full connected layers to decide whether the image is median filtered or not. The proposed approach is tested on several experiments and the experimental results demonstrate its effectiveness.

Performance analysis of peak signal-to-noise ratio and multipath source routing using different denoising method

Abstract: The problem of denoising iris pictures for iris identification systems will be discussed, as well as a novel solution based on wavelet and median filters. Different salt and pepper extraction algorithms, as well as Gaussian and speckle noises, were used. Because diverse sounds decrease picture quality during image collection, noise reduction is even more important. To reduce sounds like salt and pepper, Gaussian, and speckle, filtering (median, wiener, bilateral, and Gaussian) and wavelet transform are utilised. Provide better results as compared to other ways. A study of several efficiency indicators such as peak signal-to-noise ratio (PSNR) and mean squared error will be used to demonstrate the superiority of the proposed technique (MSE).

Extraction of significant features using GLDM for Covid-19 prediction

Abstract: Although Covid-19 caused by the SARS-COV-2 virus, is a deadliest disease, many people experienced mild symptoms and were recovered soon. In this paper, coronavirus can be easily detected using CT scan images of affected patients. Initially, images are pre-processed by filters like Median filter and Noise adaptive fuzzy switching median filter, and then the quality measurements like MSE, and PSNR are calculated. After preprocessing, segmentation is done by K-means and Robust self sparse fuzzy clustering algorithm, and then the parameters like LMSE and NAE are calculated. Finally, to get optimum results, feature extraction using GLDM is performed which helps in identifying whether it's a normal lung disease like pneumonia or the patient is affected by covid.

An Adaptive Bitonic filtering based Edge fusion algorithm for Gaussian Denoising

Abstract: It is broadly recognized that conserving the essential geometrical features of an image is crucial while denoising it. To accomplish this aim, various denoising techniques have been represented in the literature. The technique based on dual way edge fusion can efficiently solve the problem of denoising. In this paper, an efficient denoising scheme using an innovative method of calculating the image base and details is being proposed. The noisy image is thresholded to remove extra noise by the bitonic filter. Details of the discontinuities is extracted by subtracting the recovered image from the noisy image. Subsequently, details features are subtracted from the noisy image to extract the base information. After that, image features and noise are simultaneously filtered by rolling guidance filter to remove the remaining noise from the features and the significant edge information from the filtered noise. The two images are fused with maximum coefficient value to enhance the information content and visual quality of denoised image. The proposed Dual Way Residue Noise Thresholding (DWEFD) is a combination of various spatial and transform domain commutations performed parallelly. Extensive experimental results and investigations reveal that the proposed methodology is able to recover feature details of an image thereby reducing information loss along with efficient noise removal.

INCNet: Brain Tumor Detection using Inception and Optimization Techniques

Abstract: Brain tumors are one of the most serious disorders that may affect humans. The early detection of a brain tumor is critical for its treatment. The precise locations of the tumor and its spread area have been discovered. A tumor is detected by observing that it has a greater intensity in its location. For the sake of this implementation, an MRI scan image is regarded to be the system's input.Most of the existing methods have suffered from the inability to detect tumors of small size and their inability to correctly identify abnormal tissue growth in the brain. To address these limitations, a method has been proposed that includes image pre-processing, post-processing, and inception-optimization are the three primary processes in the implementation of tumor detection. Image enhancement techniques such as noise reduction, high-pass filtering, median filtering, and de-blurring are employed at the preprocessing stage. Post-processing includes the use of operations such as thresholding, segmentation using the watershed technique, and morphological operations. The implemented system is trained at various angles using the inception and optimization approach, and the precise location of the tumor is determined. The performance of the proposed method is tested in terms of precision, recall, tumor size, and accuracy. The proposed method is superior to existing approaches. The entire simulation was conducted using MATLAB R2021a.

Seismic random noise attenuation based on adaptive nonlocal median filter

Abstract: The accurate image of underground medium is determined by the quality of the seismic data, which can be improved by random noise attenuation and structural continuity enhancement. We proposed an adaptive nonlocal median filter that can protect geological structure while attenuating random noise. We combine the nonlocal idea with the weighted median filter and design the appropriate weights of the nonlocal median filter based on seismic data characteristics. The local structure is represented by the neighborhood around the center point. The directional difference of spatial vectors in the neighborhood is considered when computing the similarity. According to the local dip attribute of seismic data, the anisotropic Gaussian window is adaptively adjusted to increase the constraint along the structural direction. The proposed method can search more precisely for points with similar local structure to the filtered points and effectively attenuate seismic random noise. The continuity of events is enhanced while the goal of protecting fault information is achieved. The experimental results of the theoretical model and field data show that the adaptive nonlocal median filter can strike a balance between preserving structure information and attenuating seismic random noise.

Penerapan Metode Harmonic Mean Filter Dalam Mereduksi Gaussian Noise Pada Citra Digital

Abstract: Abstrak - Pada penelitian ini, noise dibangkitkan dengan menggunakan tools matlab dan dilakukan analisa. oleh karena itu, untuk mereduksi derau (noise) perlu dilakukan suatu proses perbaikan kualitas citra terhadap citra yang mengalami derau atau noise tersebut sehingga citra dapat dengan mudah diinterpretasikan baik oleh manusia maupun mesin. Penelitian ini menggunakan metode harmonic mean filter. Harmonic mean filter merupakan salah satu metode untuk menghitung rata-rata dari citra yang rusak g(s,t) pada sebuah blok area citra yang didefinisikan oleh 𝑆𝑥𝑦. Nilai dari citra f(x,y) yang diperbaiki pada tiap titik (x,y) hanya dihitung dengan menggunakan piksel dalam daerah yang didefinisikan oleh 𝑆𝑥𝑦. Tipe noise yang digunakan dalam penelitian ini adalah gaussian noise. Gaussian noise memiliki intensitas yang sesuai dengan distribusi normal yang memiliki rerata (mean) dan varian tertentu. Penelitian ini menguji seberapa efektif metode harmonic mean filter dalam mereduksi gaussian noise menggunakan citra rontgen. penelitian ini menggunakan matriks 5 x 5 dalam sampel perhitungannya. Hasilnya yaitu berupa angka dari hasil metode perhitungan harmonic mean filter dalam bentuk tabel.Kata kunci: Harmonic Mean Filter, Citra Digital, Gaussian Noise, Reduksi, Rontgen. Abstract - In this research, noise is generated using MATLAB tools and analyzed. Therefore, to reduce noise, it is necessary to carry out an image quality improvement process for images that experience noise or noise so that the image can be easily interpreted by both humans and machines. This study uses the Harmonic Mean Filter method. Harmonic Mean Filter is one method to calculate the average of the damaged image g(s,t) in a block image area defined by . The value of the fixed f(x,y) image at each point (x,y) is calculated using only pixels in the area defined by . The type of noise used in this study is Gaussian Noise. Gaussian Noise has an intensity that corresponds to a normal distribution which has a certain mean (mean) and variance. This study examines how effective the Harmonic Mean Filter method is in reducing Gaussian Noise using X-ray images. This study uses a 5 x 5 matrix in the calculation sample. The result is a number from the results of the Harmonic mean filter calculation method in tabular form.Keywords: Harmonic Mean Filter, Digital Image, Gaussian Noise, Reduction, X-Ray.

A Preprocessing Techniques for Seismocardiogram Signals in Removing Artifacts

Abstract: Seismocardiogram (SCG) is a non-invasive technique for cardiomechanical assessment by analyzing local vibrations on chest surface. SCG signals have numerous clinical and health awareness applications. The SCG signals utilized in this work are obtained from public domain database which are acquired by standard signal acquisition protocol. The acquired SCG signal includes artifacts such as base line wander, random noise, and predictive power line interference. The artifact-free signal helps significantly in analyzing them either in time scale or in frequency domain. In this work, an attempt has been made to remove motion-induced noises and the low frequency noise including respiratory sounds. The artifact removal methods include moving average and median filter and finite impulse response-based smoothing filter named Savitzky Golay filter. The performance of all these methods is compared using the denoising metrics such as mean square error, mean absolute error, signal to noise ratio, peak signal to noise ratio. Results demonstrate that median filter along with Butterworth filter performs better in removing the motion-induced artifact and low frequency and respiratory sounds. The methodology used in this work is helpful further for further annotation of the signals.

Application of Zero-phase Digital Filtering for Effective Denoising of FAIMS Signal

Abstract: RATIONALE FAIMS has a great potential to become a portable technology for rapid detection of chemical and biological agents. However the ion current signals, measured at the exit of the planar FAIMS directly, may contain different types of noises. The peak information in the FAIMS spectrum, such as the compensation voltage value at the maximum peak intensity (CVP ) and the peak width at half maximum (Wh ), could not be accurately determined under the weak signal condition which significantly limits the achievable instrument sensitivity, and there are no existing solutions to the problem. METHODS This study analyzed the noise type of FAIMS signal in detail, and three different signal processing algorithms, including median filtering (MF), discrete wavelet transform (DWT) and zero-phase digital filtering (ZDF), were evaluated for their performance in denoising the FAIMS signal. RESULTS The results show that the standard deviation of CVp obtained from the signal denoised by using ZDF algorithm is at least 31.82% smaller as compared to using MF and DWT algorithms. The standard deviation of Wh is at least 45.45% smaller by using ZDF algorithm. Moreover, only ZDF algorithm can keep the percentage error for the CV value of the denoised signal to be within 0.50±0.47% of the true CV value, implying the effectiveness of ZDF algorithm in denoising while retaining the integrity of the signal. CONCLUSIONS The ZDF algorithm greatly reduces the analyte peak extraction error and improves the limit of detection in FAIMS measurements.

License Plate Localization in Complex Environments Based on Improved GrabCut Algorithm

Abstract: Aiming at the problem that the existed license plate detection method lacking of accuracy and speed, an improved lightweight detection algorithm for license plate detection in natural scenarios was proposed. First, the traditional GrabCut algorithm needs to interactively provide a candidate frame in order to perform the target detection work. We replace the candidate frame by introducing the Aspect ratio of the license plate as the foreground extraction feature to automate the detection of the license plate by GrabCut algorithm. Then, in order to improve the detection precision of traditional target detection algorithms, we introduced the Wiener filter, which is widely used in the field of digital signal processing, and Combine with Bernsen algorithm to complete image noise reduction. Finally, the algorithm was tested with the CCPD dataset, which contains many vehicle images from different complex natural scenes, especially low-resolution images. The experimental results shows that improved GrabCut algorithm achieves an average accuracy of 99.34% for license plate localization and a detection speed of 0.29s/frame, which has better accuracy and real-time performance compared with traditional GrabCut and other license plate localization algorithms.

Salt and Pepper Noise Removal Method Based on the Edge-Adaptive Total Variation Model

Abstract: The traditional median filter can handle the image salt and pepper noise better. However, when the noise intensity is large, it is often necessary to enlarge the filter window to ensure the denoising effect, but the enlarged window may also cause excessive smoothing of the image, loss of texture details, and blurred edges. In view of the strong edge preservation characteristics of variational model denoising, we propose a salt and pepper noise removal method based on the edge-adaptive total variational model. Firstly, the image is segmented into edge regions and non-edge regions by edge detection operators. Secondly, the salt and pepper noise of the image is processed using the median filter and adaptive total variation model, respectively. Lastly, the non-edge regions processed by the median filter and the edge regions processed by the adaptive total variation model are extracted for splicing. The experimental results show that the method cannot only effectively remove salt and pepper noise, but also effectively protect the main edge details of the image.

A New Adaptive Filter to Remove Impulsive Noise in Color Images

Abstract: In this paper, a new adaptive vector directional distance filter (AVDDF) is proposed. The AVDDF filter is developed to remove the ‘salt and papper’ impulsive noise in color images. This algorithm consists in first step to detect pixels that are likely to have been contaminated with noise by using a threshold value. In second step, after recognizing the corrupted pixels, the smallest angular‐magnitude distance is used to replace the noisy pixels. The proposed filter is tested with several standard color images which are contaminated with various levels of ‘salt and pepper’ impulsive noise (3%, 5%, 10%, 20% and 30%). The performance of the proposed filter is measured with peak signal‐to‐noise ratio (PSNR) and normalized color difference (NCD). The simulation results show that the proposed filter permits to ameliorate the performance of the classic vector directional distance filter (VDDF). Indeed, it provides an improvement by an average of 5% and 26% in the PSNR and NCD relative to the VDDF filter, respectively, with a small increase in the execution time by 6%. Besides, the AVDDF filter allows also an enhancement by an average of 2% and 14% in the PSNR and NCD relative to adaptive vector median filter (AVMF), respectively. © 2022 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC.

Design and Simulation of a Median Filter for a CubeSat Image Processing Application Using an FPGA Architecture

Abstract: CubeSats are small satellites that can perform space missions with the advantage of low cost and short development time. Earth observation is a well-known satellite use case that has found its place in the CubeSat community. To improve the quality and the number of images that can be received from the satellite, image processing techniques can be performed. Satellite images can be disturbed, and the median filter is a pre-processing technique usually used to remove impulse noise. The aim is to develop an architecture for CubeSat onboard image processing, starting with the design of a median filter. This paper presents the design and the simulation process of a 3x3 median filter based on the Spartan 6 FPGA architecture using software components. Simulation results are generated using a test bench algorithm and a visual comparison of both the input and output images is performed.

Selective Mean Filtering for Reducing Impulse Noise in Digital Color Images

Abstract: The interest of this paper is in reduction of impulse noise in digital color images. The two main methods used for noise reduction in images are the mean and median filters. These techniques operate by replacing the test pixel in a chosen window by a new filtered pixel value. The window is made to iteratively slide across the entire image to reconstruct a new noise reduced image. The mean filters suffer from the effect of smoothing out color contrast and edges due to leveraging the unrepresentative pixels in the filtering process. The vector median filter and its variants overcome this problem by considering only the most representative pixel in the chosen window. The most representative pixel, i.e. the pixel that is of highest conformity to take the place of the test pixel, is determined by minimizing the aggregate distance from one pixel to every other pixel in the window. The problem in these median filtering approaches is that only one pixel is treated as representative of all the pixels in the chosen window. This conjecture could lead to information loss due to marginalizing other pixels that also are representative of the center pixel. In this paper, we propose a selective mean filtering process to overcome the said problem. The key idea here is to determine the most representative pixels in the window using the method of aggregate distances and then compute the mean of these pixels. This approach will perform better than the vector median filters as now a set of representative pixels are leveraged into the filtering process. Simulation results show that the proposed method performs better than the conventional vector median filtering methods in terms of noise reduction and structural similarity and thus validates the proposed approach. Moreover, the method is tested on real MRI scan images in successfully reducing impulse noise for improved medical diagnosis.

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

Abstract: 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.