Recently,ground-penetrating radar(GPR)has been extended as a well-known area to investigate the subsurface objects.However,its output has a low resolution,and it needs more processing for more interpretation.This pape...Recently,ground-penetrating radar(GPR)has been extended as a well-known area to investigate the subsurface objects.However,its output has a low resolution,and it needs more processing for more interpretation.This paper presents two algorithms for landmine detection from GPR images.The first algorithm depends on a multi-scale technique.A Gaussian kernel with a particular scale is convolved with the image,and after that,two gradients are estimated;horizontal and vertical gradients.Then,histogram and cumulative histogram are estimated for the overall gradient image.The bin values on the cumulative histogram are used for discrimination between images with and without landmines.Moreover,a neural classifier is used to classify images with cumulative histograms as feature vectors.The second algorithm is based on scale-space analysis with the number of speeded-up robust feature(SURF)points as the key parameter for classification.In addition,this paper presents a framework for size reduction of GPR images based on decimation for efficient storage.The further classification steps can be performed on images after interpolation.The sensitivity of classification accuracy to the interpolation process is studied in detail.展开更多
Curvature-driven diffusion (CDD) principles were used to develop a novel gradient based image restora- tion algorithm. The algorithm fills in blocks of missing data in a wireless image after transmission through the n...Curvature-driven diffusion (CDD) principles were used to develop a novel gradient based image restora- tion algorithm. The algorithm fills in blocks of missing data in a wireless image after transmission through the network. When images are transmitted over fading channels, especially in the severe circum- stances of a coal mine, blocks of the image may be destroyed by the effects of noise. Instead of using com- mon retransmission query protocols the lost data is reconstructed by using the adaptive curvature-driven diffusion (ACDD) image restoration algorithm in the gradient domain of the destroyed image. Missing blocks are restored by the method in two steps: In step one, the missing blocks are filled in the gradient domain by the ACDD algorithm; in step two, and the image is reconstructed from the reformed gradients by solving a Poisson equation. The proposed method eliminates the staircase effect and accelerates the convergence rate. This is demonstrated by experimental results.展开更多
This paper analyzes the characteristics of the output gradient histogram and shortages of several traditional automatic threshold methods in order to segment the gradient image better. Then an improved double-threshol...This paper analyzes the characteristics of the output gradient histogram and shortages of several traditional automatic threshold methods in order to segment the gradient image better. Then an improved double-threshold method is proposed, which is combined with the method of maximum classes variance, estimating-area method and double-threshold method. This method can automatically select two different thresholds to segment gradient images. The computer simulation is performed on the traditional methods and this algorithm and proves that this method can get satisfying result. Key words gradient histogram image - threshold selection - double-threshold method - maximum classes variance method CLC number TP 391. 41 Foundation item: Supported by the National Nature Science Foundation of China (50099620) and the Project of Chenguang Plan in Wuhan (985003062)Biography: YANG Shen (1977-), female, Ph. D. candidate, research direction: multimedia information processing and network technology.展开更多
The additional sparse prior of images has been the subject of much research in problems of sparse-view computed tomography（CT） reconstruction. A method employing the image gradient sparsity is often used to reduce t...The additional sparse prior of images has been the subject of much research in problems of sparse-view computed tomography（CT） reconstruction. A method employing the image gradient sparsity is often used to reduce the sampling rate and is shown to remove the unwanted artifacts while preserve sharp edges, but may cause blocky or patchy artifacts.To eliminate this drawback, we propose a novel sparsity exploitation-based model for CT image reconstruction. In the presented model, the sparse representation and sparsity exploitation of both gradient and nonlocal gradient are investigated.The new model is shown to offer the potential for better results by introducing a similarity prior information of the image structure. Then, an effective alternating direction minimization algorithm is developed to optimize the objective function with a robust convergence result. Qualitative and quantitative evaluations have been carried out both on the simulation and real data in terms of accuracy and resolution properties. The results indicate that the proposed method can be applied for achieving better image-quality potential with the theoretically expected detailed feature preservation.展开更多
Image restoration is often solved by minimizing an energy function consisting of a data-fidelity term and a regularization term.A regularized convex term can usually preserve the image edges well in the restored image...Image restoration is often solved by minimizing an energy function consisting of a data-fidelity term and a regularization term.A regularized convex term can usually preserve the image edges well in the restored image.In this paper,we consider a class of convex and edge-preserving regularization functions,i.e.,multiplicative half-quadratic regularizations,and we use the Newton method to solve the correspondingly reduced systems of nonlinear equations.At each Newton iterate,the preconditioned conjugate gradient method,incorporated with a constraint preconditioner,is employed to solve the structured Newton equation that has a symmetric positive definite coefficient matrix. The eigenvalue bounds of the preconditioned matrix are deliberately derived,which can be used to estimate the convergence speed of the preconditioned conjugate gradient method.We use experimental results to demonstrate that this new approach is efficient, and the effect of image restoration is reasonably well.展开更多
Purpose–Conventional image super-resolution reconstruction by the conventional deep learning architectures suffers from the problems of hard training and gradient disappearing.In order to solve such problems,the purp...Purpose–Conventional image super-resolution reconstruction by the conventional deep learning architectures suffers from the problems of hard training and gradient disappearing.In order to solve such problems,the purposeof this paperis to proposea novel image super-resolutionalgorithmbasedon improved generative adversarial networks(GANs)with Wasserstein distance and gradient penalty.Design/methodology/approach–The proposed algorithm first introduces the conventional GANs architecture,the Wasserstein distance and the gradient penalty for the task of image super-resolution reconstruction(SRWGANs-GP).In addition,a novel perceptual loss function is designed for the SRWGANs-GP to meet the task of image super-resolution reconstruction.The content loss is extracted from the deep model’s feature maps,and such features are introduced to calculate mean square error(MSE)for the loss calculation of generators.Findings–To validate the effectiveness and feasibility of the proposed algorithm,a lot of compared experiments are applied on three common data sets,i.e.Set5,Set14 and BSD100.Experimental results have shown that the proposed SRWGANs-GP architecture has a stable error gradient and iteratively convergence.Compared with the baseline deep models,the proposed GANs models have a significant improvement on performance and efficiency for image super-resolution reconstruction.The MSE calculated by the deep model’s feature maps gives more advantages for constructing contour and texture.Originality/value–Compared with the state-of-the-art algorithms,the proposed algorithm obtains a better performance on image super-resolution and better reconstruction results on contour and texture.展开更多
Objective To investigate the clinical value of different magnetic resonance (MR) pulse sequences in diagnosis of spinal metastatic tumor. Methods Fifteen patients with clinically suspected spinal metastatic tumor were...Objective To investigate the clinical value of different magnetic resonance (MR) pulse sequences in diagnosis of spinal metastatic tumor. Methods Fifteen patients with clinically suspected spinal metastatic tumor were included in this study. These patients were with documented primary tumors. Four MR pulse sequences, T1-weighted spin echo (T1WI SE), T2-weighted fast spin echo (T2WI FSE), short time inversion recovery (STIR), and gradient echo 2-D multi echo data imaging combination (GE Me-2D) were used to detect spinal metastasis. Results Fifteen vertebral bodies were entire involvement, 38 vertebral bodies were section involvement, and totally 53 vertebral bodies were involved. There were 19 focal infections in pedicle of vertebral arch, 15 metastases in spinous process and transverse process. Fifty-three vertebral bodies were abnormal in T1WI SE and GE Me-2D, 35 vertebral bodies were found abnormal in T2WI FSE, and 50 vertebral bodies were found abnormal in STIR. The verges of focal signal of involved vertebral bodies were comparatively clear in T1WI SE, comparatively clear or vague in T2WI FSE, vague in STIR, and clear in GE Me-2D.Conclusions GE Me-2D may be the most sensitive technique to detect metastases. So three sequences (T1WI SE, T2WI FSE, GE Me-2D) can demonstrate the early changes of spinal metastasis roundly.展开更多
The application of cellular neural network (CNN) has made great progress in image processing. When the selected objects extraction (SOE) CNN is applied to gray scale images, its effects depend on the choice of initial...The application of cellular neural network (CNN) has made great progress in image processing. When the selected objects extraction (SOE) CNN is applied to gray scale images, its effects depend on the choice of initial points. In this paper, we take medical images as an example to analyze this limitation. Then an improved algorithm is proposed in which we can segment any gray level objects regardless of the limitation stated above. We also use the gradient information and contour detection CNN to determine the contour and ensure the veracity of segmentation effectively. Finally, we apply the improved algorithm to tumor segmentation of the human brain MR image. The experimental results show that the algorithm is practical and effective.展开更多
Recently, exploiting low rank property of the data accomplished by the non-convex optimization has shown great potential to decrease measurements for compressed sensing. In this paper, the low rank regularization is a...Recently, exploiting low rank property of the data accomplished by the non-convex optimization has shown great potential to decrease measurements for compressed sensing. In this paper, the low rank regularization is adopted to gradient similarity minimization, and applied for highly undersampled magnetic resonance imaging(MRI) reconstruction, termed gradient-based low rank MRI reconstruction(GLRMRI). In the proposed method,by incorporating the spatially adaptive iterative singular-value thresholding(SAIST) to optimize our gradient scheme, the deterministic annealing iterates the procedure efficiently and superior reconstruction performance is achieved. Extensive experimental results have consistently demonstrated that GLRMRI recovers both realvalued MR images and complex-valued MR data accurately, especially in the edge preserving perspective, and outperforms the current state-of-the-art approaches in terms of higher peak signal to noise ratio(PSNR) and lower high-frequency error norm(HFEN) values.展开更多
Because of the combination of optimization algorithms and full wave equations, full-waveform inversion(FWI) has become the frontier of the study of seismic exploration and is gradually becoming one of the essential to...Because of the combination of optimization algorithms and full wave equations, full-waveform inversion(FWI) has become the frontier of the study of seismic exploration and is gradually becoming one of the essential tools for obtaining the Earth interior information. However, the application of conventional FWI to pure reflection data in the absence of a highly accurate starting velocity model is difficult. Compared to other types of seismic waves, reflections carry the information of the deep part of the subsurface. Reflection FWI, therefore, is able to improve the accuracy of imaging the Earth interior further. Here, we demonstrate a means of achieving this successfully by interleaving least-squares RTM with a version of reflection FWI in which the tomographic gradient that is required to update the background macro-model is separated from the reflectivity gradient using the Born approximation during forward modeling. This provides a good update to the macro-model. This approach is then followed by conventional FWI to obtain a final high-fidelity high-resolution result from a poor starting model using only reflection data.Further analysis reveals the high-resolution result is achieved due to a deconvolution imaging condition implicitly used by FWI.展开更多
基金This research was funded by the Deanship of Scientific Research at Princess Nourah Bint Abdulrahman University through the Fast-track Research Funding Program。
文摘Recently,ground-penetrating radar(GPR)has been extended as a well-known area to investigate the subsurface objects.However,its output has a low resolution,and it needs more processing for more interpretation.This paper presents two algorithms for landmine detection from GPR images.The first algorithm depends on a multi-scale technique.A Gaussian kernel with a particular scale is convolved with the image,and after that,two gradients are estimated;horizontal and vertical gradients.Then,histogram and cumulative histogram are estimated for the overall gradient image.The bin values on the cumulative histogram are used for discrimination between images with and without landmines.Moreover,a neural classifier is used to classify images with cumulative histograms as feature vectors.The second algorithm is based on scale-space analysis with the number of speeded-up robust feature(SURF)points as the key parameter for classification.In addition,this paper presents a framework for size reduction of GPR images based on decimation for efficient storage.The further classification steps can be performed on images after interpolation.The sensitivity of classification accuracy to the interpolation process is studied in detail.
基金supported by the National High-Tech Research and Development Program of China (No. 2008AA062200)the National Natural Science Foundation of China (No.60802077)the Fundamental Research Funds for the Central Universities (No. 2010QNA43)
文摘Curvature-driven diffusion (CDD) principles were used to develop a novel gradient based image restora- tion algorithm. The algorithm fills in blocks of missing data in a wireless image after transmission through the network. When images are transmitted over fading channels, especially in the severe circum- stances of a coal mine, blocks of the image may be destroyed by the effects of noise. Instead of using com- mon retransmission query protocols the lost data is reconstructed by using the adaptive curvature-driven diffusion (ACDD) image restoration algorithm in the gradient domain of the destroyed image. Missing blocks are restored by the method in two steps: In step one, the missing blocks are filled in the gradient domain by the ACDD algorithm; in step two, and the image is reconstructed from the reformed gradients by solving a Poisson equation. The proposed method eliminates the staircase effect and accelerates the convergence rate. This is demonstrated by experimental results.
文摘This paper analyzes the characteristics of the output gradient histogram and shortages of several traditional automatic threshold methods in order to segment the gradient image better. Then an improved double-threshold method is proposed, which is combined with the method of maximum classes variance, estimating-area method and double-threshold method. This method can automatically select two different thresholds to segment gradient images. The computer simulation is performed on the traditional methods and this algorithm and proves that this method can get satisfying result. Key words gradient histogram image - threshold selection - double-threshold method - maximum classes variance method CLC number TP 391. 41 Foundation item: Supported by the National Nature Science Foundation of China (50099620) and the Project of Chenguang Plan in Wuhan (985003062)Biography: YANG Shen (1977-), female, Ph. D. candidate, research direction: multimedia information processing and network technology.
基金Project supported by the National Natural Science Foundation of China(Grant No.61372172)
文摘The additional sparse prior of images has been the subject of much research in problems of sparse-view computed tomography（CT） reconstruction. A method employing the image gradient sparsity is often used to reduce the sampling rate and is shown to remove the unwanted artifacts while preserve sharp edges, but may cause blocky or patchy artifacts.To eliminate this drawback, we propose a novel sparsity exploitation-based model for CT image reconstruction. In the presented model, the sparse representation and sparsity exploitation of both gradient and nonlocal gradient are investigated.The new model is shown to offer the potential for better results by introducing a similarity prior information of the image structure. Then, an effective alternating direction minimization algorithm is developed to optimize the objective function with a robust convergence result. Qualitative and quantitative evaluations have been carried out both on the simulation and real data in terms of accuracy and resolution properties. The results indicate that the proposed method can be applied for achieving better image-quality potential with the theoretically expected detailed feature preservation.
基金supported by the National Basic Research Program (No.2005CB321702)the National Outstanding Young Scientist Foundation(No. 10525102)the Specialized Research Grant for High Educational Doctoral Program(Nos. 20090211120011 and LZULL200909),Hong Kong RGC grants and HKBU FRGs
文摘Image restoration is often solved by minimizing an energy function consisting of a data-fidelity term and a regularization term.A regularized convex term can usually preserve the image edges well in the restored image.In this paper,we consider a class of convex and edge-preserving regularization functions,i.e.,multiplicative half-quadratic regularizations,and we use the Newton method to solve the correspondingly reduced systems of nonlinear equations.At each Newton iterate,the preconditioned conjugate gradient method,incorporated with a constraint preconditioner,is employed to solve the structured Newton equation that has a symmetric positive definite coefficient matrix. The eigenvalue bounds of the preconditioned matrix are deliberately derived,which can be used to estimate the convergence speed of the preconditioned conjugate gradient method.We use experimental results to demonstrate that this new approach is efficient, and the effect of image restoration is reasonably well.
文摘Purpose–Conventional image super-resolution reconstruction by the conventional deep learning architectures suffers from the problems of hard training and gradient disappearing.In order to solve such problems,the purposeof this paperis to proposea novel image super-resolutionalgorithmbasedon improved generative adversarial networks(GANs)with Wasserstein distance and gradient penalty.Design/methodology/approach–The proposed algorithm first introduces the conventional GANs architecture,the Wasserstein distance and the gradient penalty for the task of image super-resolution reconstruction(SRWGANs-GP).In addition,a novel perceptual loss function is designed for the SRWGANs-GP to meet the task of image super-resolution reconstruction.The content loss is extracted from the deep model’s feature maps,and such features are introduced to calculate mean square error(MSE)for the loss calculation of generators.Findings–To validate the effectiveness and feasibility of the proposed algorithm,a lot of compared experiments are applied on three common data sets,i.e.Set5,Set14 and BSD100.Experimental results have shown that the proposed SRWGANs-GP architecture has a stable error gradient and iteratively convergence.Compared with the baseline deep models,the proposed GANs models have a significant improvement on performance and efficiency for image super-resolution reconstruction.The MSE calculated by the deep model’s feature maps gives more advantages for constructing contour and texture.Originality/value–Compared with the state-of-the-art algorithms,the proposed algorithm obtains a better performance on image super-resolution and better reconstruction results on contour and texture.
文摘Objective To investigate the clinical value of different magnetic resonance (MR) pulse sequences in diagnosis of spinal metastatic tumor. Methods Fifteen patients with clinically suspected spinal metastatic tumor were included in this study. These patients were with documented primary tumors. Four MR pulse sequences, T1-weighted spin echo (T1WI SE), T2-weighted fast spin echo (T2WI FSE), short time inversion recovery (STIR), and gradient echo 2-D multi echo data imaging combination (GE Me-2D) were used to detect spinal metastasis. Results Fifteen vertebral bodies were entire involvement, 38 vertebral bodies were section involvement, and totally 53 vertebral bodies were involved. There were 19 focal infections in pedicle of vertebral arch, 15 metastases in spinous process and transverse process. Fifty-three vertebral bodies were abnormal in T1WI SE and GE Me-2D, 35 vertebral bodies were found abnormal in T2WI FSE, and 50 vertebral bodies were found abnormal in STIR. The verges of focal signal of involved vertebral bodies were comparatively clear in T1WI SE, comparatively clear or vague in T2WI FSE, vague in STIR, and clear in GE Me-2D.Conclusions GE Me-2D may be the most sensitive technique to detect metastases. So three sequences (T1WI SE, T2WI FSE, GE Me-2D) can demonstrate the early changes of spinal metastasis roundly.
基金the National Natural Science Foundation of China (Grant No. 69772004) the National Basic Research Program (G1999054000) of China.
文摘The application of cellular neural network (CNN) has made great progress in image processing. When the selected objects extraction (SOE) CNN is applied to gray scale images, its effects depend on the choice of initial points. In this paper, we take medical images as an example to analyze this limitation. Then an improved algorithm is proposed in which we can segment any gray level objects regardless of the limitation stated above. We also use the gradient information and contour detection CNN to determine the contour and ensure the veracity of segmentation effectively. Finally, we apply the improved algorithm to tumor segmentation of the human brain MR image. The experimental results show that the algorithm is practical and effective.
基金the National Natural Science Foundation of China(Nos.61362001,61503176,61661031)Jiangxi Advanced Project for Post-Doctoral Research Fund(No.2014KY02)
文摘Recently, exploiting low rank property of the data accomplished by the non-convex optimization has shown great potential to decrease measurements for compressed sensing. In this paper, the low rank regularization is adopted to gradient similarity minimization, and applied for highly undersampled magnetic resonance imaging(MRI) reconstruction, termed gradient-based low rank MRI reconstruction(GLRMRI). In the proposed method,by incorporating the spatially adaptive iterative singular-value thresholding(SAIST) to optimize our gradient scheme, the deterministic annealing iterates the procedure efficiently and superior reconstruction performance is achieved. Extensive experimental results have consistently demonstrated that GLRMRI recovers both realvalued MR images and complex-valued MR data accurately, especially in the edge preserving perspective, and outperforms the current state-of-the-art approaches in terms of higher peak signal to noise ratio(PSNR) and lower high-frequency error norm(HFEN) values.
基金supported by the National Natural Science Foundation of China(Grant No.41504106&41274099)the Science Foundation of China University of Petroleum(Beijing)(Grant No.2462015YJRC012)State Laboratory of Petroleum Resource and Prospecting(Grant No.PRP/indep-3-1508)
文摘Because of the combination of optimization algorithms and full wave equations, full-waveform inversion(FWI) has become the frontier of the study of seismic exploration and is gradually becoming one of the essential tools for obtaining the Earth interior information. However, the application of conventional FWI to pure reflection data in the absence of a highly accurate starting velocity model is difficult. Compared to other types of seismic waves, reflections carry the information of the deep part of the subsurface. Reflection FWI, therefore, is able to improve the accuracy of imaging the Earth interior further. Here, we demonstrate a means of achieving this successfully by interleaving least-squares RTM with a version of reflection FWI in which the tomographic gradient that is required to update the background macro-model is separated from the reflectivity gradient using the Born approximation during forward modeling. This provides a good update to the macro-model. This approach is then followed by conventional FWI to obtain a final high-fidelity high-resolution result from a poor starting model using only reflection data.Further analysis reveals the high-resolution result is achieved due to a deconvolution imaging condition implicitly used by FWI.
