Image restoration of compressed image using classified vector quantization

To reduce communication bandwidth or storage space, image compression is needed. However, the subjective quality of compressed images may be unacceptable and the improvement of quality for compressed images may be desirable. This paper extends and modifies classified vector quantization (CVQ) to improve the quality of compressed images. The process consists of two phases: the encoding phase and the decoding phase. The encoding procedure needs a codebook for the encoder, which transforms a compressed image to a set of codeword-indices. The decoding phase also requires a different codebook for the decoder, which enhances a compressed image from a set of codeword-indices. Using CVQ to improve a compressed image's quality is different from the existing algorithm, which cannot reconstruct the high frequency components for compressed images. The experimental results show that the image quality is improved dramatically. For images in the training set, the improvement of PSNR is about 3 dB. For images, which are outside the training set, the improvement of PSNR is about 0.57 dB, which is comparable to the existing method.     

小波子带最优方向性选择和压缩感知的图像编解码

传统贝叶斯压缩感知方法所存在的局部优化问题导致了重构图像存在着较大的误差。本文提出一种基于小波自适应最优方向选择和压缩感知的图像编码方法,该方法依据图像小波变换的低频子带与高频子带所存在的相关性差异,对低频系数进行基于小波最优方向选择的编解码,以保证图像主要信息的重构质量;而针对高频子带的稀疏特性进行基于压缩感知的随机测量,以保证图像的编码效率。实验结果表明,与传统的压缩感知算法相比,本文所提出算法的解码图像具有更好的主观质量,同时在相同码率下解码图像的PSNR平均提高1~2dB。     

基于深度学习的恶劣战场环境图像恢复方法

为实现恶劣战场环境下降质图像的有效恢复、降低环境因素对战场态势感知的干扰,构建一种全新的、端到端的图像恢复方法 —–门控采样网络(GSNet).该网络以编码块-解码块为基本架构,以CNNs与门控卷积为编码与解码机制,以压缩和激励网络为编码块与解码块的连接机制,以高阶信息重要程度的重标定区分目标与背景特征,以通道粒度因子压缩方法为轻量化策略,实现对战场恶劣环境图像的快速恢复.相关实验结果表明, GSNet模型可使PSNR达到19.35 dB,并且SSIM达到0.724,无论是客观指标评价还是主观视觉效果,性能均优于对比的主流图像恢复算法;轻量级GSNet模型在较小提升PSNR、SSIM等指标的情况下,其参数量、FLOPs以及单张图像处理时间分别降低56.6%、54.6%和55.56%.     

一种保护图象轮廓细节的自适应亚抽样和插值方法

为在图象编码时减少原始图象数据,以提高压缩比和编码速度,提出了一种新的自适应亚抽样与插值的方法及可应用于多类分块图象编码的算法,该方法首先将图象划分成互不重叠的块,然后计算每个块的水平梯度与坚直梯度,再根据图象块的两个方向梯度值,将图象块分为平块、水平轮廊块、竖直轮廊块和高细节块等４类,同时对每类图象块采用不同的亚抽样与插值方法以减少原始图象数据。模拟结果表明：相对于其它的亚抽样与插值方法,该算法能够很好地保护图象中的轮廊及细节信息,从而极大地提高了重建图象的质量,尤其是图象的主观质量。另外,对于细节较多的图象,该算法在保持相当压缩比的同时,ＰＳＮＲ也提高了３．９ｄＢ;而对于细节较少的图象,压缩比与ＰＳＮＲ都略有提高。     

Random Accessible Mesh Compression Using Mesh Chartification

Previous mesh compression techniques provide decent properties such as high compression ratio, progressive decoding, and out-of-core processing. However, only a few of them supports the random accessibility in decoding, which enables the details of any specific part to be available without decoding other parts. This paper proposes an effective framework for the random accessibility of mesh compression. The key component of the framework is a wire-net mesh constructed from a chartification of the given mesh. Charts are compressed separately for random access to mesh parts and a wire-net mesh provides an indexing and stitching structure for the compressed charts. Experimental results show that random accessibility can be achieved with competent compression ratio, which is only a little worse than single-rate and comparable to progressive encoding. To demonstrate the merits of the framework, we apply it to process huge meshes in an out-of-core manner, such as out-of-core rendering and out-of-core editing.     

Foveated video quality assessment

Most image and video compression algorithms that have been proposed to improve picture quality relative to compression efficiency have either been designed based on objective criteria such as signal-to-noise-ratio (SNR) or have been evaluated, post-design, against competing methods using an objective sample measure. However, existing quantitative design criteria and numerical measurements of image and video quality both fail to adequately capture those attributes deemed important by the human visual system, except, perhaps, at very low error rates. We present a framework for assessing the quality of and determining the efficiency of foveated and compressed images and video streams. Image foveation is a process of nonuniform sampling that accords with the acquisition of visual information at the human retina. Foveated image/video compression algorithms seek to exploit this reduction of sensed information by nonuniformly reducing the resolution of the visual data. We develop unique algorithms for assessing the quality of foveated image/video data using a model of human visual response. We demonstrate these concepts on foveated, compressed video streams using modified (foveated) versions of H.263 that are standard-compliant. We rind that quality vs. compression is enhanced considerably by the foveation approach     

基于CS理论的嵌入式监控系统视频编码

将压缩感知理论应用于视频编解码中,给出了视频图像帧内和帧间编码的CS采样图像编码与重构新方法.通过仿真实验表明在图像编解码系统中,与传统视频编码方法相比,结合CS理论的方法能得到更高质量的重构图像,PSNR有较大提高.     

面向水下图像的质量评价方法

目的 针对目前水下图像质量评价方法少和现有方法存在局限性等问题,提出一种无参考并且无需手工设计特征的水下图像质量评价方法。方法 提出的水下图像质量评价方法将深度学习网络框架与随机森林回归模型相结合,首先采用深度神经网络提取水下图像的特征;然后使用提取的特征和标定的水下图像质量分数训练回归模型;最终,利用训练好的回归模型预测水下图像的质量。结果 在本文收集的水下图像数据集和水下图像清晰化算法处理结果上评测本文方法,并与多种质量评价方法进行比较,其中包括预测结果与主观质量分数比较、水下图像清晰化结果评测比较、预测结果与主观质量分数相关性比较、鲁棒性比较等。主观实验结果表明本文的评价方法可以相对准确地给出符合人类视觉感知的水下图像质量分数,并且具有更好的鲁棒性。定量实验结果表明本文方法与其他方法相比,预测的图像质量分数与主观分数具有更高的相关性。结论 提出的水下图像质量评价方法无需参考图像,省去了手工设计的特征,充分利用了深度学习网络的学习和表征能力。本文方法的准确性较好,普适性和鲁棒性较高,预测的质量分数与人类视觉感知具有较高的一致性。本方法适用于原始的水下图像和水下图像清晰化算法的处理结果。     

滑动平均和改进权重函数的快速非局部平均图像去噪算法

非局部平均算法(NL-means)是一种有效的高斯噪声去除方法,由于其实现时效率低下,很难应用到实际中。针对非局部平均算法的低效率问题,提出一种快速的非局部平均去噪算法(FNLM)。首先,为了实现对算法的加速,采用滑动平均和权重对称技术。其次,算法在加速时一般会影响到去噪效果,为了使算法加速的同时保证去噪效果,提出一种改进的权重计算函数。最后,对新算法进行了一定的实验分析,实验结果显示提出的快速算法FNLM与原始的非局部平均算法相比,效率得到了很大提升,与其他的经典算法相比,在效率和效果上都非常有竞争力。     

快速去块效应的线性规划方法

许多现有的图像压缩算法在高压缩比下会产生恼人的块效应,消除块效应的后处理方法一直以来都是图像处理领域的重要研究方向。消除块效应可以认为是从不准确的采样数据出发,尽可能恢复原始图像,这也是压缩传感理论所做的。因此利用压缩传感理论,给出了一种新的去块效应方法,将去块效应问题归结为一个无需调校任何其他参数的线性规划问题,最终采用GPU实现,得以快速求解。大量的实验结果表明,该方法能快速有效地去除块效应,改善了图像的视觉效果,同时提高了图像的PSNR。     

Perceptual quality assessment of SAR image compression

Image quality assessment (IQA) is of fundamental importance for image compression applications. Traditional IQA measures for synthetic aperture radar (SAR) image compression do not consider the properties of the human visual system (HVS). Since human beings are still the final users in most SAR image applications, the objective evaluation coordinate to human perception is the most acceptable and practical IQA method. In this article, we carried out an extensive subjective quality assessment experiment in which four different SAR image compression techniques and a total of 300 test images were estimated by 100 human observers. The test result provides a straightforward comparison of these four compression methods. Then the performances of seven classical objective IQA algorithms were validated based on obtained subjective evaluations. Moreover, we propose a novel objective approach by taking into account the HVS and SAR image characteristics. Experimental results demonstrate that the proposed metric correlates well with human perception for SAR image compression.     

一种基于SPIHT算法的感兴趣区域编码新方法*

在研究JPEG2000中的感兴趣区域(ROI)编码算法优缺点的基础上,结合嵌入式编码理论和人眼视觉特性,利用等级树分集(SPIHT)编码算法的特性,提出了一种高效的ROI编码方法,压缩后的码流具有嵌入式特点,支持渐进传输.实验结果表明:尽管在低码率下,该算法整体峰值信噪比较低,但主观视觉效果明显好于SPIHT算法,结果令人满意.     

视频质量评价方法综述

论述了视频质量的主观和客观评价方法,重点阐述了视频质量的客观评价方法.对目前常用的视频质量客观评价指标峰值信噪比和均方误差进行了分析,并指出其存在的问题;将客观评价方法归结为基于人眼视觉系统生理特征和基于结构失真的两大类,并结合仿真实验予以评述;最后对视频质量评价技术的发展进行了预测和展望.     

面向立体视频的自适应数字水印算法

鉴于立体视频是未来视频技术的一个重要发展方向,对其产品的知识产权进行保护需要提前解决,以避免盗版和侵权现象。为此,在前期数字水印和立体视频研究工作的基础上,通过详细分析立体视频的特点后,给出了一种面向立体视频的自适应数字水印新算法。新算法是将水印信息添加到增强层码流的视差矢量数据上,这种水印信息的嵌入不仅对基本码流没有影响,而且不会影响到基本码流重建后的视频图像质量。该水印算法采用自适应策略,使得嵌入水印后的图像质量损失达到最小。其水印的嵌入和提取操作均基于压缩域进行,无需视频码流的完全解码,而且水印的提取使用盲检法,无需视频原图像。整个算法简单、快速、高效。通过不同测试序列的仿真实验结果显示,水印的嵌入对原图像影响很小,使用自适应算法进行水印嵌入后图像的峰值信噪比下降值均小于非自适应算法。     

基于PCA硬阈值收缩的平滑投影Landweber图像压缩感知重构

利用压缩感知理论对图像进行测量和重构时,基于分块思想可有效提高重构速度,但同时会带来较强的块效应.为了解决该问题,在编码端提出了一种基于边缘检测的自适应分块压缩感知测量方案;在解码端提出了一种基于主成分分析(PCA)的平滑投影Landweber(SPL)重构法,该算法运用PCA训练出适合于图像结构的稀疏字典,用于进行硬阈值收缩,从而有效消除了块效应,提升了重构图像的质量.为了提高硬阈值收缩效率和减少训练复杂度,采用了3种基于块的PCA硬阈值收缩方案:全局PCA、局部PCA和分层PCA.仿真实验结果表明:所提出的自适应压缩感知测量方案与SPL重构法相结合,和传统分块压缩感知方案相比,峰值信噪比(PSNR)值均提升了1～3 dB;本文算法,无论在传统分块压缩感知方案下还是在自适应分块压缩感知方案下,与基于方向小波阈值收缩的SPL重构算法相比,均获得了更高的PSNR值.     

基于改进StOMP算法图像压缩感知重构*

在图像压缩感知重建中,一些算法能够取得好的重构效果,但耗时较长;一些算法耗时较短,但又不能取得精确解。针对重构效果和耗时不能兼得的问题,本文基于小波域稀疏,选用常规观测矩阵进行观测采样,通过对观测结果预定义滤波、选取信号硬阈值,引入共轭梯度下降算法,对分段正交匹配追踪（StOMP）重建算法进行改进。提出重建图像的边缘相似度概念,并对不同压缩比下的观测信号重建进行实验仿真。结果表明,相对于改进前StOMP算法,改进后StOMP算法在迭代收敛时间较短的情况下,重构效果提升。在主观评价上,重建图像噪声点明显减少;客观评价上,PSNR值提高,达到预期效果。     

基于U-Net的多尺度低照度图像增强网络

低照度是夜晚拍摄时常见的一种现象,不充分的光照会使图像细节损失严重,降低图像视觉质量。针对现有低照度图像增强方法对不同尺度特征的感知和表达能力存在不足的问题,提出一种基于U-Net的多尺度低照度图像增强网络（MSU-LIIEN）。采用特征金字塔作为基本处理框架,实现对低照度图像的特征提取。在特征金字塔构建的3个分支结构中均使用U-Net作为骨干网,对提取到的浅层图像特征进行编码与解码操作,同时引入结构细节残差融合块以增强网络模型提取和表征低照度图像特征信息的能力。在此基础上,对提取到的特征信息逐层融合,恢复正常光照图像。实验结果表明,MSU-LIIEN在LOL-datasets和Brighting Train数据集中相比于性能排名第二的KinD模型,平均峰值信噪比分别提高16.21%和46.67%,且在主观视野感受和客观评价指标方面均优于所有对比的经典模型,不但能有效提升低照度图像的整体亮度,而且能很好地保持图像中的细节信息和清晰的物体边缘轮廓,使增强后的图像整体画面真实自然。     

Colorization in YCbCr color space and its application to JPEG images

This paper presents a colorization method in YCbCr color space, which is based on the maximum a posteriori estimation of a color image given a monochrome image as is our previous method in RGB color space. The presented method in YCbCr space is much simpler than that in RGB space and requires much less computation time, while both methods in YCbCr and RGB space produce color images with comparable PSNR values. The proposed colorization in YCbCr is applied to JPEG compressed color images aiming at better recovery of downsampled chrominance planes. Experimental results show that colorization in YCbCr is usually effective for quality improvement of JPEG color images.     

多层感知分解的全参考图像质量评估

目的 图像质量评估是计算机视觉、图像处理等领域的基础研究课题之一,传统评估方法常基于图像低层视觉特征而忽略了高层语义信息,这也在一定程度上影响了客观指标和主观视觉质量的一致性。近年来,感知损失被广泛应用于图像风格化、图像复原等研究中,通过使用预训练的深度网络对图像进行多层语义分解,在相关问题上取得了较好的效果。受感知损失启发,提出一种多层感知分解的全参考图像质量评估方法。方法 首先使用预训练的深度网络对图像进行多层语义分解,获取多层特征图,再计算失真图像与参考图像之间的相似度,以及它们的不同层级特征图之间的相似度,最终得出兼顾了高层语义信息的图像质量分数。结果 针对传统方法PSNR（peak signal-to-noise ratio）、SSIM（structure similarity）、MS-SSIM（multi-scale structure similarity）及FSIM（feature similarity）进行实验,结果表明,本文方法能够有效提升传统图像质量评估方法的性能,在SRCC（Spearman rank order correlation coefficient）、KRCC（Kendall rank order correlation coefficient）、PLCC（Pearson linear correlation coefficient）和RMSE（root mean squared error）客观指标上均有相应提升。通过使用本文框架,PSNR、SSIM、MS-SSIM、FSIM方法在TID2013数据库上SRCC指标分别获得0.02、0.07、0.06和0.04的提升。结论 本文提出的一种多层感知分解的全参考图像质量评估方法,结合传统方法与深度学习方法,兼顾了图像低层视觉特征和高层语义信息,从而有效地提升了传统方法的评估性能,使客观评估结果更加符合主观视觉感受,同时,本文提出的评估框架能够适用于多种传统方法的性能提升。     

A new method for evaluating the subjective image quality of photographs: dynamic reference

The Dynamic Reference (DR) method has been developed for subjective image quality experiments in which original or undistorted images are unavailable. The DR method creates reference image series from test images. Reference images are presented to observers as a slide show prior to evaluating their quality. As the observers view the set of reference images, they determine the overall variation in quality within the set of test images. This study compared the performance of the DR method to that of the standardized absolute category rating (ACR) and paired comparison (PC) methods. We measured the performance of each method in terms of time effort and discriminability. The results showed that the DR method is faster than the PC method and more accurate than the ACR method. The DR method is especially suitable for experiments that require highly accurate results in a short time.