Super-Resolution Reconstruction of Images Based on Microarray Camera

In the field of images and imaging, super-resolution (SR) reconstruction of images is a technique that converts one or more low-resolution (LR) images into a highresolution (HR) image. The classical two types of SR methods are mainly based on applying a single image or multiple images captured by a single camera. Microarray camera has the characteristics of small size, multi views, and the possibility of applying to portable devices. It has become a research hotspot in image processing. In this paper, we propose a SR reconstruction of images based on a microarray camera for sharpening and registration processing of array images. The array images are interpolated to obtain a HR image initially followed by a convolution neural network (CNN) procedure for enhancement. The convolution layers of our convolution neural network are 3×3 or 1×1 layers, of which the 1×1 layers are used to improve the network performance particularly. A bottleneck structure is applied to reduce the parameter numbers of the nonlinear mapping and to improve the nonlinear capability of the whole network. Finally, we use a 3×3 deconvolution layer to significantly reduce the number of parameters compared to the deconvolution layer of FSRCNN-s. The experiments show that the proposed method can not only ameliorate effectively the texture quality of the target image based on the array images information, but also further enhance the quality of the initial high resolution image by the improved CNN.     

Multicriteria maximum likelihood neural network approach to positron emission tomography

The emerging technology of positron emission image reconstruction is introduced in this paper as a multicriteria optimization problem. We show how selected families of objective functions may be used to reconstruct positron emission images. We develop a novel neural network approach to positron emission imaging problems. We also studied the most frequently used image reconstruction methods, namely, maximum likelihood under the framework of single performance criterion optimization. Finally, we introduced some of the results obtained by various reconstruction algorithms using computer‐generated noisy projection data from a chest phantom and real positron emission tomography (PET) scanner data. Comparison of the reconstructed images indicated that the multicriteria optimization method gave the best in error, smoothness (suppression of noise), gray value resolution, and ghost‐free images. © 2001 John Wiley & Sons, Inc. Int J Imaging Syst Technol 11, 361–364, 2000     

Image SR-based NLM and DCNN improved IBP with cubic B-spline

ABSTRACT

Image super-resolution (SR) techniques aim to estimate high-resolution (HR) image from low-resolution (LR) image. Existing SR method has slow convergence and recovery of high-frequency details are inaccurate. To overcome these issues, two algorithms have been proposed for image SR based on non-local means improved iterative back projection (NLM-IIBP), deep convolutional neural network improved iterative back projection (DCNN-IIBP) to produce high-resolution images with low noise, minimal blur by restoring high-frequency details. In NLM-IIBP denoised images have been interpolated using cubic B-spline interpolation and processed using IIBP based on guided bilateral method. NLM preserves the edges effectively, but does not consider high dimensional information and over smoothing during noise minimization. To further improve the resolution, NLM is replaced by DCNN. DCNN denoising method suppresses different noises at different noise levels. The proposed algorithms have been analysed and compared with existing approaches using various parameters to prove the effectiveness.     

彩色图像多尺度引导的深度图像超分辨率重建

为获得更优的深度图像超分辨率重建结果,本文构建了彩色图像多尺度引导深度图像超分辨率重建卷积神经网络。该网络使用多尺度融合方法实现高分辨率(HR)彩色图像特征对低分辨率(LR)深度图像特征的引导,有益于恢复图像细节信息。在对LR深度图像提取特征的过程中,构建了多感受野残差块(MRFRB)提取并融合不同感受野下的特征,然后将每一个MRFRB输出的特征连接、融合,得到全局融合特征。最后,通过亚像素卷积层和全局融合特征,得到HR深度图像。实验结果表明,该算法得到的超分辨率图像缓解了边缘失真和伪影问题,有较好的视觉效果。     

A neural network approach to microwave imaging

We present a neural network approach to microwave imaging for medical diagnosis. The problem is to reconstruct the complex permittivity of the biological tissues illuminated by the transverse magnetic (TM) incident waves. In order to avoid the inherent ill‐posedness of the inverse scattering problem, we introduce a stochastic process based on Markov random field and a priori knowledge. A coupled gradient neural network is proposed to deal with the mixed‐variable problem because the reconstructed dielectric permittivities are continuous complex variables and the line processes, which can preserve the edges of the reconstructed image, are binary variables. We report the numerical results of a simple human forearm model. We also point out the advantages and the limitations of this method. © 2001 John Wiley & Sons, Inc. Int J Imaging Syst Technol 11, 159–163, 2000     

Fast image vector quantization using a modified competitive learning neural network approach

The basic goal of image compression through vector quantization (VQ) is to reduce the bit rate for transmission or data storage while maintaining an acceptable fidelity or image quality. The advantage of VQ image compression is its fast decompression by table lookup technique. However, the codebook supplied in advance may not handle the changing image statistics very well. The need for online codebook generation became apparent. The competitive learning neural network design has been used for vector quantization. However, its training time can be very long, and the number of output nodes is somewhat arbitrarily decided before the training starts. Our modified approach presents a fast codebook generation procedure by searching for an optimal number of output nodes evolutively. The results on two medical images show that this new approach reduces the training time considerably and still maintains good quality for recovered images. © 1997 John Wiley & Sons, Inc. Int J Imaging Syst Technol, 8, 413–418, 1997     

Application of Hybrid Learning Neural Fuzzy Systems in Reliability Prediction

This study presents a hybrid learning neural fuzzy system for accurately predicting system reliability. Neural fuzzy system learning with and without supervision has been successfully applied in control systems and pattern recognition problems. This investigation modifies the hybrid learning fuzzy systems to accept time series data and therefore examines the feasibility of reliability prediction. Two neural network systems are developed for solving different reliability prediction problems. Additionally, a scaled conjugate gradient learning method is applied to accelerate the training in the supervised learning phase. Several existing approaches, including feed‐forward multilayer perceptron (MLP) networks, radial basis function (RBF) neural networks and Box–Jenkins autoregressive integrated moving average (ARIMA) models, are used to compare the performance of the reliability prediction. The numerical results demonstrate that the neural fuzzy systems have higher prediction accuracy than the other methods. Copyright © 2005 John Wiley & Sons, Ltd.     

Computer‐aided diagnosis system for the detection of bronchiectasis in chest computed tomography images

A computer‐aided diagnosis (CAD) system has been developed for the detection of bronchiectasis from computed tomography (CT) images of chest. A set of CT images of the chest with known diagnosis were collected and these images were first denoised using Wiener filter. The lung tissue was then segmented using optimal thresholding. The Pathology Bearing Regions (PBRs) were then extracted by applying pixel‐based segmentation. For each PBR, a gray level co‐occurrence matrix (GLCM) was constructed. From the GLCM texture features were extracted and feature vectors were constructed. A probabilistic neural network (PNN) was constructed and trained using this set of feature vectors. The images together with the PBRs and the corresponding feature vector and diagnosis were stored in an image database. Rules for diagnosis and for determining the severity of the disease were generated by analyzing the images known to be affected by bronchiectasis. The rules were then validated by a human expert. The validated rules were stored in the Knowledge Base. When a physician gives a CT image to the CAD system, it first transforms the image into a set of feature vectors, one for each PBR in the image. It then performs the diagnosis using two techniques: PNN and mahalanobis distance measure. The final diagnosis and the severity of the disease are determined by correlating the diagnosis determined by both the techniques in consultation with the knowledge base. The system also retrieves similar cases from the database. Thus, this system would aid the physicians in diagnosing bronchiectasis. © 2009 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 19, 290–298, 2009     

Bidirectional multiple‐valued neural network for pattern recognition and associative recall

A new neural network algorithm based on the counter‐propagation network (CPN) architecture, named MVL‐CPN, is proposed in this paper for bidirectional mapping and recognition of multiple‐valued patterns. The MVL‐CPN is capable of performing a mathematical mapping of a set of multiple‐valued vector pairs by self‐organization. The use of MVL‐CPN reduces considerably the number of nodes required for the input layers as well as the number of synaptic weights compared to the binary CPN. The training of the network is stable because all synaptic weights are monotonically nonincreasing. The bidirectional mapping and associative recall features of the MVL‐CPN are tested by using various sets of quaternary patterns. It is observed that the MVL‐CPN can converge within three or four iterations. The high‐speed convergence characteristics of the network can lead to the possibility of using this architecture for real‐time applications. An important advantage of the proposed type of neural network is that it can be implemented in VLSI with reduced number of neurons and synaptic weights when compared to a larger binary network needed for the same application. © 2000 John Wiley & Sons, Inc. Int J Imaging Syst Technol 11, 125–129, 2000     

Multicriteria second‐order neural networks approach to imaging through turbulence

Atmospheric turbulence can greatly limit the spatial resolution in optical images obtained of space objects when imaged with ground‐based telescopes. Two widely used algorithms to remove atmospheric turbulence in this class of images are blind de‐convolution and speckle imaging. Both algorithms are effective in removing atmospheric turbulence, but they use different types of prior knowledge and have different strengths and weaknesses. We have developed a multicriteria cross entropy minimization approach to imaging through atmospheric turbulence and a second‐order neural network implementations. Our simulations illustrated the efficiency of our method. © 2003 Wiley Periodicals, Inc. Int J Imaging Syst Technol 13, 146–151, 2003; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ima.10037     

A novel blind super‐resolution technique based on the improved Poisson maximum a posteriori algorithm

Image restoration has received considerable attention. In many practical situations, unfortunately, the blur is often unknown, and little information is available about the true image. Therefore, the true image is identified directly from the corrupted image by using partial or no information about the blurring process and the true image. In addition, noise will be amplified to induce severely ringing artifacts in the process of restoration. This article proposes a novel technique for the blind super‐resolution, whose mechanism alternates between de‐convolution of the image and the point spread function based on the improved Poisson maximum a posteriori super‐resolution algorithm. This improved Poisson MAP super‐resolution algorithm incorporates the functional form of a Wiener filter into the Poisson MAP algorithm operating on the edge image further to reduce noise effects and speed restoration. Compared with that based on the Poisson MAP, the novel blind super‐resolution technique presents experimental results from 1‐D signals and 2‐D images corrupted by Gaussian point spread functions and additive noise with significant improvements in quality. © 2003 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 12, 239–246, 2002; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ima.10032     

SRResNet Performance Enhancement Using Patch Inputs and Partial Convolution-Based Padding

Due to highly underdetermined nature of Single Image Super-Resolution (SISR) problem, deep learning neural networks are required to be more deeper to solve the problem effectively. One of deep neural networks successful in the Super-Resolution (SR) problem is ResNet which can render the capability of deeper networks with the help of skip connections. However, zero padding (ZP) scheme in the network restricts benefits of skip connections in SRResNet and its performance as the ratio of the number of pure input data to that of zero padded data increases. In this paper. we consider the ResNet with Partial Convolution based Padding (PCP) instead of ZP to solve SR problem. Since training of deep neural networks using patch images is advantageous in many aspects such as the number of training image data and network complexities, patch image based SR performance is compared with single full image based one. The experimental results show that patch based SRResNet SR results are better than single full image based ones and the performance of deep SRResNet with PCP is better than the one with ZP.     

Integrated segmentation of brain tumor images for radiotherapy and neurosurgery

Segmentation of brain tumor images is an important task in diagnosis and treatment planning for cancer patients. To achieve this goal with standard clinical acquisition protocols, conventionally, either classification algorithms are applied on multimodal MR images or atlas‐based segmentation is used on a high‐resolution monomodal MR image. These two approaches have been commonly regarded separately. We propose to integrate all the available imaging information into one framework to be able to use the information gained from the tissue classification of the multimodal images to perform a more precise segmentation on the high‐resolution monomodal image by atlas‐based segmentation. For this, we combine a state of the art regularized classification method with an enhanced version of an atlas‐registration approach including multiscale tumor‐growth modeling. This contribution offers the possibility to simultaneously segment subcortical structures in the patient by warping the respective atlas labels, which is important for neurosurgical planning and radiotherapy planning. © 2013 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 23, 59–63, 2013     

基于BP神经网络的ECT图像重建算法

针对电容层析成像图像重建问题的病态性,利用COMSOL软件建立系统模型,并结合MATLAB实现正问题的求解。依据BP神经网络所具有的理想的非线性映射和联想记忆功能实现了由检测电容值到重建图像灰度值之间的非线性映射,避免了传统算法中对灵敏度矩阵求解的繁琐,克服了因线性化处理所导致的成像精度低的缺点。在MATLAB平台下,采用2种滤波方法进行滤波,对图像增强修复,提高了图像质量。     

MAP fusion method for superresolution of images with locally varying pixel quality

Superresolution is a procedure that produces a high‐resolution image from a set of low‐resolution images. Many of superresolution techniques are designed for optical cameras, which produce pixel values of well‐defined uncertainty, while there are still various imaging modalities for which the uncertainty of the images is difficult to control. To construct a superresolution image from low‐resolution images with varying uncertainty, one needs to keep track of the uncertainty values in addition to the pixel values. In this paper, we develop a probabilistic approach to superresolution to address the problem of varying uncertainty. As direct computation of the analytic solution for the superresolution problem is difficult, we suggest a novel algorithm for computing the approximate solution. As this algorithm is a noniterative method based on Kalman filter‐like recursion relations, there is a potential for real‐time implementation of the algorithm. To show the efficiency of our method, we apply this algorithm to a video sequence acquired by a forward looking sonar system. © 2008 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 18, 242–250, 2008; Published online in Wiley InterScience (www.interscience.wiley.com).     

Multimodal medical image fusion based on discrete Tchebichef moments and pulse coupled neural network

Multimodal medical image fusion plays a vital role in clinical diagnoses and treatment planning. In many image fusion methods‐based pulse coupled neural network (PCNN), normalized coefficients are used to motivate the PCNN, and this makes the fused image blur, detail loss, and decreases contrast. Moreover, they are limited in dealing with medical images with different modalities. In this article, we present a new multimodal medical image fusion method based on discrete Tchebichef moments and pulse coupled neural network to overcome the aforementioned problems. First, medical images are divided into equal‐size blocks and the Tchebichef moments are calculated to characterize image shape, and energy of blocks is computed as the sum of squared non‐DC moment values. Then to retain edges and textures, the energy of Tchebichef moments for blocks is introduced to motivate the PCNN with adaptive linking strength. Finally, large firing times are selected as coefficients of the fused image. Experimental results show that the proposed scheme outperforms state‐of‐the‐art methods and it is more effective in processing medical images with different modalities. © 2017 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 27, 57–65, 2017     

Synthetic‐aperture technique for high‐resolution composite imaging of the inside walls of tubular specimens

High‐resolution survey of the inside walls of tubular specimens is a unique application of synthetic‐aperture composite imaging. This article describes the data acquisition process, 3D motion estimation and compensation, image registration, and superposition for the formation of high‐resolution composite images from conventional video sequences. Experiment results from the survey of an oil well are used to demonstrate the capability of the technique. © 2004 Wiley Periodicals, Inc. Int J Imaging Syst Technol 14, 167–169, 2004; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ima.20020     

基于颜色空间变换和CNN的自适应去模糊方法

目的研究数字图像中的去模糊问题,从受损的模糊图像中恢复出清晰图像。方法针对现有图像去模糊算法无法保留图像高频信息及容易产生振铃效应等问题,提出一种基于Y通道反卷积和卷积神经网络的两阶段自适应去模糊算法(SDYCNN)。在第1阶段,将数字图像转换至YUV颜色空间,根据图像无参考质量评价分数与模糊核尺寸之间的对应关系,在Y通道内自适应确定模糊核尺寸并进行反卷积增强;第2阶段将第1阶段中的反卷积增强作为预处理方式,通过4层卷积神经网络建立反卷积增强后的图像与清晰图像之间的映射关系,实现图像去模糊。结果轻微模糊图像在第1阶段便能够得到较好的去模糊效果,严重模糊图像经过第1阶段的反卷积增强,也有助于神经网络中特征的快速提取。结论实验结果表明,该算法不仅对于模糊图像具有良好的恢复效果,运算效率也有显著提升。     

A deep learning model integrating convolution neural network and multiple kernel K means clustering for segmenting brain tumor in magnetic resonance images

In medical imaging, segmenting brain tumor becomes a vital task, and it provides a way for early diagnosis and treatment. Manual segmentation of brain tumor in magnetic resonance (MR) images is a time‐consuming and challenging task. Hence, there is a need for a computer‐aided brain tumor segmentation approach. Using deep learning algorithms, a robust brain tumor segmentation approach is implemented by integrating convolution neural network (CNN) and multiple kernel K means clustering (MKKMC). In this proposed CNN‐MKKMC approach, classification of MR images into normal and abnormal is performed by CNN algorithm. At next, MKKMC algorithm is employed to segment the brain tumor from the abnormal brain image. The proposed CNN‐MKKMC algorithm is evaluated both visually and objectively in terms of accuracy, sensitivity, and specificity with the existing segmentation methods. The experimental results demonstrate that the proposed CNN‐MKKMC approach yields better accuracy in segmenting brain tumor with less time cost.