A two-stage convolutional sparse prior model for image restoration

Image restoration (IR) from noisy, blurred or/and incomplete observed measurement is one of the important tasks in image processing community. Image prior is of utmost importance for recovering a high quality image. In this paper, we present a two-stage convolutional sparse prior model for efficient image restoration. The multi-view features prior is first obtained by convolving the image with the Fields-of-Experts (FoE) filters and then the resulting multi-view features are represented by convolutional sparse coding (CSC) prior. By taking advantage of the convolutional filters, the proposed two-stage model inherits the strengths of multi-view features and CSC priors. The assembled multi-view features contain high-frequency, redundancy, and large range of feature orientations, which are favor to be represented by CSC and consequently for better image recovery. Augmented Lagrangian and alternating direction method of multipliers are employed to decouple the nonlinear optimization problem in order to iteratively approach the optimum solution. The results of various experiments on image deblurring and compressed sensing magnetic resonance imaging (CS-MRI) reconstruction consistently demonstrate that the proposed algorithm efficiently recovers image and presents advantages over the current leading restoration approaches.     

Patchwise scaling method for content-aware image resizing

Image resizing becomes more and more important in content-aware image displaying. This paper proposes a patchwise scaling method to resize an image to emphasize the important areas and preserve the globally visual effect (smoothness, coherence and integrity). This method for resizing image is based on optimizing the image distance presented in this paper. The image distance is defined based on so-called local bidirectional similarity measurement and smoothness measurement to quantify the quality of resizing outputs. The original image is divided into small important patches and unimportant patches based on an important map. The important map is generated automatically using a novel combination of image edge and saliency measurement. A scaling factor is computed for each small patch. The resized image is produced by iteratively optimizing, which is based on our image distance, the scaling factor for each small patch. Experiments of different type images demonstrate that our method can be effectively used in image processing applications to locally shrink and enlarge important areas while preserving image quality.     

Virtual view synthesis using layered depth image generation and depth-based inpainting for filling disocclusions and translucent disocclusions

View synthesis is an efficient solution to produce content for 3DTV and FTV. However, proper handling of the disocclusions is a major challenge in the view synthesis. Inpainting methods offer solutions for handling disocclusions, though limitations in foreground-background classification causes the holes to be filled with inconsistent textures. Moreover, the state-of-the art methods fail to identify and fill disocclusions in intermediate distances between foreground and background through which background may be visible in the virtual view (translucent disocclusions). Aiming at improved rendering quality, we introduce a layered depth image (LDI) in the original camera view, in which we identify and fill occluded background so that when the LDI data is rendered to a virtual view, no disocclusions appear but views with consistent data are produced also handling translucent disocclusions. Moreover, the proposed foreground-background classification and inpainting fills the disocclusions with neighboring background texture consistently. Based on the objective and subjective evaluations, the proposed method outperforms the state-of-the art methods at the disocclusions.     

基于分等级对象语义图模型的复杂目标自动识别方法研究

目标自动识别是图像处理领域的研究热点。针对现有方法的不足,该文提出一种新的基于分等级对象语义图模型的复杂目标自动识别方法。该方法通过构建分等级对象语义图模型增强对目标与背景间、目标部件间语义约束的利用,引入置信对象网络统计局部特性,利用消息机制传递对象间相互影响,实现概率语义分析。训练中还将产生式和判别式方法结合,提高了目标识别的准确度。在自然和遥感部分目标类别数据集上的测试结果表明,该方法能完成对多种类型和复杂结构目标的识别和提取,具有一定的实用价值。     

一种基于像素差值编码度量不确定SAR图像相似性的新方法

在SAR图像目标识别、图像匹配等应用中,定义合理可靠的相似度尤为重要。该文在分析现有SAR图像相似度基础上,提出一种基于像素差值编码的相似度准则。首先将SAR图像按照相邻像素灰度差异生成编码图像,然后以编码图像之间的一致性作为相似度。该文从理论上证明了该相似度对SAR图像中相干斑噪声、部分遮挡和模糊等因素的鲁棒性和适应性,还讨论了将该准则应用于SAR图像匹配时,如何针对不确定性,给出一定置信水平下所有合理的匹配位置。理论和实验结果表明该文提出的相似度准则对SAR图像上相干斑噪声、部分遮挡以及模糊不敏感,能有效应用于不确定SAR图像的匹配。     

An ensemble image quality assessment algorithm based on deep feature clustering

To effective handle image quality assessment (IQA) where the images might be with sophisticated characteristics, we proposed a deep clustering-based ensemble approach for image quality assessment toward diverse images. Our approach is based on a convolutional DAE-aware deep architecture. By leveraging a layer-by-layer pre-training, our proposed deep feature clustering architecture extracted a fixed number of high-level features at first. Then, it optimally splits image samples into different clusters by using the fuzzy C-means algorithm based on the engineered deep features. For each cluster, we simulated a particular fitting function of differential mean opinion scores with each assessed image’s PSNR, SIMM, and VIF scores. Comprehensive experimental results on TID2008, TID2013 and LIVE databases have demonstrated that compared to the state-of-the-art counterparts, our proposed IQA method can reflect the subjective quality of images more accurately by seamlessly integrating the advantages of three existed IQA methods.     

A survey on analysis and implementation of state-of-the-art haze removal techniques

Haze is a poor-quality state described by the opalescent appearance of the atmosphere which reduces the visibility. It is caused by high concentrations of atmospheric air pollutants, such as dust, smoke and other particles that scatter and absorb sunlight. The poor visibility can result in the failure of multiple computer vision applications such as smart transport systems, image processing, object detection, surveillance etc. One of the major issues in the field of image processing is the restoration of images that are corrupted due to different degradations. Typically, the images or videos captured in the outside environment have low contrast, colour fade and restricted visibility due to suspended particles of the atmosphere that directly influence the image quality. This can cause difficulty in identifying the objects in the captured hazy images or frames. To address this problem, several image dehazing techniques have been developed in the literature, each of which has its own advantages and limitations, but effective image restoration remains a challenging task. In recent times, various learning (Machine learning & Deep learning) based methods greatly condensed the drawbacks of manual design of haze related features and reduces the difficulty in efficient restoration of images with less computational time and cost. The current state-of-the-art methods for haze free images, mainly from the last decade, are thoroughly examined in this survey. Moreover, this paper systematically summarizes the hardware implementations of various haze removal methods in real time. It is with the hope that this current survey acts as a reference for researchers in this scientific area and to provide a direction for future improvements based on current achievements.     

基于各向异性自适应高斯加权方向窗的非局部三维Otsu图像门限分割

针对传统3维Otsu(3D-Otsu)门限分割方法中的滤噪性能和小目标保持性能的不足,该文提出一种基于各向异性自适应高斯加权方向窗的3D-Otsu门限分割的新方法。新方法改进了3D-Otsu的邻域窗口设置方法,采用中心点的局部特征来自适应地确定邻域各向异性高斯加权方向窗口的尺寸、尺度和滤波方向。然后,提出非局部多方向相似度测量来更有效地捕捉图像中的模式冗余。最终,结合像素点灰度值、加权均值、加权中值构建3维直方图,并基于最大类间方差计算门限矢量进行分割。实验结果表明：与目前广泛使用的2维Otsu, 2维最大熵以及传统3维Otsu方法相比,新方法有着更好的门限分割效果,并具有更好的滤噪性能和小目标保持性能。     

一种新的SAR图像相干斑抑制算法

针对经典局域统计自适应滤波器的窗口大小问题,提出了一种新的用于抑制合成孔径雷达图像相干斑噪声的滤波算法.基于SAR图像的统计特性,窗口大小根据区域特征自动调整,尽可能将窗口调整为包含同一均匀区域,从而在保持细节的同时尽最大可能地抑制斑点噪声.实验结果表明,本算法不仅有效抑制了SAR图像的相干斑,保持图像的边缘细节.而且有着良好的图像视觉效果.     

光无线局域网上行链路中的光束对准设计

光链路是无线局域网安全性的物理层解决方案,其中上行链路为便携式终端设备设计的LD发射光束对准是关键问题之一。在Windows平台上展示了一种自动对准的软硬件控制方法,利用Windows中的Direct Show函数库,对激光光斑与目标的对准情况进行视频信号采集和图像识别,为简化识别算法,以蓝色为特征颜色并以一个特定的形状来标识对准目标,以650 nm红色LD为识别导向光束。使之位于红外传输光束中心位置左右,通过识别LD红色光斑与蓝色目标并判断它们之间偏差的方位和距离,将数据输出到下位机控制的微型云台,在二维方位上校准光斑与目标的偏差,同时通过红外LD实现了发送端到接收目标的光信号传输。