De-noising Effect Research Articles

A Novel Lidar Signal Denoising Method Based on Convolutional Autoencoding Deep Learning Neural Network

The lidar is susceptible to the dark current of the detector and the background light during the measuring process, which results in a significant amount of noise in the lidar return signal. To reduce noise, a novel denoising method based on the convolutional autoencoding deep-learning neural network is proposed. After the convolutional neural network was constructed to learn the deep features of lidar signal, the signal details were reconstructed by decoding part to obtain the denoised signal. To verify the feasibility of the proposed method, both the simulated signals and the actually measured signals by Mie-scattering lidar were denoised. Some comparisons with the wavelet threshold denoising method and the variational modal decomposition denoising method were performed. The results show the denoising effect of the proposed method was significantly better than the other two methods. The proposed method can eliminate complex noise in the lidar signal while retaining the complete details of the signal.

Nuchal Translucency Thickness Measurement in Fetal Ultrasound Images to Analyze Down Syndrome

Down syndrome (DS) is a chromosomal condition associated with intellectual disability and developmental delay in infants. The most salient pre-natal marker to identify DS during the early stages of gestation is the Nuchal translucency (NT) thickness. Accurate NT measurement from ultrasound (US) images becomes challenging due to the presence of speckle noise, weak edges and other artifacts. This paper proposes a semi-automated approach for early identification and calibration of NT thickness by estimating the distance between the edges of NT. A comparative study is performed on six traditional filters to determine the effective denoising filter for US datasets. This paper also presents an evaluation of two popular segmentation algorithms, the Active Contour and Chan-Vese-based segmentation. The study reveals that Wiener filter outperformed other filters based on the performance metrics and achieved peak signal noise ratio of 42.88 dB and mean squared error of −3.45. The prime feature NT region has been segmented effectively using enhanced double region-based active contour segmentation and obtained Dice Similarity Index (DSI- 84%) and Jaccard Index (76%). This technique has been validated by using Bland Altman’s plot to determine the correlation between clinical and proposed method.

Edge detection algorithm based on valley lines and watershed

The purpose of edge detection algorithm is to highlight the edge in the image, which is also the premise and key of image segmentation. However, the traditional edge detection algorithm will amplify the influence of noise in the process of derivative operation, and the noisy edges will often appear, which is not conducive to the accurate extraction of edge. To solve this problem, we propose an edge detection algorithm combining valley lines and watershed algorithm. In this algorithm, the bilateral filter is used to replace the Gaussian filter to achieve edge preserving and denoising. The rewritten Canny operator is used to calculate the low value points and get the valley lines of the image. This process can achieve the first denoising. Then, the valley lines are used to replace the discrete points of water injection and the watershed algorithm is used to calculate the segmentation results of the image. At this time, there are many pseudo edges around the main boundary, and the pseudo edges are deleted by setting the threshold, so as to achieve the effect of secondary denoising. The experimental results show that our method has good performance in noise resistance and edge continuity compared with other and traditional edge detection algorithms.

Defining of the Minimum Wavelet Filter Coefficients Bit-Width for 3D Medical Imaging

Medical imaging uses a variety of modalities to provide visual information about a patient. Various methods are used to process this data. Many of them are based on discrete wavelet transform (DWT). Its use will allow effective denoising and compression of 2D and 3D images. This paper proposes a new approach to linear time-invariant wavelet filtering using quantized filter coefficients when using which the computational errors have different signs and allow to partially compensate each other as a result of which the processed image is of high quality. The analysis of the quantization noise of the direct multilevel DWT filter coefficients is carried out. The derived formulas demonstrate the relationship between the quantization accuracy of these coefficients and the processing quality of digital 3D images. The derived formulas for calculating the minimum accuracy of the wavelet filter coefficients representation in the computing devices memory allow minimizing the effect of quantization noise on the result of 3D images processing. Modelling of 3D medical tomographic images DWT processing showed that a decrease in the ratio of the average voxel brightness to the maximum allowable value with increasing color depth of images leads to faster achievement of high quality compared to the results of theoretical analysis with an increase in the value of the scaling degree of the wavelet filter coefficients. The obtained theoretical and practical results open up the possibility for reducing the computational complexity of software and hardware implementation of wavelet processing of 3D medical visual data on modern microelectronic devices (field-programmable gate arrays, application-specific integrated circuits, etc.).

Extreme Learning Machine Denoising Algorithm Based Analysis of Transvaginal 3-Dimensional Ultrasonic Image for the Diagnostic Effect of Intrauterine Adhesion

The aim was to analyze the application values and diagnostic effects of transvaginal 3-dimensional (3D) ultrasonic image based on extreme learning machine denoising algorithm (ELMDA) in the diagnosis of intrauterine adhesions (IUA). The speckle noise in the 3D ultrasound image was removed with the ELMDA. Its peak signal-to-noise ratio (PSNR) and the mean square error (MSE) were compared with those of the median filter algorithm (MFA) with the anisotropic diffusion algorithm (ADA) and wavelet threshold. The ELMDA was used in the diagnosis of 3D ultrasound images to compare the accuracy of hysteroscopy with transvaginal 3D ultrasound and two-dimensional (2D) ultrasound in the diagnosis of IUA. The results showed that the MSE of ELMDA was dramatically smaller than those of ADA and WT-MFA and its PSNR was higher than those of the other two algorithms ( P < 0.05) when the noise variance was constant. The diagnostic accuracy of mild and moderate adhesions by 2D ultrasound was statistically different ( P < 0.05) compared with hysteroscopy. But the diagnosis results of severe adhesions were consistent, and the diagnosed cases were both 6 (11.11%) with no statistical difference ( P > 0.05). In addition, there was no statistically great difference in the diagnostic accuracy of IUA by transvaginal 3D ultrasound and hysteroscopy ( P > 0.05), and the diagnosis results of moderate and severe adhesions were consistent (both 20 cases (37.04%) and 6 cases (11.11%), respectively) with no statistical difference ( P > 0.05). The diagnostic accuracy of 3D ultrasound was 96.30%, while that of 2D ultrasound was 90.74%, showing a statistical difference ( P < 0.05). In conclusion, ELMDA had a good effect of denoising, and there was a high accuracy of the application of 3D transvaginal ultrasound to diagnose IUA, which had reliable clinical application value.

An adaptive recognition method for take-off action images of back-style high jump based on feature extraction

In the process of human movement, the behavioral characteristics of human body are complex and diverse. Thus, it is difficult to use accurate shape to represent the uncertainty of human behavior, which brings difficulties for 3D modeling in the state of motion. To solve this problem, an adaptive image recognition method based on feature extraction is proposed in this paper. Firstly, the image noise threshold was obtained by wavelet coefficient conversion. Then the wavelet coefficient matrix was constructed on the basis of decomposed image multi-resolution, and the image noise was removed according to the threshold. Based on this, Gaussian function in Canny method is introduced to obtain the edge coordinates of moving targets by convolution operation. After that, the image rotation and translation matrices were resolved according to the Murkowski distance to obtain the motion feature vectors. Finally, the hidden Markov model was constructed to segment the background and the target according to the feature sequence and the target edge coordinates, so as to realize the adaptive recognition of the take-off action images of the back-style high jump. Experimental results showed that this method had the advantages of high denoising effect, accurate feature extraction, error free edge division and accurate recognition result.

Fuzzy System Based Medical Image Processing for Brain Disease Prediction.

The present work aims to explore the performance of fuzzy system-based medical image processing for predicting the brain disease. The imaging mechanism of NMR (Nuclear Magnetic Resonance) and the complexity of human brain tissues cause the brain MRI (Magnetic Resonance Imaging) images to present varying degrees of noise, weak boundaries, and artifacts. Hence, improvements are made over the fuzzy clustering algorithm. A brain image processing and brain disease diagnosis prediction model is designed based on improved fuzzy clustering and HPU-Net (Hybrid Pyramid U-Net Model for Brain Tumor Segmentation) to ensure the model safety performance. Brain MRI images collected from a Hospital, are employed in simulation experiments to validate the performance of the proposed algorithm. Moreover, CNN (Convolutional Neural Network), RNN (Recurrent Neural Network), FCM (Fuzzy C-Means), LDCFCM (Local Density Clustering Fuzzy C-Means), and AFCM (Adaptive Fuzzy C-Means) are included in simulation experiments for performance comparison. Results demonstrate that the proposed algorithm has more nodes, lower energy consumption, and more stable changes than other models under the same conditions. Regarding the overall network performance, the proposed algorithm can complete the data transmission tasks the fastest, basically maintaining at about 4.5 s on average, which performs remarkably better than other models. A further prediction performance analysis reveals that the proposed algorithm provides the highest prediction accuracy for the Whole Tumor under DSC (Dice Similarity Coefficient), reaching 0.936. Besides, its Jaccard coefficient is 0.845, proving its superior segmentation accuracy over other models. In a word, the proposed algorithm can provide higher accuracy, a more apparent denoising effect, and the best segmentation and recognition effect than other models while ensuring energy consumption. The results can provide an experimental basis for the feature recognition and predictive diagnosis of brain images.

Research on De-noising of Downhole Engineering Parameters by Wavelet based on Improved Threshold Function

The measurement of downhole engineering parameters is greatly disturbed by the working environment. Effective de-noising methods are required for processing logging-while-drilling (LWD) acquisition signals, in order to obtain downhole engineering parameters accurately and effectively. In this paper, a new de-noising method for measuring downhole engineering parameters was presented, based on a feedback method and a wavelet transform threshold function. Firstly, in view of the mutability and density of downhole engineering data, an improved wavelet threshold function was proposed to de-noise the signal, so as to overcome the shortcomings of data oscillation and deviation caused by the traditional threshold function. Secondly, due to the unknown true value, traditional single denoising effect evaluation cannot meet the requirements of quality evaluation very well. So the root mean square error (RMSE), signal-to-noise ratio (SNR), smoothness (R) and fusion indexs (F) are used as the evaluation parameters of the de-noising effect, which can determine the optimal wavelet decomposition scale and the best wavelet basis. Finally, the proposed method was verified based on the measured downhole data. The experimental results showed that the improved wavelet de-noising method could reduce all kinds of interferences in the LWD signal, providing reliable measurement for analyzing the working status of the drilling bit.

Denoising Speech Based on Deep Learning and Wavelet Decomposition

The work proposed a denoising speech method using deep learning. The predictor and target network signals were the amplitude spectra of the wavelet-decomposition vectors of the noisy audio signal and clean audio signal, respectively. The output of the network was the amplitude spectrum of the denoised signal. Besides, the regression network used the input of the predictor to minimize the mean square error between its output and input targets. The denoised wavelet-decomposition vector was transformed back to the time domain by the output amplitude spectrum and the phase of the wavelet-decomposition vector. Then, the denoised speech was obtained by the inverse wavelet transform. This method overcame the problem that the frequency and time resolution of the short-time Fourier transform could not be adjusted. The noise reduction effect in each frequency band was improved due to the gradual reduction of the noise energy in the wavelet-decomposition process. The experimental results showed that the method has a good denoising effect in the whole frequency band.

Wavelet method optimised by ant colony algorithm used for extracting stable and unstable signals in intelligent substations

Partial discharge (PD) signals are an important index to evaluate the operation state of intelligent substations. The correct distinction of PD pulse and interference pulse has become a challenging task. Because of the noise and the low signal-to-noise ratio, the stable signals become non-stationary. The selection of a wavelet basis, the selection rule of threshold λ and the design of the threshold function are the key factors affecting the final denoising effect. Therefore, an enhanced ant colony optimisition wavelet (ACOW) algorithm was applied to find the global optimal threshold through the continuous derivative threshold function and the ant colony optimisation (ACO) algorithm. At the same time the efficiency of adaptive search calculation, was also significantly improved. The method of the ACOW algorithm was compared with the soft wavelet method, gradient-based wavelet method and the genetic optimisation wavelet (GOW) method. Using these four methods to denoise four typical signals, different mean square errors (MSE), magnitude errors (ME) and time costs were obtained. Interestingly, the results show that the ACOW method can achieve the minimum MSE and has less time cost. It generates significantly smaller waveform distortion than the other three threshold estimation methods. In addition, the high efficiency and good quality of the output signals are beneficial to the diagnosis of local discharge signals in intelligent substations.

Random noise suppression of seismic data based on deep convolution neural network

Random noise suppression of seismic data is essential in seismic data processing. Since the seismic data recorded by the geophone is usually noisy, this kind of noisy data can be regarded as a manifestation of low signal-to-noise ratio. Low SNR data will affect subsequent processing of seismic data, such as migration and imaging. In this paper, we aim to improve the imaging quality of seismic data and propose an intelligent noise reduction framework for convolutional neural network to adaptively learn seismic signals from noisy seismic data. In order to speed up network training and avoid gradient disappearance during training, we add residual learning and batch normalization methods to the network, and use ReLU activation function and Adam optimization algorithm to optimize the network. In addition, the two datasets, Marmousi and F3, are used to train and test the network. A fully trained network can not only retain weak features in learning, but also remove random noise. First, fully train the network, extract random noise from it, and retain the learned seismic data features, and then estimate the waveform features in the test set by reconstructing the seismic data. The processing results of synthetic records and field data show the potential of deep convolutional neural network in random noise suppression tasks, and experimental verification shows that the deep convolutional neural network framework has a good denoising effect.

Research on Denoising of Brain MRI of Alzheimer's Disease Based on BM3D Algorithm

Alzheimer's disease is a chronic degenerative disease of the acquired nervous system. Early detection and early treatment are particularly important. According to the technical requirements of brain magnetic imaging, the important step before brain magnetic imaging segmentation is image preprocessing and image denoising, which can more accurately segment the effective lesion features of magnetic imaging through image recognition. The algorithm studied in this paper is the BM3D method of image preprocessing and denoising. The BM3D noise algorithm model is a non-local denoising method, which is a spatial algorithm, and the other transform method is a conversion algorithm, which has a significant denoising effect. The image is grouped by two similar blocks through the BM3D algorithm and then collaboratively filtered and aggregated. The experimental analysis shows that the image processed by the BM3D algorithm is significantly better than the original image, and it can solve the problem of noise in the image and blurring of the boundary between the tissues.

HashHeat: A hashing-based spatiotemporal filter for dynamic vision sensor

Neuromorphic event-based dynamic vision sensors (DVS) have much faster sampling rates and a higher dynamic range than frame-based imagers. However, they are sensitive to background activity (BA) events that are unwanted. We propose HashHeat, a hashing-based spatiotemporal BA filter for DVS. It is the first spatiotemporal filter that doesn't scale with the DVS output size and doesn't store the 32-bits timestamps. We not only give the visual denoising effect of the filter but also use two metrics for quantitatively analyzing the filter's global performance and local performance respectively, where we introduce a novel metric for evaluating global performance. The experimental results show that HashHeat achieves similar global performance as baseline filters, but increases the signal to noise ratio by about 1.5x to nearly 5x compared with other baseline filters concerning the local performance. The hardware implementation enables HashHeat to output a labeled event every 10ns and meets the real-time requirement. And it can reduce the storage cost by 128x to 256x compared with baseline filters.

Towards robust and understandable fault detection and diagnosis using denoising sparse autoencoder and smooth integrated gradients

Industrial applications of fault detection and diagnosis face great challenges as they require not only accurate identification of faulty statuses but also the effective understandability of the results. In this paper, a two-step robust and understandable fault detection and diagnosis framework is developed to address this challenge by exploiting denoising sparse autoencoder and smooth integrated gradients. Specifically, denoising sparse autoencoder(DSAE) is utilized to detect faults in the first step. DSAE is more robust to noise corruption and has better generalization performance compared to the existing autoencoder-based methods. In the second step, smooth integrated gradients(SIG) is used to diagnose the root-cause variables of the faults detected. Smooth integrated gradients can provide a denoising effect on the feature importance. The proposed framework is evaluated through an application to the Tennessee Eastman process. As proved in the experiments, the presented DSAE-SIG method not only achieves higher diagnosis accuracy but also successfully identifies the potential root-cause variables of process disturbances.

Effect of De-noising by Wavelet Filtering and Data Augmentation by Borderline SMOTE on the Classification of Imbalanced Datasets of Pig Behavior

Classification of imbalanced datasets of animal behavior has been one of the top challenges in the field of animal science. An imbalanced dataset will lead many classification algorithms to being less effective and result in a higher misclassification rate for the minority classes. The aim of this study was to assess a method for addressing the problem of imbalanced datasets of pigs' behavior by using an over-sampling method, namely Borderline-SMOTE. The pigs' activity was measured using a triaxial accelerometer, which was mounted on the back of the pigs. Wavelet filtering and Borderline-SMOTE were both applied as methods to pre-process the dataset. A multilayer feed-forward neural network was trained and validated with 21 input features to classify four pig activities: lying, standing, walking, and exploring. The results showed that wavelet filtering and Borderline-SMOTE both lead to improved performance. Furthermore, Borderline-SMOTE yielded greater improvements in classification performance than an alternative method for balancing the training data, namely random under-sampling, which is commonly used in animal science research. However, the overall performance was not adequate to satisfy the research needs in this field and to address the common but urgent problem of imbalanced behavior dataset.

MR-guided motion-corrected PET image reconstruction for cardiac PET-MR.

Simultaneous PET-MR imaging has shown potential for the comprehensive assessment of myocardial health from a single examination. Furthermore, MR-derived respiratory motion information has been shown to improve PET image quality by incorporating this information into the PET image reconstruction. Separately, MR-based anatomically guided PET image reconstruction has been shown to perform effective denoising, but this has been so far demonstrated mainly in brain imaging. To date the combined benefits of motion compensation and anatomical guidance have not been demonstrated for myocardial PET-MR imaging. This work addresses this by proposing a single cardiac PET-MR image reconstruction framework which fully utilises MR-derived information to allow both motion compensation and anatomical guidance within the reconstruction. Methods: Fifteen patients underwent a 18F-FDG cardiac PET-MR scan with a previously introduced acquisition framework. The MR data processing and image reconstruction pipeline produces respiratory motion fields and a high-resolution respiratory motion-corrected MR image with good tissue contrast. This MR-derived information was then included in a respiratory motion-corrected, cardiac-gated, anatomically guided image reconstruction of the simultaneously acquired PET data. Reconstructions were evaluated by measuring myocardial contrast and noise and compared to images from several comparative intermediate methods using the components of the proposed framework separately. Results: Including respiratory motion correction, cardiac gating, and anatomical guidance significantly increased contrast. In particular, myocardium-to-blood pool contrast increased by 143% on average (p<0.0001) compared to conventional uncorrected, non-guided PET images. Furthermore, anatomical guidance significantly reduced image noise compared to non-guided image reconstruction by 16.1% (p<0.0001). Conclusion: The proposed framework for MR-derived motion compensation and anatomical guidance of cardiac PET data was shown to significantly improve image quality compared to alternative reconstruction methods. Each component of the reconstruction pipeline was shown to have a positive impact on the final image quality. These improvements have the potential to improve clinical interpretability and diagnosis based on cardiac PET-MR images.

GPR B-Scan Image Denoising via Multi-Scale Convolutional Autoencoder with Data Augmentation

Ground-penetrating radar (GPR), as a non-invasive instrument, has been widely used in civil engineering. In GPR B-scan images, there may exist random noise due to the influence of the environment and equipment hardware, which complicates the interpretability of the useful information. Many methods have been proposed to eliminate or suppress the random noise. However, the existing methods have an unsatisfactory denoising effect when the image is severely contaminated by random noise. This paper proposes a multi-scale convolutional autoencoder (MCAE) to denoise GPR data. At the same time, to solve the problem of training dataset insufficiency, we designed the data augmentation strategy, Wasserstein generative adversarial network (WGAN), to increase the training dataset of MCAE. Experimental results conducted on both simulated, generated, and field datasets demonstrated that the proposed scheme has promising performance for image denoising. In terms of three indexes: the peak signal-to-noise ratio (PSNR), the time cost, and the structural similarity index (SSIM), the proposed scheme can achieve better performance of random noise suppression compared with the state-of-the-art competing methods (e.g., CAE, BM3D, WNNM).

Optimized Double-Regional Filtering Algorithm on MRI Three-Dimensional Reconstructed Images for the Evaluation of Effects of Delivery on the Pelvis of Primiparas

This study was to explore the denoising and segmentation effect of dual-domain image denoising (DDID) algorithm, and the Galois field (GF) and nonlocal means (NLM) algorithms were introduced for comparative analysis. 40 primiparas in the hospital from January 2018 to January 2020 were divided into an experimental group (caesarean section (CS), group E) and a control group (vaginal delivery (VD), group C). The peak signal-to-noise ratio (PSNR) and segmentation parameters of DDID algorithm were compared with GF algorithm and NLM algorithm. It was found that the DDID showed higher overall accuracy (OA) and lower false positive rate (FPR) and false negative rate (FNR). The PSNR of DDID was higher than the other two algorithms. GF algorithm showed the highest edge retention index (ERI). The incidence of pelvic organ prolapse (POP) in group E and group C was 9/20 (45%) and 5/20 (25%), respectively, with extreme difference ( P &lt; 0.05 ). Evaluation of the effects of delivery on the pelvis of primiparas with MRI three-dimensional (3D) reconstructed images based on the optimized DDID showed a superior and stable denoising effect and good segmentation, so it was worthy of clinical promotion and application.

Image Denoising Using Adaptive and Overlapped Average Filtering and Mixed-Pooling Attention Refinement Networks

Cameras are essential parts of portable devices, such as smartphones and tablets. Most people have a smartphone and can take pictures anywhere and anytime to record their lives. However, these pictures captured by cameras may suffer from noise contamination, causing issues for subsequent image analysis, such as image recognition, object tracking, and classification of an object in the image. This paper develops an effective combinational denoising framework based on the proposed Adaptive and Overlapped Average Filtering (AOAF) and Mixed-pooling Attention Refinement Networks (MARNs). First, we apply AOAF to the noisy input image to obtain a preliminarily denoised result, where noisy pixels are removed and recovered. Next, MARNs take the preliminary result as the input and output a refined image where details and edges are better reconstructed. The experimental results demonstrate that our method performs favorably against state-of-the-art denoising methods.

De-noising method of tunnel blasting signal based on CEEMDAN decomposition-wavelet packet analysis

Aiming at the measured vibration signals collected during tunnel blasting construction, a noise reduction method based on the overall average empirical mode decomposition method (CEEMDAN decomposition) combined with wavelet packet analysis was in troduced. First, a series of multiple intrinsic modal components were obtained by CEEMDAN decomposition, and the modal components containing noise were selected using correlation coefficients, checked by the spectrogram and the variance contribution rate of the modal components. Then, the wavelet packet threshold noise reduction method was used to process the modal components containing noise. Finally, the unprocessed modal components and the de-noised components were reconstructed to obtain the final pure blasting vibration signal. At the same time, the feasibility of this noise reduction method has been verified by wavelet packet energy spectrum analysis. This method combines the advantages of CEEMDAN decomposition and wavelet packet analysis. Compared with existing methods, the de-noising effect is better, and it can be applied to the de-noising processing of similar tunnel blasting signals.