Peak Signal To Noise Ration Research Articles

Singular value decomposition: A useful technique for image denoising

A key function of image processing is picture denoising, which improves the quality of images by eliminating extraneous noise while keeping crucial information in tact. Singular Value Decomposition (SVD) is a linear algebraic technique that reduces the original datas complexity and scale by breaking down the matrices and extracting the important information. With the power of decomposition which utilizes the non-local self-similarity property of an image to achieve satisfactory denoising performance, SVD denoising has become a potent tool in image processing. In this paper, SVD is outlined and its working, applications, and challenges as a denoising technique in image denoising are discussed. The author discovered that Singular Value Decomposition can be a significant factor in image denoising by applying it to the image. As a result, Singular Value Decomposition could be thought as a helpful image denoising approach in the image processing sequence that will raise the images Peak Signal-to-Noise Ration (PSNR) and improve the quality of the image.

United multi-task learning for abdominal contrast-enhanced CT synthesis through joint deformable registration

Synthesizing abdominal contrast-enhanced computed tomography (CECT) images from non-enhanced CT (NECT) images is of great importance, in the delineation of radiotherapy target volumes, to reduce the risk of iodinated contrast agent and the registration error between NECT and CECT for transferring the delineations. NECT images contain structural information that can reflect the contrast difference between lesions and surrounding tissues. However, existing methods treat synthesis and registration as two separate tasks, which neglects the task collaborative and fails to address misalignment between images after the standard image pre-processing in training a CECT synthesis model. Thus, we propose an united multi-task learning (UMTL) for joint synthesis and deformable registration of abdominal CECT. Specifically, our UMTL is an end-to-end multi-task framework, which integrates a deformation field learning network for reducing the misalignment errors and a 3D generator for synthesizing CECT images. Furthermore, the learning of enhanced component images and the multi-loss function are adopted for enhancing the performance of synthetic CECT images. The proposed method is evaluated on two different resolution datasets and a separate test dataset from another center. The synthetic venous phase CECT images of the separate test dataset yield mean absolute error (MAE) of 32.78±7.27 HU, mean MAE of 24.15±5.12 HU on liver region, mean peak signal-to-noise rate (PSNR) of 27.59±2.45 dB, and mean structural similarity (SSIM) of 0.96±0.01. The Dice similarity coefficients of liver region between the true and synthetic venous phase CECT images are 0.96±0.05 (high-resolution) and 0.95±0.07 (low-resolution), respectively. The proposed method has great potential in aiding the delineation of radiotherapy target volumes.

Image Steganography Based on Chaos Function and Randomize Function

The exchange of data is not limited to personal text information or information about institutions and governments, but includes digital media transferred via the Internet including everything, whether texts, images or videos and audio, or animation. These media need high-security protection and high speed during its transmission from one site to another. In this study, a new method is suggested for hiding a gray-level image within a larger color image based on the proposed steganography map that merged chaotic function and randomize function. The size of the chaos and randomize functions is 16 bytes. Experimental results obtained a successful method based on mean squared error, signal-to-noise ratio, peak signal noise rate, embedding capacity, entropy, and histogram. This method can rapidly hide and extract ciphertext in and from the gray image. The original image and the stego image are difficult to distinguish because the correlation between them is very close to 1, indicating that attackers cannot easily differentiate these images with the naked eye. This condition can successfully hide information on the Internet.

Noise Removal in Traffic Sign Detection Systems

The application of Traffic sign detection and recognition is growing in traffic assistant driving systems and automatic driving systems. It helps drivers and automatic driving systems to detect and recognize the traffic signs effectively. However, it is found that it may be difficult for these systems to work in challenging environments like rain, haze, hue, etc. To help the detection systems to have better performance in challenging conditions like rain and haze, we propose the use of a deep learning technique based on a Convolutional Neural Network to process visual data. The processed data could be used in the detection. We are using the NoiseNet model [11], a noise reduction network for our architecture. The model is trained to enhance images in patches instead of as a whole. The training is done using the Challenging Unreal and Real Environment - Traffic Sign Detection Dataset(CURE-TSD) which contains videos of different roads in various challenging situations. The enhanced images obtained are compared using the object detection algorithms YOLO and Faster RCNN. The Mean Absolute Error(MAE) of original and enhanced images are calculated and compared for two classes of images - rain and haze for both the algorithms. The proposed approach achieved an average Peak Signal to Noise Ration(PSNR) of 25.30 and an Structural Similarity(SSIM) of 0.88. The average MAE values of YOLO and Faster RCNN model reduced by 0.11 and 0.30 respectively on using enhanced images.

Complex Wavelets and SPIHTL for Edge Feature Enhancement and Optimum Feature Selection with Deep Learning Algorithm for Video Salient Object Detection

Cognitive property of the human visual system to detect salient objects in image and video sequences to correlate with neighbouring objects for information retrieval and understanding is developed for computer vision application. Detecting salient objects with accurate boundaries and features is addressed in this work by using the directional selectivity property of complex wavelets. Computation complexity of processing complex wavelet sub band is addressed by companding the features into eight sub bands and optimum number of features are selected using SPIHTL (Set Partition In Hierarchical Trees List) algorithm. Five layered deep learning algorithm is designed with 2D FFNN ( Feed Forward Neural Network) structure and reordering units to detect salient object features. The 3D DTCWT (Dual Tree Complex Wavelet Transform) model combined with 2D SPIHTL encoder and deep learning algorithm is modelled in MATLAB and is trained to identify 120 objects with PSNR (Peak Signal To Noise Ration) of 40dB. The designed model is demonstrated to detect generic objects. Keywords: Video saliency, DTCWT, Directional Features, SPIHTL, Deep Learning, 2D Neural Network

Editorial Real-Time Implementation of VVC Standard for Consumer Electronic Devices

We are pleased to present this Special Issue on this timely and important topic. The next generation ISO/IEC MPEG and ITU-T VCEG video coding standard called versatile video coding (VVC) was released in July 2020 <xref ref-type="bibr" rid="ref1">[1]</xref>, <xref ref-type="bibr" rid="ref2">[2]</xref>. Gls vvc includes some coding tools enabling an important coding gains estimated to 35&#x0025; of bit-rate reductions for the same quality in terms of peak signal to noise ration (PSNR) compared with high efficiency video coding (HEVC) in random access (RA) coding configurations. For all intra (AI) configuration the improvement is a bit lower and estimated in a 25&#x0025;. Moreover, subjective comparison recently conducted between HEVC and the VVC reference software, has shown that this gain is even higher reaching up to 50&#x0025; of bit-rate reduction for the same perceived quality <xref ref-type="bibr" rid="ref3">[3]</xref> - <xref ref-type="bibr" rid="ref4">[4]</xref>. Unfortunately the coding gain is associated to an increase in the algorithm complexity at both encoder and decoder sides. The VVC encoder is estimated to be <inline-formula> <tex-math notation="LaTeX">$8\times $ </tex-math></inline-formula> more complex than HEVC in RA configuration and <inline-formula> <tex-math notation="LaTeX">$25\times $ </tex-math></inline-formula>. The complexity of the VVC decoder is doubled compared to HEVC. Therefore, complexity reduction and optimization of VVC is critical to design of software and hardware real time VVC codecs for successful deployment on video consumer devices including among others TVs, mobile phones and tablets.

Image Compression Based on Hybrid Domain Attention and Postprocessing Enhancement.

Deep learning-based image compression methods have made significant achievements recently, of which the two key components are the entropy model for latent representations and the encoder-decoder network. Both the inaccurate estimation of the entropy estimation model and the existence of information redundancy in latent representations lead to a reduction in the compression efficiency. To address these issues, the study suggests an image compression method based on a hybrid domain attention mechanism and postprocessing improvement. This study embeds hybrid domain attention modules as nonlinear transformers in both the main encoder-decoder network and the hyperprior network, aiming at constructing more compact latent features and hyperpriors and then model the latent features as parametric Gaussian-scale mixture models to obtain more precise entropy estimation. In addition, we propose a solution to the errors introduced by quantization in image compression by adding an inverse quantization module. On the decoding side, we also provide a postprocessing enhancement module to further increase image compression performance. The experimental results show that the peak signal-to-noise rate (PSNR) and multiscale structural similarity (MS-SSIM) of the proposed method are higher than those of traditional compression methods and advanced neural network-based methods.

Classification of breast cancer with deep learning from noisy images using wavelet transform.

In this study, breast cancer classification as benign or malignant was made using images obtained by histopathological procedures, one of the medical imaging techniques. First of all, different noise types and several intensities were added to the images in the used data set. Then, the noise in images was removed by applying the Wavelet Transform (WT) process to noisy images. The performance rates in the denoising process were found out by evaluating Peak Signal to Noise Rate (PSNR) values of the images. The Gaussian noise type gave better results than other noise types considering PSNR values. The best PSNR values were carried out with the Gaussian noise type. After that, the denoised images were classified by Convolution Neural Network (CNN), one of the deep learning techniques. In this classification process, the proposed CNN model and the VggNet-16 model were used. According to the classification result, better results were obtained with the proposed CNN model than VggNet-16. The best performance (86.9%) was obtained from the data set created Gaussian noise with 0.3 noise intensity.

High-fidelity sub-Nyquist ghost imaging with tri-directional probing

Ghost imaging is an unconventional imaging method, which has invoked many applications in various fields. However, it is still a major challenge to achieve high-fidelity high-resolution images at a sub-Nyquist sampling rate. Here, we present a ghost imaging method that illuminates an object with three directional Tetris-like patterns, which can greatly trade off the contradiction between the high resolution and high detection signal-to-noise ratio. As the projected patterns gradually shrink during the detection, the image is also gradually recovered from low to high resolution. In addition, this method can recover complex chromatic objects without any compromising image quality by adaptively abandoning unnecessary patterns at sampling rates well below the Nyquist limit. Meanwhile, the dynamic probing scheme has an excellent noise-removal capability. The simulation and experiment demonstrate that the sampling rate to recover a high-fidelity image is only 10.7% for a scene of a 9.7% duty cycle. For a very noisy scene whose peak signal–noise rate (PSNR) is 10.18 dB [the structural similarity index (SSIM) is 0.068], this scheme increases the PSNR to 18.63 dB [structural similarity index (SSIM) to 0.73]. Therefore, the proposed method may be useful for ghost imaging in the low sampling rate regime or complex chromatic objects reconstruction.

Deep neural networks for magnetic resonance elastography acceleration in thermal-ablation monitoring.

To develop a deep neural network for accelerating magnetic resonance elastography (MRE) acquisition, to validate the ability to generate reliable MRE results with the down-sampled k-space data, and to demonstrate the feasibility of the proposed method in monitoring the stiffness changes during thermal ablation in a phantom study. MRE scans were performed with 60Hz excitation on porcine ex-vivo liver gel phantoms in a 0.36T MRI scanner to generate the training dataset. The acquisition protocol was based on a spin-echo MRE pulse sequence with tailored motion-sensitive gradients to reduce echo time (TE). A U-Net based deep neural network was developed and trained to interpolate the missing data from down-sampled k-space. We calculated the errors of 80 sets magnitude/phase images reconstructed from the zero-filled, compressive sensing (CS) and deep learning (DL) method for comparison. The peak signal-to-noise rate (PSNR) and structural similarity index (SSIM) of the magnitude/phase images were also calculated for comparison. The stiffness changes were recorded before, during, and after ablation. The mean stiffness values over the region of interest (ROI) were compared between the elastograms reconstructed from the fully sampled k-space and interpolated k-space after thermal ablation. The mean absolute error (MAE), PSNR, and SSIM of the proposed DL approach were significantly better than the results from the zero-filled method (p<0.0001) and CS (p<0.0001). The stiffness changes before and after thermal ablation assessed by the proposed approach (before: 7.7±1.1kPa, after: 11.9±4.0kPa, 4.2-kPa increase) gave close agreement with the values calculated from the fully sampled data (before: 8.0±1.0kPa, after: 12.6±4.2kPa, 4.6-kPa increase). In contrast, the stiffness changes computed from the zero-filled method (before: 4.9±1.4kPa, after: 5.6±2.8kPa, 0.7-kPa increase) were substantially underestimated. This study demonstrated the capability of the proposed DL method for rapid MRE acquisition and provided a promising solution for monitoring the MRI-guided thermal ablation.

Fusion of CT and MR Liver Images Using Multiresolution Analysis Methods

There are various medical imaging techniques such as Computed Tomography (CT) and Magnetic Resonance (MR) techniques. Both techniques give complex features of the region to be imaged. This study proposes an approach that uses Multiresolution Analysis (MRA) methods to fuse CT and MR liver images to obtain as detailed images as possible for medical diagnostic purposes. The transform coefficients are obtained by applying MRA methods to the images. Images are combined by applying 3 different fusion rules to these transform coefficients. Peak Signal to Noise Rate (PSNR), Structural Similarity Index Measure (SSIM) and Mean Square Error (MSE) values are calculated to evaluate the fused images. When comparing the methods, the best result was obtained using complex-valued curvelet transform.

Adaptive Colour Space based Robust Image Watermarking Using Serial DWT-HD-SVD domain

Watermarking is used for the authentication and protection of digital image data. The color image watermarking is a challenging task and may endup to be lengthy. Various transformations based methods are used forenhacingwatermarkrobustnessand to keep itinvisible. Robustness of watermarking methods have been improved using the hybrid DWT-SVD domain. In this paper an improved hybrid transformation method is prpoposed, using the combination of the Hesenberg Decomposition (HD) with the existing DWT-SVD method. The LAB colour space is adopted at the pre-processing stage for the entropy maximization. The entropy analysis is used for justification of the colour space adaptation.Performance of the existing and proposed method is tested on colour images using parametric analysis of normalized correlation (NC),Peak signal to noise ration (PSNR) and SSIM.Efficiency of existing waatermarking methods may detoriate undervarious attacks. The Impact of the various watermark attacks are also evaluated over the extraction quality.

State of the Art on Video Streaming and Applications

The real-time and live video streaming are playing an important role in our daily lives; therefore enhancing the quality of real-time and live video streaming over wired and wireless packet erasure networks which becomes more and more important topic nowadays. The main objective of this tutorial paper is to highlight state of the art on video streaming and the most important methods for enhancing video streaming over wired and wireless packet erasure networks to the computer/communication engineering students and new researchers. Firstly, it presents an overview of layered coding compression techniques (like: H.264/SVC) which lead to better real-time video streaming quality than single layer coding techniques (like: H.264/AVC). Secondly, it highlights how to solve the channel errors and lost packets problems in video streaming via implementing some of Forward Error Correction (FEC) techniques. Furthermore, it presents Backward Error Correction (BEC) techniques for enhancing the video streaming. After that it talks about the implementation of rateless FEC techniques with feedback for solving the real-time and live video streaming issues. Then it presents some recent researches on the Unequal Error (or Loss) Protection (UEP) techniques which are more powerful and achieve better Peak Signal to Noise Rate (PSNR) than the previous Equal Error Protection (EEP) techniques. Also it highlights the adaptive UEP scheme which brings significant improvements in terms of average PSNR over static UEP scheme for same used bandwidth. Furthermore, it talks about the multipath video streaming which leads to better loss characteristics and better load distribution on wireless networks than the single path video streaming. Finally, it presents some recent researches that based on hybrid solution of rateless FEC and advanced BEC techniques for enhancing the real-time and live video streaming quality over 3G, Wi-Fi and WiMAX wireless networks.

A Review : Comparative Study of Edge Detection Techniques

This article defines research about edge detection systems with their functions, pros and cons. For this purpose we will compare existing edge detection procedures lying under the 3 categories a) 1st order derivative/gradient methods (sobel, prewitt's, robert's operator) b) 2nd order derivatives / zerocrossing (laplacian of gaussian, difference of gaussian) and c) optimal methods (canny operator). Comparison is finished on the foundation of side detected pictures with their PSNR (peak signal to noise ration) and MSE (mean square error) values for exclusive strategies.

Hybrid image noise reduction algorithm based on genetic ant colony and PCNN

Pulse Coupled Neural Network (PCNN) has gained widespread attention as a nonlinear filtering technology in reducing the noise while keeping the details of images well, but how to determine the proper parameters for PCNN is a big challenge. In this paper, a method that can optimize the parameters of PCNN by combining the genetic algorithm (GA) and ant colony algorithm is proposed, which named as GACA, and the optimized procedure is named as GACA-PCNN. Firstly, the noisy image is filtered by median filter in the proposed GACA-PCNN method; then, the noisy image is filtered by GACA-PCNN constantly and the median filtering image is used as a reference image; finally, a set of parameters of PCNN can be automatically estimated by GACA, and the pretty effective denoising image will be obtained. Experimental results indicate that GACA-PCNN has a better performance on PSNR (peak signal noise rate) and a stronger capacity of preserving the details than previous denoising techniques.

Residual domain dictionary learning for compressed sensing video recovery

For compressed sensing (CS) recovery, the reconstruction quality is highly dependent on the sparsity level of the representation for the signal. Motivated by the observation that the temporal residual image is much sparser than its original image, a temporal residual-domain dictionary learning method for CS video recovery is proposed in this paper. The adaptive basis is learned from inter-frame differences by Karhunen---Loeve transform (KLT) to represent the residuals. And a block-based motion estimation/motion compensation (ME/MC) residual reconstruction strategy is incorporated for the CS video recovery. Experimental results on common test sequences at various sampling rates illustrate that the proposed algorithm gains great improvements over existing approaches. For some video sequences, the proposed method outperforms the state-of-art method near 1 dB in terms of peak signal noise rate (PSNR) at some higher sampling rate.

Multifocus watermarking approach based on discrete cosine transform.

Image fusion process consolidates data and information from various images of same sight into a solitary image. Each of the source images might speak to a fractional perspective of the scene, and contains both "pertinent" and "immaterial" information. In this study, a new image fusion method is proposed utilizing the Discrete Cosine Transform (DCT) to join the source image into a solitary minimized image containing more exact depiction of the sight than any of the individual source images. In addition, the fused image comes out with most ideal quality image without bending appearance or loss of data. DCT algorithm is considered efficient in image fusion. The proposed scheme is performed in five steps: (1) RGB colour image (input image) is split into three channels R, G, and B for source images. (2) DCT algorithm is applied to each channel (R, G, and B). (3) The variance values are computed for the corresponding 8 × 8 blocks of each channel. (4) Each block of R of source images is compared with each other based on the variance value and then the block with maximum variance value is selected to be the block in the new image. This process is repeated for all channels of source images. (5) Inverse discrete cosine transform is applied on each fused channel to convert coefficient values to pixel values, and then combined all the channels to generate the fused image. The proposed technique can potentially solve the problem of unwanted side effects such as blurring or blocking artifacts by reducing the quality of the subsequent image in image fusion process. The proposed approach is evaluated using three measurement units: the average of Q(abf), standard deviation, and peak Signal Noise Rate. The experimental results of this proposed technique have shown good results as compared with older techniques.

Bursty interference-oriented video quality assessment method

The frequent bursty interference leads to wireless throughput variability in future networks, which results in video quality of experience (QoE) degradation. It is highly desirable to be able to predict video quality to meet QoE requirements. There has been a great deal of studies on video quality assessment, but only limited work has been reported for assessing video quality under bursty interference environment. In this paper, we seek to ameliorate this by developing a bursty interference-oriented video quality assessment algorithm. First, a subjective experiment has been carried out and a hysteresis model was proposed by analyzing the experiment data. Simulation result shows that in burst traffic environment, the model has a better correlation with human visual system (HVS) effect. Then we proposed an objective quality assessment algorithm by taking the video color, brightness, motion and other spatial features together with Structural Similarity Index Measurement (SSIM) into consideration, which outperforms Peak Signal Noise Rate (PSNR), Visual Information Fidelity (VIF) and SSIM in bursty environment.

Modified Bit-Planes Sobel Operator: A New Approach to Edge Detection

detection of edges in images is a vital operation with applications in various fields. There are a number of methods developed already for the same. We have developed a for extraction of edges which is a modification of the existing Sobel operator. We have first extracted the bit planes of each image and have applied the Sobel operator on each bit plane for enhanced results. After this we have recreated the image by adding up the edges of all the bit-planes in their order of importance. This is a fairly simple global method which yields very good results The computations are simpler and faster as well. Pratts figure of merit (FOM) has been used to quantify the measure of edges. The values of Peak Signal to Noise Ration (PSNR) and Mean Square Error (MSE) have been calculated to assess the performance of the new algorithm in comparison to the previous existent one in presence of additive Gaussian noise. The results favor our new algorithm clearly.

Denoising Multi-Channel Images in Parallel MRI by Low Rank Matrix Decomposition

Parallel magnetic resonance imaging (pMRI) techniques can speed up MRI scan through a multi-channel coil array receiving signal simultaneously. Nevertheless, noise amplification and aliasing artifacts are serious in pMRI reconstructed images at high accelerations. This study presents a patch-wise denoising method for pMRI by exploiting the rank deficiency of multi-channel coil images and sparsity of artifacts. For each processed patch, similar patches are searched in spatial domain and throughout all coil elements, and arranged in appropriate matrix forms. Then, noise and aliasing artifacts are removed from the structured matrix by applying sparse and low rank matrix decomposition method. The proposed method has been validated using both phantom and in vivo brain data sets, producing encouraging results. Specifically, the method can effectively remove both noise and residual aliasing artifact from pMRI reconstructed noisy images, and produce higher peak signal noise rate (PSNR) and structural similarity index matrix (SSIM) than other state-of-the-art denoising methods.