Shearlet Transform Research Articles

Optimal Deep Neural Network model based multimodality fused medical image classification

Medical image fusion and classification has been utilized to get valuable significant data multimodality medical image data. The fundamental commitment of this paper is to extricate the compelling features from the fused medical image that is Multimodality Medical images (MMI) for the classification model. At first, MMI are considered for the fusion procedure by shearlet Transform and optimization model to get fused images. The medical images of fusion has demonstrated to be valuable for propelling the clinical dependability of utilizing medical imaging for medical diagnostics and investigation, this procedure is considered as the principal phase. In send phase, few features are extracted from the Fused multimodality Medical Image (FMMI). For example, texture and XOR pattern models for classification, the point of recognition as given fused image as begin or malignant by Deep Neural Network (DNN) with Optimization. The significant purpose of optimization is to improve the classification accuracy. To optimize the DNN structure, we proposed the Discrete Gravitational Search Algorithm (DGSA), and it depends on increasing speed and force of DNN weights. When the optimal structure finds, the test image comprises the input of the structure for classification. The experimental results stated that the proposed model shows effective performance with the fusion factor of 6.52 and spatial frequency of 26.8.

Edge preserved multispectral image compression using PCA and hybrid transform

In multispectral image compression, it is difficult to obtain sparse approximation for images with rich details. The existing methods produce better results under some constraints on image content. In order to obtain robust multispectral image compression, the Principal Component Analysis (PCA) method is used to reduce the spectral redundancy and the sparse approximation is obtained by using Extended Shearlet Transform (EST) and Tetrolet Transform(TT). The anisotropic property of EST is used to preserve smooth images with global structures of an image whereas the TT is used to preserve the local structures. The performance of proposed method is compared with the existing methods in terms of rate distortion and information preservation perspectives. The Compression Ratio (CR) and Peak Signal to Noise Ratio (PSNR) are used as rate-distortion measure. The Mean Structural Similarity Index Metric (MSSIM) and Kappa coefficient (K) are information preservation measures. The simulation results show that the proposed EPMI-HT method outperforms the existing hybrid methods for all kinds of image content at high CR with retaining edge information.

Image Fusion Based on Nonsubsampled Shearlet Transform

In current days, an image fusion is a powerful method and developing field in the area of image processing. The image fusion is the process of combining two or more images into a single image then the resulting image will appear more informative than any of the input images. It is the process of assimilation of numerous input images into a new single fused image with highly informative than the input image. There are various image fusion transform techniques are proposed. Out of that techniques a Non-subsampled shearlet transform includes shift invariant property, highly directionality, feasible and more efficient information as compared to previous techniques such as wavelet transform(WT), DWT, LWT, MWT, CWT, Curvelet transform, Counterlet transform, and Nonsubsampled Counterlet transform(NSCT). This NSST technique is carried out by adjusting the levels with filter banks. Downsampling is used to reconstruct. NSST decomposition provides a simple hierarchical framework for image fusion with different geographical resolution.

Automatic Grading System for Diagnosis of Breast Cancer Exploiting Co-occurrence Shearlet Transform and Histogram Features

ObjectivesBreast cancer (BC) is one of the most commonly reported health issues worldwide, especially in females. Early detection and diagnosis of BC can greatly reduce mortality rates. Samples obtained with different imaging methods such as mammography, computerized tomography, magnetic resonance, ultrasound, and biopsy are used in the diagnosis of BC. Histopathological images obtained from a biopsy contain vital information about the stage of the BC. Computer-aided systems are important tools to assist pathologists in the early detection of BC. Material and methodsIn the current study, the use of gray-level co-occurrence matrix (GLCM) of Shearlet Transform (ST) coefficients were first scrutinized as textural features. ST is an advanced decomposition-based method that can analyze images in various directions and is sensitive to edge singularities. These features make ST more robust than other decomposition methods such as Fourier and wavelet. Color channel histogram features were also utilized for a second level of evaluation in the diagnosis of the BC stage. These features are considered one of the most important building blocks that pathologists consider in the course of grading histopathological images. Then, by combining these two features, the classification results were re-assessed utilizing Support Vector Machine (SVM) as a classifier. ResultsThe assessments were performed on a BreaKHis dataset containing benign and malignant histopathological samples. The average accuracy scores were reported as being 98.2%, 97.2%, 97.8%, and 97.3% in the sub-databases with 40×, 100×, 200×, and 400× magnification factors, respectively. ConclusionsThe obtained results showed that the proposed method was quite efficient in histopathological image classification. Despite the relative simplicity of the approach, the obtained results were far superior to previously reported results.

Wall Thickness Measurement of Pipes Using Digital Radiography and Three Wavelet Methods

Digital X-ray radiography is a well established method for non-destructive detection of corrosion and measurement of wall thickness of pipes. Due mainly to the unavoidable detection of scattered X-rays as well as the ever-present system electronic noise, the quality of the acquired radiographs is often low; the quality can however be enhanced by applying various image processing methods. In this research, Wavelet Transform (WT), Gabor Filters (GF) and Shearlet Transform (ST) have been utilized in a background removing algorithm to achieve the required radiographic image quality improvement. The algorithm first decomposes the image into different levels using the WT, GF or the ST method, and the automatic thresholding level, which is based on Rayleigh distribution, is then applied. The output images (with the background removed) were then reconstructed by subtracting the modified component from the original radiograph. The results of the study showed that the reconstructed images yield significantly greater contrast than the original radiograph, the wall thickness measurements have higher precision and defect detection has greater specificity for defect in all regions. Also, the outcomes show that the algorithm has high analytical efficiency and is capable of accurately measuring pipe wall thicknesses and corrosion regions.

The Subsurface Structure on the CE-3 Landing Site: LPR CH-1 Data Processing by Shearlet Transform

The Lunar penetrating radar (LPR) carried by Chang’E-3 (CE-3) mission is an important application of radar in lunar exploration. An opportunity of significance to detect the information of regolith and the subsurface structure on the landing site is offered by LPR aboard the Yutu rover. On the basis of a data processing flow, a low-frequency radar image has been available for mapping subsurface structure. The noise interfered data give a huge challenge for geological stratification and interpretation. To solve the limitation imposed by noise, we adopt the shearlet transform as a promising tool for data analysis and noise attenuation. The different distributions of the noise and signal in the shearlet domain decrease the difficulty of noise attenuation. To optimize the denosing strategy, we replace a local adaptive thresholding function with the conventional hard threshold. The quality of the processed data is improved, which is helpful for geological stratification and interpretation. Finally, by combining these data with the regional geology and previous research, especially the LPR data, we can provide an interpretation of the LPR CH-1.

Self-Supervised Light Field Reconstruction Using Shearlet Transform and Cycle Consistency

The image-based rendering approach using Shearlet Transform (ST) is one of the state-of-the-art Densely-Sampled Light Field (DSLF) reconstruction methods. It reconstructs Epipolar-Plane Images (EPIs) in image domain via an iterative regularization algorithm restoring their coefficients in shearlet domain. Consequently, the ST method tends to be slow because of the time spent on domain transformations for dozens of iterations. To overcome this limitation, this letter proposes a novel self-supervised DSLF reconstruction method, CycleST, which applies ST and cycle consistency to DSLF reconstruction. Specifically, CycleST is composed of an encoder-decoder network and a residual learning strategy that restore the shearlet coefficients of densely-sampled EPIs using EPI reconstruction and cycle consistency losses. Besides, CycleST is a self-supervised approach that can be trained solely on Sparsely-Sampled Light Fields (SSLFs) with small disparity ranges ($\leqslant$ 8 pixels). Experimental results of DSLF reconstruction on SSLFs with large disparity ranges (16 - 32 pixels) from two challenging real-world light field datasets demonstrate the effectiveness and efficiency of the proposed CycleST method. Furthermore, CycleST achieves ~ 9x speedup over ST, at least.

Identification of Colon Cancer Using Multi-Scale Feature Fusion Convolutional Neural Network Based on Shearlet Transform

Colon cancer identification is of great significance in medical diagnosis. Real-time, objective and accurate inspection results will facilitate medical professionals to explore symptomatic treatment promptly. However, the existing methods depend on hand-crafted features which require extensive professional expertise and long inspection period. Therefore, we propose a multi-scale feature fusion convolutional neural network (MFF-CNN) based on shearlet transform to identify histopathological image of colon cancer. The characteristic of the framework is the shearlet coefficients of histopathological image in multiple decomposition scales were extracted as supplementary feature which were also fed to the network with the original pathological image. After feature learning and feature fusion, the MFF-CNN based on shearlet transform can achieve the identification accuracy of 96% and average F-1 score of 0.9594 for colorectal adenocarcinoma epithelium (TUM) and normal colon mucosa (NORM). The false negative rate and false positive rate can be reduced to 5.5% and 2.5%, respectively. The superior performance of the network opens a new perspectives for real-time, objective and accurate diagnosis of cancer.

An Infrared and Visible Image Fusion Algorithm Based on LSWT-NSST

Regarding the problems of image distortion, edge blurring, Gibbs phenomena in the traditional wavelet transform algorithm and the loss of subtle features in the Non-Subsampled Shearlet Transform (NSST), and considering the physical characteristics of infrared and visible images, an infrared and visible image fusion algorithm based on the Lifting Stationary Wavelet Transform (LSWT) and Non-Subsampled Shearlet Transform is proposed in this paper. First, since LSWT can quickly calculate and has all advantages of traditional WT, it is utilized to decompose infrared and visible images to obtain low-frequency coefficients and multi-scale and multi-directional high-frequency coefficients, respectively. Second, NSST multi-scale decomposition is used to extract the target features and detailed features of the image from the high and low-frequency sub-bands to obtain new high and low-frequency sub-bands. Third, according to the physical characteristics that low and high-frequency coefficients represent, different fusion rules are designed. Discrete Cosine Transform (DCT) and Local Spatial Frequency (LSF) are introduced in the low-frequency sub-band, and LSF adaptive weighted fusion rules are used in the DCT domain. The fusion strategy improves the regional contrast in the high-frequency sub-band with the spectral characteristics of human vision. Finally, the Inverse Lifting Stationary Wavelet Transform (ILSWT) is used to reconstruct the fusion coefficients to obtain the final fused images. To verify the advantages of the proposed algorithm in this paper, the classic and advanced 9 IR and VI fusion algorithms are selected for subjective and objective comparison. In the objective evaluation, a comprehensive ranking index is designed based on 9 classical indicators. Simulation experiments with 10 IR and VI fusion algorithms prove that the proposed algorithm has better performance and flexibility. The results show that the proposed algorithm in this paper fuses the images with clear edges, prominent targets, and good visual perception, and it outperforms state-of-the-art image fusion algorithms.

Composite material terahertz image fusion based on PCNN and RGEDIM under non-subsampled shearlet transform

In order to solve the problem that defects of different scales have different terahertz imaging characteristics in fiber reinforced composites, the fusion processing method of two terahertz images with complementary defect information was studied. To reduce the Gibbs phenomenon, Non-subsampled Shearlet Transform (NSST) with the property of shift- invariance was used to decompose source images and get their low-frequency subband and high frequency subband coefficients. Regional variance was used as connection strength factor of the Pulse Coupled Neural Network (PCNN) in the low frequency coefficient fusion, which is more according with human visual characteristics. In the fusion of high frequency coefficients, the Regional Gradient Energy of Direction Information Measure (RGEDIM) was introduced to extract the edge, texture and other details of the image and integrated them into the final image, the impact of noise on image fusion was reduced better. Finally, the fusion image was obtained through NSST inverse transform. The experimental results show that this method is superior to wavelet, Non-subsampled Contourlet Transform (NSCT) and traditional PCNN method, the fusion image has more mutual information and contains more original image information, all the defects of the source image can be clearly seen on the fusion image.

Denoising of Uncertain Type Noise Images by Spatial Feature Classification in Nonsubsampled Shearlet Transform

Most denoising methods are designed to deal standard images with specific type noise, which do not perform well when denoising real noisy images contain uncertain types of noise. However, underwater image is a typical uncertain type noise image. To solve this problem, this paper presents a method using spatial feature classification jointing nonsubsampled shearlet transform (NSST) for denoising uncertain type noise images. Justifiable granule is employed to solve the problem of parameter selection. The raw image was decomposed by using the NSST to get one low frequency subband and several high frequency subbands. Then, the preliminary binary map is built, the binary map is employed to decide whether a coefficient contains spatial feature or not. And we employ justifiable granule to solve the difficulty of parameter selection. The high subbands coefficients are classified into two classes by fuzzy support vector machine classification: the texture class and the noise class. At last, the adaptive Bayesian threshold is used to shrink the coefficients. Simulation results show the proposed method is effective in uncertain type noise images(also have good performance in specific type noise). The method we proposed has been compared with other popular denoising methods and get excellent subjective performance and PSNR improvement.

Low-Light Image Enhancement Based on Nonsubsampled Shearlet Transform

To improve the observability of low-light images, a low-light image enhancement algorithm based on nonsubsampled shearlet transform (NSST) is presented (LIEST). The proposed algorithm can synchronously achieve contrast improvement, noise suppression, and the enhancement of specific directional details. An enhancement framework of low-light noisy images is first derived, and then, according to the framework, a low-light noisy image is decomposed into low-pass subband coefficients and bandpass direction subband coefficients by NSST. Then, in the NSST domain, an illumination map is estimated based on a bright channel of the low-pass subband coefficients, and noise is simultaneously suppressed by shrinking the bandpass direction subband coefficients. Finally, based on the estimated illumination map, the low-pass subband coefficients, and the shrunken bandpass direction subband coefficients, inverse NSST is implemented to achieve low-light image enhancement. Experiments demonstrate that the LIEST exhibits superior performance in improving contrast, suppressing noise, and highlighting specific details as compared to seven similar algorithms.

Classification of Brain MRI for Alzheimer’s Disease Detection Based on Ensemble Machine Learning

Background: Machine learning is now a powerful tool to help improve medical disorders diagnosis. One of its critical applications is the classification or clustering of neurodegenerative disease by pattern recognition methods based on biomedical signals and medical images. Early detection of these diseases is always useful and vital. In this study, we focused on Alzheimer’s disease (AD) as a type of dementia leading to problems with memory, thinking, and behavior. This disease was named after Dr. Alois Alzheimer in 1906 when he inspected a female patient who died of an unusual mental illness. According to recent studies, four stages are introduced for AD, including pre-dementia, early AD, moderate AD, and advanced AD. There are several methods for AD diagnosis that include mental status evaluation, physical exam, and neurological exam, based on different imaging techniques such as magnetic resonance imaging (MRI). Several methods have been introduced until now for the classification and detection of AD using machine learning algorithms, such as the classification of AD with discrete wavelet transform (DWT) and single linear discriminant analysis (LDA) classifiers and differentiation of AD from normal based on T2-weighted MRI with shearlet transform (ST) and K-nearest neighbors (KNN) classifiers. Objectives: In this work, we proposed a methodology based on DWT with three-level decomposition feature extraction (Figure 1). Methods: Based on statistics (mean, variance, skewness) of features and principal component analysis (PCA) for dimension reduction, we used five classifiers, including multi-layer perceptron neural network (MLPNN) (Figure 2), KNN, support vector machine (SVM) (Figure 3), and naive Bayesian (NB) with the majority vote method to fuse them into one ensemble classifier. Results: The proposed methodology was evaluated using 100 T2-weighted MRI of AD and cognitive normal (CN) subjects, which were chosen from the Harvard Medical School website. The accuracy, specificity, and sensitivity achieved from our methodology were 95%, 90%, and 100%, respectively by using a 10-fold cross-validation strategy. Conclusion: Our study showed that the stacking method for classification of AD and CN was better than using one classifier and comparable with state-of-the-art methods.

A Robust Pan-Sharpening Scheme for Improving Resolution of Satellite Images in the Domain of the Nonsubsampled Shearlet Transform

Pan-sharpening is a substantial post-processing task for captured multispectral remotely sensed satellite imagery. Its main purpose is to fuse the high spectral characteristics of the multispectral (MS) images with the high spatial information of the panchromatic (Pan) image to output a sharper MS image (pan-sharpened) that encompasses higher spectral and spatial resolutions. In this paper, we investigate a conception of a new pan-sharpening scheme using the pulse coupled neural network (PCNN) in the nonsubsampled shearlet transform (NSST) domain. This can be done based on two main steps. In the first step, the input MS and Pan images are individually decomposed into multi-scaled and multi-directional coefficients by NSST. Second, the PCNN is applied to the low-frequency coefficients, which are merged by a weighted firing energy fusion rule utilizing the PCNN firing times. The detail coefficients with higher matching value are chosen to be the fused detail coefficients. Lastly, the pan-sharpened image is generated by the inverse NSST. WorldView-2, GeoEye-1, and QuickBird satellite datasets are employed in the experiments which demonstrate that the investigated scheme gained the ability in preserving both the high spatial details and high spectral characteristics simultaneously without involving abundant computation time. In addition, various image quality metrics such as CC, RMSE, RASE, ERGAS, SAM, Q4, QNR, and SCC are adopted to assess the spectral and spatial qualities of the pan-sharpened image. The experimental results and performance analysis illustrated that our scheme improved performance efficiency and achieved superiority over other conventional techniques.

A Novel Medical Image Fusion Approach Based on Nonsubsampled Shearlet Transform

A Novel Medical Image Fusion Approach Based on Nonsubsampled Shearlet Transform

Seismic Random Noise Reduction Using Adaptive Threshold Combined Scale and Directional Characteristics of Shearlet Transform

Random noise attenuation is an important step in seismic processing. A method with an adaptive threshold based on scale and directional characteristics of shearlet transform is proposed to attenuate seismic random noise. The method can adaptively calculate threshold according to the decomposed scales and directions, where the directional characteristics are embodied by using the L1 norm of coefficients. Compared to the universal threshold and the threshold based on scale methods only, the scale and directional threshold methods can remove random noise and reserve an effective signal to the utmost extent. The synthetic and field data examples demonstrate that the proposed method can significantly improve the denoising effect and has the stronger denoising ability.

A Research for Implementing Image Interpolation using Inpainting and Shearlet Transform

This paper proposes a way to improve the compression ratio of images by expunging some parts of the image prior transmission. The remaining data besides essential details for recovering the removed regions are encoded to produce the final data. At the decoding side an inpainting method is applied to retrieve the removed region. The Shearlet Transform is used for the smoothing purpose of the recovered image. This transform can identify the location of singularities of a function and also the orientation of discontinuity curves. The Shearlet Transform has the ability to provide a very accurate geometrical characterization of general discontinuity occurring in images.

Extracting Reflected Waves From Acoustic Logging Data Based on the Shearlet Transform

Acoustic reflection imaging in well logging is an important technology for surveying near-borehole geological structures, especially for exploration in heterogeneous reservoirs. The key component of this method is the extraction of reflected waves from full waveforms. Due to the high impedance of the borehole wall interface, it is quite difficult to separate reflections from the dominant mode waves. This letter introduces a method for identifying reflected waves based on the shearlet transform. This method, which is characterized by multiscale and multidirectional decomposition capabilities, can be used to identify different components of acoustic logging signals. The effectiveness and accuracy of the proposed method are verified in comparison with other methods with synthetic data. Finally, field data results further validate this method.

Feature Extraction with Discrete Non-Separable Shearlet Transform and Its Application to Surface Inspection of Continuous Casting Slabs

A new feature extraction technique called DNST-GLCM-KSR (discrete non-separable shearlet transform-gray-level co-occurrence matrix-kernel spectral regression) is presented according to the direction and texture information of surface defects of continuous casting slabs with complex backgrounds. The discrete non-separable shearlet transform (DNST) is a new multi-scale geometric analysis method that provides excellent localization properties and directional selectivity. The gray-level co-occurrence matrix (GLCM) is a texture feature extraction technology. We combine DNST features with GLCM features to characterize defects of the continuous casting slabs. Since the combination feature is high-dimensional and redundant, kernel spectral regression (KSR) algorithm was used to remove redundancy. The low-dimension features obtained and labels data were inputted to a support vector machine (SVM) for classification. The samples collected from the continuous casting slab industrial production line—including cracks, scales, lighting variation, and slag marks—and the proposed scheme were tested. The test results show that the scheme can improve the classification accuracy to 96.37%, which provides a new approach for surface defect recognition of continuous casting slabs.

Application of discrete transforms with selective coefficients for blind image watermarking

AbstractWatermarking scheme is not helpful in conveying the original image continually so as identifying the owner's mark from the watermarked image. In the current situation, our chosen topic is significant in real‐time applications like fraud recognition, emergency clinic, and government divisions. In this study, an optimal multiblind watermarking model is proposed for the watermark detection process. Our proposed model is a combination of intelligent domain transforms like Discrete Shearlet Transform and Discrete Curvelet Transform (DCurT). The imperceptibility necessity of the plan is accomplished utilizing optimal coefficients that are performed by applying in DCurT with metaheuristic optimization model that is Grasshopper Algorithm. It is played by a chasing behavior of gathering of grasshoppers and cooperation process, here, Random Grasshopper Optimization is utilized for watermarking. The secret image is embedded with a lower band of optimal coefficients with DCurT, and here, the band is DCur (1, 5). The secret data is embedded in the host images to make it secure, and then, the extraction of the embedding process takes place inversely. For experimentation analysis, 20 digital images are considered, and different attacks are applied in the proposed watermarking model. Thus, the watermarked image looks lossless compared with the host image. Moreover, recent literary works and domain transform are also utilized for strength investigation of the proposed watermarking model.