Noise Reduction Filter Research Articles

Effective Noise Reduction Filters for Precise Temperature Measurement Using Brillouin Distributed Optical Fiber Sensors

This study explores the effectiveness of various noise reduction filters in accurately measuring temperature distributions over a 38.2 km single-mode fiber (SMF). Brillouin optical time-domain analysis (BOTDA) sensor is utilized to gather Brillouin gain spectra (BGSs), which are denoised using bilateral filter (BF), guided filter (GF), adaptive Wiener filter (AWF), non-local means filter (NLMF), average filter (AF) and disc filter (DF). The temperature distributions over the SMF are then determined by applying least-squares curve fitting (LSCF). The study assesses the efficacy of noise reduction filters considering signal-to-noise ratio (SNR), uncertainty in temperature measurement (UTM), experimental spatial resolution (ESR) and signal processing speed (SPS). Among six different filters, NLMF outperforms other filters which can provide SNR improvement of 10.22 dB for which the UTM can be improved by 58.93% without deteriorating the ESR of the sensor. The noise reduction using such filter can also provide 6.2% faster SPS. Therefore, NLMF can be considered as an effective noise reduction filter for the precise temperature measurement using BOTDA sensors. DUJASE Vol. 8 (1) 42-50, 2023 (January)

Exploring the Impact of Noise and Image Quality on Deep Learning Performance in DXA Images.

Segmentation of the femur in Dual-Energy X-ray (DXA) images poses challenges due to reduced contrast, noise, bone shape variations, and inconsistent X-ray beam penetration. In this study, we investigate the relationship between noise and certain deep learning (DL) techniques for semantic segmentation of the femur to enhance segmentation and bone mineral density (BMD) accuracy by incorporating noise reduction methods into DL models. Convolutional neural network (CNN)-based models were employed to segment femurs in DXA images and evaluate the effects of noise reduction filters on segmentation accuracy and their effect on BMD calculation. Various noise reduction techniques were integrated into DL-based models to enhance image quality before training. We assessed the performance of the fully convolutional neural network (FCNN) in comparison to noise reduction algorithms and manual segmentation methods. Our study demonstrated that the FCNN outperformed noise reduction algorithms in enhancing segmentation accuracy and enabling precise calculation of BMD. The FCNN-based segmentation approach achieved a segmentation accuracy of 98.84% and a correlation coefficient of 0.9928 for BMD measurements, indicating its effectiveness in the clinical diagnosis of osteoporosis. In conclusion, integrating noise reduction techniques into DL-based models significantly improves femur segmentation accuracy in DXA images. The FCNN model, in particular, shows promising results in enhancing BMD calculation and clinical diagnosis of osteoporosis. These findings highlight the potential of DL techniques in addressing segmentation challenges and improving diagnostic accuracy in medical imaging.

A New Data Processing Approach for the SHPB Test Based on PSO-TWER

This study addresses the challenge of accurately determining the arrival time of stress wave signals in SHPB test data processing. To eliminate human error, we introduce the time-window energy ratio method and evaluate six filters for noise reduction using box fractal dimensions. A mathematical model is established to optimize the stress equilibrium and impact process, which is solved using particle swarm optimization, resulting in the PSO-TWER method. We explore the impact of inertia weight and calculation methods on optimization outcomes, defining a stress equilibrium evaluation index. The results indicate that time-window length significantly affects arrival-time outputs, and the dynamic inertia weight factor enhances optimization convergence. The method accurately determines arrival times and effectively screens test data, providing a robust approach for SHPB test data processing.

Sound-field image denoising using deep neural network considering physical characteristics of sound

This paper presents a method of denoising utilizing deep neural networks (DNN) in sound field images measured by optical methods. Optical sound measurement has attracted much attention because of its capability to visualize sound field with high spatial resolution from a distance, which is difficult with conventional microphones. However, optical methods have the issue of the measured sound-field images often being heavily distorted by noise. This is caused by the weak phase of light changed by sound. Conventionally, noise-reduction filters considering the physical properties of sound have been known to be used for sound-field denoising. In this background, we propose a DNN-based sound-field denoising method that takes into account the physical properties of sound. In the DNN, a network architecture originally used for denoising natural images is employed, and we integrate loss functions based on the Helmholtz equation. Additionally, a 2D sound field dataset obtained from numerical acoustic simulation with random parameters is used during the training. Numerical and experimental data comparison experiments showed that the proposed DNN-based sound-field denoising outperformed the previous non-DNN methods.

Impact of enhancement filters of a CMOS system on halo artifact expression at the bone-to-implant interface.

To assess the impact of enhancement filters on the formation of halo artifacts in radiographs of dental implants obtained with a complementary metal oxide semiconductor (CMOS) system. Digital radiographs of dental implants placed in dry human mandibles were processed with the Noise Reduction smoothing filter, as well as the Sharpen 1, Sharpen 4, and Sharpen UM high-pass filters available in the CLINIVIEW™ software (Instrumentarium Dental, Tuusula, Finland). Subjective analysis involved evaluating the left, right, and apical surfaces of each implant for the presence of much, few, or no halo. The objective analysis involved measurement of the halo area using the Trainable Weka Segmentation plugin (ImageJ, National Institutes of Health, Bethesda, MD, USA). Data were analyzed using Friedman's test (subjective analysis) and ANOVA (objective analysis) (α = 5%). In the subjective evaluation, the Sharpen 4 filter produced more radiographs with much halo present, and in the objective evaluation, a bigger halo area when compared to the original images and the Noise Reduction filter for all surfaces (p < 0.05). When evaluating dental implants, priority should be given to original images and those enhanced with smoothing filters since they exhibit fewer halo artifacts. Post-processing tools, such as enhancement filters, may improve the image quality and assist some diagnostic tasks. However, little is known regarding the impact of enhancement filters in halo formation on CMOS systems, which have been increasingly used in dental offices.

Design of Electronic Filter for Noise and Vibration Reduction in Brushed DC Motor

With the recent conversion of internal combustion engines to electric vehicles, new noise issues have arisen, and among them, the noise generated by internal vehicle auxiliary systems is being considered. This study introduces an electronic filter designed with a motor model featuring vibration components, aiming to minimize the noise and vibrations generated by a Brushed DC (BDC) motor commonly employed in vehicle internal systems. It introduces a method to identify the connectors and internal parameters used in the motor for the matching of the model and experimental motor, and to measure and estimate these parameters. The model is separated and executed to ensure convergence, and it is validated by comparing the analysis results with the measured values. A filter is designed using the model to reduce current oscillations in the motor, confirming a subsequent reduction in noise and vibration. This research suggests the potential to attenuate noise and vibration in already produced motors by attaching only a filter without modifying the internal motor structure. Moreover, it is anticipated that a filter can be designed to predict and mitigate the noise and vibration components of the motor based on changes in load.

Adaptive Gaussian Wiener Filter for CT-Scan Images with Gaussian Noise Variance

Medical imaging plays an important role in modern healthcare, with Computed Tomography (CT) being essential for high-resolution cross-sectional imaging. However, Gaussian noise often occurs within the CT scan images and makes it difficult for image interpretation and reduces the diagnostic accuracy, creating a significant obstacle to fully utilizing CT scanning technology. Existing denoising techniques have a hard time balance between noise reduction and preserving the important image details, failing to enable the optimal diagnostic precision. This study introduces Adaptive Gaussian Wiener Filter (AGWF), a novel filter aims to denoise CT scan images that have been corrupted with various Gaussian noise variance without compromising the image details. The AGWF combines the Gaussian filter for initial noise reduction, followed by the implementation of Wiener filter, which can adaptively estimate noise variance and signal power in localized regions. This approach not only outperforms other existing techniques but also showcases a remarkable balance between noise reduction and image detail preservation. The experiment evaluates 300 images from the dataset and each image is corrupted with Gaussian noise variance to ensure a comprehensive evaluation of the AGWF’s performance. The evaluation indicated that AGWF can improve the Signal-to-Noise Ratio (SNR) value and reduce the Root Mean Square Error (RMSE) and Mean Square Error (MSE) value, showing a qualitative improvement in CT scan imagery. The proposed method holds promising potential for advancing medical imaging technology with the implementation of deep learning.

An advanced deep neural network for fundus image analysis and enhancing diabetic retinopathy detection

Diabetic retinopathy (DR) involves retina damage due to diabetes, often leading to blindness. It is diagnosed via color fundus injections, but the manual analysis is cumbersome and error-prone. While computer vision techniques can predict DR stages, they are computationally intensive and struggle with complex data extraction. In this research, our prime objective was to automate the process of DR classification into its various stages using convolutional neural network (CNN) models. We employed the performance of fifteen pre-trained models with our novel proposed diabetic retinopathy network (DRNet13) model. We aimed to discern the most efficient model for accurate diabetic retinopathy (DR) staging based on fundus images from five DR classes. We preprocessed the image using a median filter for noise reduction and Gamma correction for image enhancement. We expanded our dataset from 3662 to 7500 images to create a more generalized training model through various augmentation techniques. We also evaluated multiple evaluation metrics, including accuracy, precision, F1-score, Sensitivity, Specificity, Area under the curve (AUC), Mean Squared Error (MSE), False Positive Rate (FPR), False Negative Rate (FNR), in addition to confusion matrices for an in-depth comparison of the performance of these models. Feature maps were employed to illuminate decision making areas in the DRNet13 model, which achieved a 97 % accuracy rate for DR detection, surpassing other CNN architectures in speed and efficiency. Despite a few misclassifications, the model's capability to identify critical features demonstrates its potential as an impactful diagnostic tool for timely and accurate identification of diabetic retinopathy.

Neutrosophic Fuzzy Logic based SVM approach for Enhanced Skin Cancer Prediction

In this study, a thorough methodology is used to present a unique way for improving skin cancer prediction accuracy. The research uses sophisticated preprocessing methods, such as the Frost filter for noise reduction and histogram equalization for contrast enhancement, to boost contrast on dermoscopic pictures from various sources, using an ISIC 2020 dataset. These actions greatly raise the dermoscopic pictures’ overall quality and usefulness for diagnosis. Utilizing labeled data for training, we offer a Fuzzy-based C-means clustering technique based on Neutrosophic Logic during the segmentation phase. In order to overcome ambiguities in skin lesion segmentation, the neutrosophic set—a groundbreaking idea in philosophy—is used. The suggested model enhances the accuracy of segmentation by modifying the neutrosophic set functions. For precise prediction, the approach combines Support Vector Machine (SVM) classification with Histogram of Oriented Gradient (HOG) feature extraction. While SVM, a supervised learning algorithm, diagnoses skin lesions based on the collected features, HOG features capture gradient information. To improve object recognition and classification, the HOG-SVM architecture is made to methodically collect and quantify essential information using dermoscopic pictures. The use of Neutrosophic Fuzzy Logic, which combines the benefits of fuzzy clustering with neutrosophic sets to produce more precise and nuanced predictions, sets the suggested method apart. The integration of different approaches into a holistic solution for skin cancer prediction is what makes the proposed study innovative. Findings and performance analysis show of the HOG-SVM method exhibits an outstanding accuracy of 98.69%, outperforming LR, KNN, and GNB methods. Python software is used to accomplish the suggested approach. This discovery opens up a possible path for better skin cancer diagnosis and advances the rapidly developing fields of dermatology and medical image processing.

Evaluation of Speckle Noise Reduction Filters and Machine Learning Algorithms for Ultrasound Images

Evaluation of Speckle Noise Reduction Filters and Machine Learning Algorithms for Ultrasound Images

Comparative Evaluation of Filters for Speckle Noise Reduction in a Clinical Liver Ultrasound Image

Comparative Evaluation of Filters for Speckle Noise Reduction in a Clinical Liver Ultrasound Image

Efficient Classification of Marine Debris using SVM with Noise Removal and Feature Extraction Techniques with Improved Performances

Marine debris is a significant environmental issue, necessitating the development of precise and efficient technologies for classifying and reducing its impact. This research tries to address the issues of image noise in ocean trash classification using four alternative filters the Gaussian, Bilateral, Mean, and Alpha-Trimmed Mean (ATM) filters for noise reduction and SVM for classification. The accuracy of categorization algorithms can be considerably impacted by the presence of noise in photos of maritime debris. In this study, we suggest a two-pronged strategy: To effectively decrease noise and improve the quality of the input photos, we first apply filters. These filters were chosen with care to protect significant details while eliminating extraneous noise. Second, we classify the debris into various categories based on its visual attributes using SVM, a powerful ML technique. The ATM filter significantly reduces noise and improves the clarity of photographs of ocean trash, according to the results of our experiments. This work proposes a novel approach for classifying marine debris using advanced machine learning algorithms. For improved classification accuracy, we suggest combining Support Vector Machines (SVM) with Adaptive Thresholding Mean (ATM) filtering and Histogram of Oriented Gradients (HOG) feature extraction. According to this research, the ATM filter is a promising option for noise reduction in ocean trash imaging, potentially increasing the precision of subsequent classification algorithms and assisting in efficient environmental monitoring and marine ecosystem conservation efforts. The language used for execution is Python.

Parameter Optimization of Noise-Reduction Filters of Hi-Performance Power Supplies

For a noise-reduction filter, the parameters of the manufactured device often do not fit the expected parameters during development, what is caused by the spread of the parameters of the electronic components or their parasitic parameters, as well as due to the presence of its own capacitance and inductance of the frequency response assembly. In this article, the influence of parasitic parameters of components and fasteners on the filter frequency response is considered. Utilizing the electronic simulator software, the filter parameters were modeled in the frequency range from 10 kHz to 1 GHz. Based on the simulation results, recommendations were given for optimizing the filter mounting elements and changing the inductance of low-frequency chokes.

A position-adaptive noise-reduction method using a deep denoising filter bank for dedicated breast positron emission tomography images.

Dedicated breast positron emission tomography (db-PET) is more sensitive than whole-body positron emission tomography and is thus expected to detect early stage breast cancer and determine treatment efficacy. However, it is challenging to decrease the sensitivity of the chest wall side at the edge of the detector, resulting in a relative increase in noise and a decrease in detectability. Longer acquisition times and injection of larger amounts of tracer improve image quality but increase the burden on the patient. Therefore, this study aimed to improve image quality via reconstruction with shorter acquisition time data using deep learning, which has recently been widely used as a noise reduction technique. In our proposed method, a multi-adaptive denoising filter bank structure was introduced by training the training data separately for each detector area because the noise characteristics of db-PET images vary at different locations. Input and ideal images were reconstructed based on 1- and 7-min collection data, respectively, using list mode data. The deep learning model used residual learning with an encoder-decoder structure. The image quality of the proposed method was superior to that of existing noise reduction filters such as Gaussian filters and nonlocal mean filters. Furthermore, there was no significant difference between the maximum standardized uptake values before and after filtering using the proposed method. Taken together, the proposed method is useful as a noise reduction filter for db-PET images, as it can reduce the patient burden, scan time, and radiotracer amount in db-PET examinations.

An effective adaptive fuzzy filter for speckle noise reduction

An effective adaptive fuzzy filter for speckle noise reduction

Optimization of Median Modified Wiener Filter for Improving Lung Segmentation Performance in Low-Dose Computed Tomography Images

In low-dose computed tomography (LDCT), lung segmentation effectively improves the accuracy of lung cancer diagnosis. However, excessive noise is inevitable in LDCT, which can decrease lung segmentation accuracy. To address this problem, it is necessary to derive an optimized kernel size when using the median modified Wiener filter (MMWF) for noise reduction. Incorrect application of the kernel size can result in inadequate noise removal or blurring, degrading segmentation accuracy. Therefore, various kernel sizes of the MMWF were applied in this study, followed by region-growing-based segmentation and quantitative evaluation. In addition to evaluating the segmentation performance, we conducted a similarity assessment. Our results indicate that the greatest improvement in segmentation performance and similarity was at a kernel size 5 × 5. Compared with the noisy image, the accuracy, F1-score, intersection over union, root mean square error, and peak signal-to-noise ratio using the optimized MMWF were improved by factors of 1.38, 33.20, 64.86, 7.82, and 1.30 times, respectively. In conclusion, we have demonstrated that by applying the MMWF with an appropriate kernel size, the optimization of noise and blur reduction can enhance segmentation performance.

Measurements of gas-liquid two-phase flow dynamics using high-speed neutron imaging

Gas-liquid two-phase flow appears in many heat-exchanging devices. The two-phase flow dynamics should be clarified to understand the phenomena in such devices. Although many measurement techniques have been applied to two-phase flow experiments, measuring the gas-liquid interfacial structure is difficult, which changes temporally and spatially. In this study, high-speed neutron imaging is used to measure two-phase flow dynamics, and the accuracy of the void fraction measurement is investigated. In high-speed neutron imaging, image blurring and distortion occur due to light and object motion intensifying. As a result, the quantitative accuracy might decrease. So, the rotating stainless-steel calibration disk, which simulates the bubble behavior in water, is observed by high-speed neutron imaging. Several noise reduction filters are tested to remove the blur and noise in the acquired images. Finally, the air-water two-phase flow is visualized by high-speed neutron imaging, and noise filtering is applied.

Enhancement and labelling of OCT images

Abstract Optical Coherence Tomography (OCT) is a noninvasive technique for visualizing retinal cross-sections, assisting in diagnosing and monitoring retinal diseases. This study presents an effective OCT image noise reduction and segmentation method. Due to the high resolution of our primary image database, a visual comparison of filter performance was challenging. To address this, we utilized a high-noise image dataset from Duke University, enabling a better evaluation of noise reduction filters. Our method was compared against eight widely-used image filters, including Gaussian, Low pass,Wavelet domain filtering, Lee filter, Anisotropic diffusion, Bilateral filter, Total variational filter, and BM3D filter. Both visual and quantitative analyses were conducted using no-reference performance parameters, namely Gradient Magnitude Similarity Deviation (GMSD), Standard Deviation of Wavelet Coefficients (SDWC), Focus Measure with Tengrade Variance (FMTV), Perception-based Image Quality Evaluator (PIQE), and Visual Information Fidelity (VIF). The results demonstrated the superiority of our proposed filter in terms of noise reduction performance while maintaining the sharpness of retinal layers. Quantitative analysis revealed notable performance gains, including improvements of 63.27% to 83.24% with GMSD, consistent edge strength enhancement of 9.4% to 9.97% using SDWC, gains in image quality between 51.99% and 54.64% with FMTV, performance improvements ranging from 7.25% to 23.97% in terms of PIQE, and a substantial increase in performance varying from 16.31% to 27.69% when assessing the impact on retinal layer quality using VIF. Using the Kruskal-Wallis test, our proposed noise reduction method shows statistical significance for all quantitative parameters. Additionally, our clustering algorithm effectively separated the foreground, including retinal layers and vitreous detachment, from the background and identified an area representing the region between retinal layers where fluid accumulates. We’ve successfully achieved OCT image enhancement, along with distinct foreground and background segmentation.

Cuckoo Search Optimization-Based Bilateral Filter for Multiplicative Noise Reduction in Satellite Images

&lt;div&gt;Speckle noise degrades the visual appearance and the quality of a synthetic aperture radar (SAR) image. The reduction of speckle noise is the first step in any remote-sensing device. To improve the noisy SAR images, a variety of adaptive and nonadaptive noise reduction filters were used. In order to eliminate speckle noise present in SAR images, an adaptive cuckoo search optimization-based speckle reduction bilateral filter has been designed in this article. To test the ability to eliminate multiplicative noise, the suggested filter’s effectiveness was compared to that of several de-speckling approaches. It has been measured with different assessment metrics such as PSNR, EPI, SSIM, and ENL. When compared to conventional de-noising filters, the proposed filter shows promising results for lowering speckle noise and retaining edge properties. In addition, the PSNR value has increased as compared to the PMD method and this method has been shown to be efficient in reducing speckle noise in SAR images.&lt;/div&gt;

Snow Attenuation of Infrasound Signals and Wind Noise

Abstract Infrasound monitoring of geophysical and human-sourced activity, such as explosions, eruptions, earthquakes, and avalanches, may be impacted by burial of sensors in a snowpack. We studied the frequency-dependent attenuation of sound, vertical sound speed, and wind noise reduction in a natural snowpack using earthquakes as a natural source of infrasound. Our experiment used a 2 m vertical distribution of sensors deployed during the 2021–2022 winter in the Sawtooth mountains of Idaho. We recorded 38 earthquakes and their primary infrasound propagating vertically upward through the snow. For these events, average attenuation appeared negligible between 0 and 8 Hz, rose to 0.1 dB/cm at ∼12 Hz, and remained at this level for higher frequencies. Using timing lags across the vertical profile the speed of infrasound in the snow was measured at 160 ± 51 m/s, significantly subsonic and indicative of pore-space waves. Finally, we quantified the significant reduction in the amount of background noise measured for snow-buried compared with free-air sensors. The 24 dB of reduction of noise observed for high wind speed conditions was evident with only 20 cm of snow burial and demonstrates the powerful capability of snowpack as a noise reduction filter. This study provides a clearer understanding of how snow burial filters infrasound and minimizes noise levels and has implications for monitoring of both natural and human-made hazards.