Non-subsampled Shearlet Transform Research Articles

Local bit-plane neighbour dissimilarity pattern in non-subsampled shearlet transform domain for bio-medical image retrieval.

This paper introduces a novel descriptor non-subsampled shearlet transform (NSST) local bit-plane neighbour dissimilarity pattern (NSST-LBNDP) for biomedical image retrieval based on NSST, bit-plane slicing and local pattern based features. In NSST-LBNDP, the input image is first decomposed by NSST, followed by introduction of non-linearity on the NSST coefficients by computing local energy features. The local energy features are next normalized into 8-bit values. The multiscale NSST is used to provide translational invariance and has flexible directional sensitivity to catch more anisotropic information of an image. The normalised NSST subband features are next decomposed into bit-plane slices in order to capture very fine to coarse subband details. Then each bit-plane slices of all the subbands are encoded by exploiting the dissimilarity relationship between each neighbouring pixel and its adjacent neighbours. Experiments on two computed tomography (CT) and one magnetic resonance imaging (MRI) image datasets confirms the superior results of NSST-LBNDP when compared to many recent well known relevant descriptors both in terms of average retrieval precision (ARP) and average retrieval recall (ARR).

Intelligent Fusion of Infrared and Visible Image Data Based on Convolutional Sparse Representation and Improved Pulse-coupled Neural Network

Multi-source information can be obtained through the fusion of infrared images and visible light images, which have the characteristics of complementary information. However, the existing acquisition methods of fusion images have disadvantages such as blurred edges, low contrast, and loss of details. Based on convolution sparse representation and improved pulse-coupled neural network this paper proposes an image fusion algorithm that decompose the source images into high-frequency and low-frequency subbands by non-subsampled Shearlet Transform (NSST). Furthermore, the low-frequency subbands were fused by convolutional sparse representation (CSR), and the high-frequency subbands were fused by an improved pulse coupled neural network (IPCNN) algorithm, which can effectively solve the problem of difficulty in setting parameters of the traditional PCNN algorithm, improving the performance of sparse representation with details injection. The result reveals that the proposed method in this paper has more advantages than the existing mainstream fusion algorithms in terms of visual effects and objective indicators.

An Accurate and Lightweight Method for Human Body Image Super-Resolution.

In this paper, we propose a new method to super-resolve low resolution human body images by learning efficient multi-scale features and exploiting useful human body prior. Specifically, we propose a lightweight multi-scale block (LMSB) as basic module of a coherent framework, which contains an image reconstruction branch and a prior estimation branch. In the image reconstruction branch, the LMSB aggregates features of multiple receptive fields so as to gather rich context information for low-to-high resolution mapping. In the prior estimation branch, we adopt the human parsing maps and nonsubsampled shearlet transform (NSST) sub-bands to represent the human body prior, which is expected to enhance the details of reconstructed human body images. When evaluated on the newly collected HumanSR dataset, our method outperforms state-of-the-art image super-resolution methods with ∼8× fewer parameters; moreover, our method significantly improves the performance of human image analysis tasks (e.g. human parsing and pose estimation) for low-resolution inputs.

Hyperspectral Image Classification Based on Stacked Contractive Autoencoder Combined With Adaptive Spectral-Spatial Information

Hyperspectral image (HSI) contain abundant spectral and spatial information, enabling the accurate classification of ground objects. However, many existing machine learning methods have poor performance, and some existing CNN-based methods require high computational power, which considerably limits their real-world applications. To address these issues, in this paper, we propose an alternative HSI classification method based on the stacked contractive autoencoder (SCAE) and adaptive spectral-spatial information to improve the accuracy of HSI classification. Specifically, the non-subsampled shearlet transform (NSST) with the guided filtering (NG) enhances spatial structure information. Subsequently, we present an adaptive spatial information extraction method to extract the spatial information of pixels. Furthermore, we propose an HSI classification network, called SCAE-LR, for feature extraction and classification. The SCAE is implemented to extract the adaptive spectral-spatial feature, and a logistic regression (LR) layer is employed for classification. Extensive experiments on the Indian Pines data set and the Pavia University data set demonstrate the superior performance of our method.

Change Detection in SAR Images Based on Improved Non-Subsampled Shearlet Transform and Multi-Scale Feature Fusion CNN

Traditional methods for change detection in synthetic aperture radar (SAR) images have difficulty in obtaining results from the generated differential image (DI) owing to speckle noise. In recent years, many deep learning (DL)-based methods have emerged because of their outstanding anti-noise and self-learning ability. However, they are limited by the requirement of abundant high-precision labels. Therefore, in this paper, we propose a novel unsupervised method based on improved non-subsampled shearlet transform (NSST) and multi-scale feature fusion convolutional neural network (CNN) for change detection. First, this method improves the traditional NSST algorithm and proposes a novel pseudo-label generator to obtain more pseudo-labels with higher confidence. It is noteworthy that the more accurate the pseudo-labels are, the better the change detection results will be. Second, this method designs a multi-scale feature fusion block in the network to make the feature images contain more complete information and reduces the number of pooling layers to avoid losing feature image details. The main idea of this method is to eliminate the step of generating the DI and directly obtain results from the original images. The theoretical analysis and final results conducted on three real data sets prove its validity. Furthermore, to verify the generality and potential of the propose method, we apply it to the cross-region change detection and compare it with the supervised method, which achieve satisfactory results.

A Novel Comprehensive Watermarking Scheme for Color Images

Watermarking technology is commonly used to solve various problems in digital rights management and multimedia security. If a watermarking scheme with multiple purposes applies single method, it will easily cause the destruction of the hidden messages in particular attacks. For the copyright protection and tamper detection of color images, this research proposed a robust-fragile watermarking scheme. The two different embedding schemes embed the watermark into the R layer and G layer after NSST (nonsubsampled shearlet transform) and DWT (discrete wavelet transform) transformation. The hash sequence generated by the R layer and the G layer is served as fragile watermarks and is embedded into the B layer by the LSB (least significant bit) method. Finally, an improved rotation correction is applied to better extract the watermark under the rotation attack. Experimental results show that the proposed method is more accurate than the existing ones in terms of rotation angle correction and can effectively resist general attacks such as noise, filtering, and JEPG compression. Moreover, the proposed fragile watermark can locate the tamper position when malicious tamper occurs. Except cropping attack, the true-positive rate (TPR) reaches 1 for all attacks.

A multisource image fusion method for multimodal pig-body feature detection

The multisource image fusion has become an active topic in the last few years owing to its higher segmentation rate. To enhance the accuracy of multimodal pig-body feature segmentation, a multisource image fusion method was employed. Nevertheless, the conventional multisource image fusion methods can not extract superior contrast and abundant details of fused image. To superior segment shape feature and detect temperature feature, a new multisource image fusion method was presented and entitled as NSST-GF-IPCNN. Firstly, the multisource images were resolved into a range of multiscale and multidirectional subbands by Nonsubsampled Shearlet Transform (NSST). Then, to superior describe fine-scale texture and edge information, even-symmetrical Gabor filter and Improved Pulse Coupled Neural Network (IPCNN) were used to fuse low and high-frequency subbands, respectively. Next, the fused coefficients were reconstructed into a fusion image using inverse NSST. Finally, the shape feature was extracted using automatic threshold algorithm and optimized using morphological operation. Nevertheless, the highest temperature of pig-body was gained in view of segmentation results. Experiments revealed that the presented fusion algorithm was able to realize 2.102-4.066% higher average accuracy rate than the traditional algorithms and also enhanced efficiency.

Feature Extraction Based on Non-Subsampled Shearlet Transform (NSST) with Application to SAR Image Data

Considering the defaults in synthetic aperture radar (SAR) image feature extraction, an SAR target recognition method based on non-subsampled Shearlet transform (NSST) was proposed with application to target recognition. NSST was used to decompose an SAR image into multilevel representations. These representations were translation-invariant, and they could well reflect the dominant and detailed properties of the target. During the machine learning classification stage, the joint sparse representation was employed to jointly represent the multilevel representations. The joint sparse representation could represent individual components independently while considering the inner correlations between different components. Therefore, the precision of joint representation could be enhanced. Finally, the target label of the test sample was determined according to the overall reconstruction error. Experiments were conducted on the MSTAR dataset to examine the proposed method, and the results confirmed its validity and robustness under the standard operating condition, configuration variance, depression angle variance, and noise corruption.

Multistage multimodal medical image fusion model using feature‐adaptive pulse coupled neural network

AbstractMedical image fusion focuses to fuse complementary diagnostic details for better visualization of comprehensive information and interpretation of various diseases and its treatment planning. In this paper, a multistage multimodal fusion model is presented based on nonsubsampled shearlet transform (NSST), stationary wavelet transform (SWT), and feature‐adaptive pulse coupled neural network. Firstly, NSST is employed to decompose the source images into optimally sparse multi‐resolution components followed by SWT. Secondly, structural features are extracted by a weighted sum‐modified Laplacian and applied to an adaptive model to map feature weights for low‐band SWT component fusion, and texture feature‐based fusion rule is applied to fuse high‐band SWT components. High‐frequency NSST components are fused using the absolute maximum and sum of absolute difference based rule to retain complex directional details. Experimental results show that the proposed method obtains significantly better fused medical images compared to others with excellent visual quality and improved computational measures.

Defogging Technology Based on Dual-Channel Sensor Information Fusion of Near-Infrared and Visible Light

Since the method to remove fog from images is complicated and detail loss and color distortion could occur to the defogged images, a defogging method based on near-infrared and visible image fusion is put forward in this paper. The algorithm in this paper uses the near-infrared image with rich details as a new data source and adopts the image fusion method to obtain a defog image with rich details and high color recovery. First, the colorful visible image is converted into HSI color space to obtain an intensity channel image, color channel image, and saturation channel image. The intensity channel image is fused with a near-infrared image and defogged, and then it is decomposed by Nonsubsampled Shearlet Transform. The obtained high-frequency coefficient is filtered by preserving the edge with a double exponential edge smoothing filter, while low-frequency antisharpening masking treatment is conducted on the low-frequency coefficient. The new intensity channel image could be obtained based on the fusion rule and by reciprocal transformation. Then, in color treatment of the visible image, the degradation model of the saturation image is established, which estimates the parameters based on the principle of dark primary color to obtain the estimated saturation image. Finally, the new intensity channel image, the estimated saturation image, and the primary color image are reflected to RGB space to obtain the fusion image, which is enhanced by color and sharpness correction. In order to prove the effectiveness of the algorithm, the dense fog image and the thin fog image are compared with the popular single image defogging and multiple image defogging algorithms and the visible light-near infrared fusion defogging algorithm based on deep learning. The experimental results show that the proposed algorithm is better in improving the edge contrast and the visual sharpness of the image than the existing high-efficiency defogging method.

An Optimal Registration on Shearlet domain with Novel Weighted Energy fusion for Multi-Modal Medical Images

Medical images usually display various attributes of data on human viscera and abnormal tissue in different modalities. The fusion of images, ensures effective deployment of all relevant information from several modalities into a single image. Finally, the paper leads to a novel multi-modal medical image fusion method based on Non-Subsampled Shearlet Transform (NSST) called Denoised Optimum B-Spline Shearlet Image Fusion(DOBSIF) that is based on real-time and standard radiological datasets. The proposed, novel registration based fusion technique DOBSIF, is compared with existing state-of-art techniques like PCA, DWT, SWT etc., and in comparison the proposed DOBSIF works effectively for both the color and grayscale images. To improve the fusion function, a novel pre-fusion method is carried out with the help of the Whale Optimization Algorithm(WOA), using the ideal B-spline-based registration method and in addition a novel Weighted Energy fusion rule is applied to derive relevant information from the source images. The proposed work initially follows normal pre-processing measures such as Gaussian filtering, edge sharpening, and resizing followed by optimal registration, fusion with novel fusion rule and segmentation of tumor part from the fused image. The visual quality of the fused image was evaluated by expert radiologists and the fused resultant image is robust, subject to the modified Active Contour(ACM) model for lesion identification.

Wood Defect Detection Based on Depth Extreme Learning Machine

The deep learning feature extraction method and extreme learning machine (ELM) classification method are combined to establish a depth extreme learning machine model for wood image defect detection. The convolution neural network (CNN) algorithm alone tends to provide inaccurate defect locations, incomplete defect contour and boundary information, and inaccurate recognition of defect types. The nonsubsampled shearlet transform (NSST) is used here to preprocess the wood images, which reduces the complexity and computation of the image processing. CNN is then applied to manage the deep algorithm design of the wood images. The simple linear iterative clustering algorithm is used to improve the initial model; the obtained image features are used as ELM classification inputs. ELM has faster training speed and stronger generalization ability than other similar neural networks, but the random selection of input weights and thresholds degrades the classification accuracy. A genetic algorithm is used here to optimize the initial parameters of the ELM to stabilize the network classification performance. The depth extreme learning machine can extract high-level abstract information from the data, does not require iterative adjustment of the network weights, has high calculation efficiency, and allows CNN to effectively extract the wood defect contour. The distributed input data feature is automatically expressed in layer form by deep learning pre-training. The wood defect recognition accuracy reached 96.72% in a test time of only 187 ms.

A Fusion Method for Atomic Force Acoustic Microscopy Cell Imaging Based on Local Variance in Non-Subsampled Shearlet Transform Domain

Atomic force acoustic microscopy (AFAM) is a measurement method that uses the probe and acoustic wave to image the surface and internal structures of different materials. For cellular material, the morphology and phase images of AFAM reflect the outer surface and internal structures of the cell, respectively. This paper proposes an AFAM cell image fusion method in the Non-Subsampled Shearlet Transform (NSST) domain, based on local variance. First, NSST is used to decompose the source images into low-frequency and high-frequency sub-bands. Then, the low-frequency sub-band is fused by the weight of local variance, while a contrast limited adaptive histogram equalization is used to improve the source image contrast to better express the details in the fused image. The high-frequency sub-bands are fused using the maximum rule. Since the AFAM image background contains a lot of noise, and improved segmentation algorithm based on the Otsu algorithm is proposed to segment the cell region, and the image quality metrics based on the segmented region will make the evaluation more accurate. Experiments with different groups of AFAM cell images demonstrated that the proposed method can clearly show the internal structures and the contours of the cells, compared with traditional methods.

Non Subsampled Shearlet Transform Based Fusion of Multiple Exposure Images

Fusion of multiple exposure images has attracted attention over the past decade and several algorithms have been developed, so as to capture the entire dynamic range of the scene in a single image. Capturing images with changes in exposure settings leads to a set of multiple exposure images with different areas of the scene highlighted in different image. Weak edges and fine textures of the image are lost during an under or over exposure. Also for objective evaluation we need to measure and both the structural and textural information in the images simultaneously. To address this issue an algorithm based on the Non-subsampled shearlet transform (NSST) for fusing multiple exposure images is proposed so as to depict clearly the dimly lit, brightly lit and well lit regions in a single fused image. In the proposed algorithm NSST decomposition is first performed on the images to obtain the multi-scale and multi-direction representations. The high frequency bands are fused by retaining the pixels with the highest value coefficients at each sub band at each level. Whereas the low frequency bands are fused by averaging operation. Proposed method leads to better results in visual quality.

A novel statistical approach for multiplicative speckle removal using t-locations scale and non-sub sampled shearlet transform

One of the most interesting problems in denoising of images includes despeckling of multiplicative noise. This paper proposes a novel statistical processor in the framework of Non-sub Sampled Shearlet Transform (NSST) to reduce the effect of the multiplicative noise on images given preserving of structural and visual quality of image. First, we indicate that NSST coefficients of logarithmically transformed images can be statistically meaningful modeled by t-location scale (TLS). For designing our processor, we employ Minimum Mean Squared Error (MMSE) estimator to reduce noise distortion. We show by using TLS as the prior distribution, non-linear noise suppression behavior is obtained in test images. Finally, we compare our method by the state-of-the-art algorithms like soft and hard thresholding and also with well-known adaptive filters in this area like Wiener, Frost, Lee and one recent method in shearlet denoising framework.

Multi-Focus Image Fusion Based on Non-Subsampled Shearlet Transform and Spiking Cortical Model

Multi-Focus Image Fusion Based on Non-Subsampled Shearlet Transform and Spiking Cortical Model

Synthetic-Aperture Radar Image Despeckling based on Improved Non-Local Means and Non-Subsampled Shearlet Transform

When Synthetic-Aperture (SAR) image is transformed into wavelet domain and other transform domains, most of the coefficients of the image are small or zero. This shows that SAR image is sparse. However, speckle can be seen in SAR images. The non-local means is a despeckling algorithm, but it cannot overcome the speckle in homogeneous regions and it blurs edge details of the image. In order to solve these problems, an improved non-local means is suggested. At the same time, in order to better suppress the speckle effectively in edge regions, the non-subsampled Shearlet transform (NSST) is applied. By combining NSST with the improved non-local means, a new type of despeckling algorithm is proposed. Results show that the proposed algorithm leads to a satisfying performance for SAR images.

Multimodal image fusion based on global-regional-local rule in NSST domain

An accurate representation of the source image is vital in image fusion processing. However, among most of the traditional multi-scale decomposition based image fusion methods, they fail to accurately interpret the source images, only capturing local information without considering the regional and global information. To overcome this drawback, a novel image fusion method based on the global-regional-local rule is proposed in the nonsubsampled shearlet transform (NSST) domain. First, the source paired images are decomposed into the low-pass subbands and high-pass subbands with the NSST transformation. Second, for comprehensively representing the statistical correlation of source image, three-level statistical models, that is, global contextual hidden Markov model (G-CHMM), regional contextual hidden Markov model (R-CHMM), and local contextual hidden Markov model (L-CHMM) are established for the high-pass subbands. In the R-CHMM, a novel feature vector is extracted for getting the region map of source image by the fuzzy cluster method (FCM). Third, to get accurate active measures of the source image, a global-regional-local fusion rule based on the statistical characteristics extracted from global-regional-local CHMM is designed, which is used to fuse the high-pass subbands. The low-pass subbands are fused with the choose-max rule based on the local gradient measure. Finally, the fused image is obtained by the inverse NSST. Experimental results on serials of infrared and visible images, medical images and multi-focus images demonstrate the advantage of the proposed method in terms of detail preserving.

Locally optimum watermark decoder in NSST domain using RSS-based Cauchy distribution

It is important to consider the tradeoff between imperceptibility and robustness requirements in developing an image-watermarking technique, some statistical model-based image watermarking schemes have been designed in the past decade. The effectiveness of a statistical watermark decoder depends highly on the modeling of the transform-domain coefficients and the use of hypothesis testing. In this study, a multiplicative image watermarking scheme is proposed in the nonsubsampled Shearlet transform (NSST) domain, where NSST coefficients are modeled by ranked set sample (RSS) based Cauchy statistical distribution and locally most powerful (LMP) test criterion is applied. Digital image watermarking technique consists of two parts, namely, embedding and extracting. In the embedding process, to achieve relatively good imperceptibility and robustness, watermark data is inserted into the significant NSST directional subband that has the highest energy value, by modifying nonlinearly the significant NSST coefficients. In the extracting phase, a scheme is proposed for designing a blind NSST domain watermark decoder incorporating the RSS based Cauchy statistical distribution and LMP test criterion. Here, the RSS method is used to estimate the location parameter and shape parameter of Cauchy statistical distribution instead of traditional maximum-likelihood (ML) method, which can provide the Cauchy model with higher precision parameters. Experimental results on a set of standard test images show significant improvements in imperceptibility and robustness using the proposed method compared with the best known state-of-the-art approaches.

Infrared and visible image fusion based on nonlinear enhancement and NSST decomposition

In multi-scale geometric analysis (MGA)-based fusion methods for infrared and visible images, adopting the same representation for the two types of images will result in the non-obvious thermal radiation target in the fused image, which can hardly be distinguished from the background. To solve the problem, a novel fusion algorithm based on nonlinear enhancement and non-subsampled shearlet transform (NSST) decomposition is proposed. Firstly, NSST is used to decompose the two source images into low- and high-frequency sub-bands. Then, the wavelet transform (WT) is used to decompose high-frequency sub-bands to obtain approximate sub-bands and directional detail sub-bands. The “average” fusion rule is performed for fusion for approximate sub-bands. And the “max-absolute” fusion rule is performed for fusion for directional detail sub-bands. The inverse WT is used to reconstruct the high-frequency sub-bands. To highlight the thermal radiation target, we construct a non-linear transform function to determine the fusion weight of low-frequency sub-bands, and whose parameters can be further adjusted to meet different fusion requirements. Finally, the inverse NSST is used to reconstruct the fused image. The experimental results show that the proposed method can simultaneously enhance the thermal target in infrared images and preserve the texture details in visible images, and which is competitive with or even superior to the state-of-the-art fusion methods in terms of both visual and quantitative evaluations.