边缘指导的BDCT压缩图像的后处理算法

基于BDCT的压缩图像由于频域的量化使重建回的图像出现“块失真”和“振铃”效应,严重影响了压缩图像的质量,提出一种针对这两种失真分别进行处理的后处理算法．首先,根据压缩图像块的属性把像素块分为不同的种类,针对不同的块采用3种不同的模式来去除“块失真”;然后,在边缘检测结果的指导下,对不同的区域采用不同的滤波器进行滤波,来消除“振铃”．大量的实验结果表明,该算法在主观视觉效果和客观标准评上都取得了良好的效果．     

一种空间域自适应去块效应算法

基于块离散余弦变换的图像压缩编码,在低比特率时其重构图像的块边界上会产生严重的方块效应。提出了一种空间域的自适应去块效应算法。首先把块效应建模成阶梯函数,然后根据判断准则把图像块分成平滑块,纹理块和边缘块。对平滑块,首次导出了块效应强度的表达式,然后根据块效应强度和平滑度进行自适应平滑滤波;对边缘块和纹理块,采用简单的Sigma滤波器对块边界周围的像素滤波。大量的仿真实验结果表明提出的算法有优越的去块效应性能。     

加强边缘感知的盲去模糊算法

目的 拍摄过程中,如果摄像机进行了错误的聚焦,就会得到模糊的图像,如何将模糊图像变得清晰成为一个亟待解决的问题。目前关于图像的去模糊方法多采用基于模糊核约束的卷积模型。但是由于实际应用中很难准确获取模糊核的信息,同时计算机也存在精度限制,计算结果与实际物理模型有偏差,因而去模糊的主要挑战为：如何精确地估计模糊核,以及如何在复原过程中减弱由于精度限制造成的振铃效应。方法 振铃效应是指图像的灰度剧烈变化处产生的震荡,类似于钟被敲击后产生的波状空气震荡。在图像复原过程中,此效应通常发生在梯度变化较大的边缘区域附近。本文对此进行研究,在去模糊过程中引入边缘信息作为约束条件,以改善模糊核的估计,并通过抑制边缘区域的反卷积,抑制图像复原过程中的振铃效应。算法主要分为如下3个部分：1）设计了适用于模糊图像的边缘提取算法;2）利用边缘信息设计了加强边缘感知的反卷积算法;3）提出并设计了安全检测子,以保证算法在边缘区域复原的完整性。结果 实验结果表明,在没有先验知识的情况下,本文方法可以较好地恢复图像细节,并有效抑制振铃效应。较之传统的去模糊处理算法,本文方法在性能上有较大提高。比如,相比于Chan、Krishnan以及Hu的方法,本文方法在峰值信噪比指标上分别提高了25.73%、3.52%和4.43%,在结构相似性指标上分别提高了7.67%、1.63%和3.59%。同时,与基于深度学习的方法相比,本文方法不依赖于数据集,鲁棒性更强。结论 本文方法可以较好地恢复图像细节,并抑制振铃效应,同时比深度学习方法适用范围更广。     

一种有效的块DCT编码图像去块效应算法

块离散余弦变换在低比特率时其恢复图像的块边界上会出现明显可见的方块效应,从而降低了图像的视觉质量。提出一种基于小波域特征分析的去块效应算法,目的是尽量消除块效应的同时充分保护图像的边缘信息。图像空域中的块效应,在小波域高频子带表现为平行线效应、竖直线效应或者网格效应。通过为各个子带设置自适应操作算子去抑制块边界系数能量聚集现象,使得空域中的块效应得到消除。仿真结果表明在不同比特率下,对不同类型的图像,新算法都能得到较好的去块效应效果,并且具有较高的运算速度。     

一种基于图象小波极值表示的减少方块效应的算法

针对低比特率JPEG压缩图象会产生严重的方块效应问题和根据后处理技术可有效地减少方块效应的认识,提出了一种新的基于图象小波极值描述的后处理算法。该算法根据解压缩图象方块效应在一级和二级小波变换系数中的极值特性,对位于图象背景区、平滑边缘区及阶梯状边缘区的方块边界的小波极值分别加以处理。实验结果表明,该算法能有效地减少方块效应,且能改善译码图象的信噪比和主观视觉质量。     

盲复原图像振铃效应评价

线性空不变盲复原算法通常会在较为明显的边缘处产生振铃效应。振铃现象主要受噪声,复原算法种类,以及复原算法参数的选择的影响。提出了一种盲复原图像振铃效应评价方法。方法根据振铃效应的不同类型,使用Gabor滤波器,共生向量等方法分别对其进行评价,最后提出整体的振铃效应评价方法。实验结果表明,该方法可以有效地评价不同复原算法和不同复原参数下的复原图像中的振铃效应,评价结果符合主观评价结果。     

Support Vector Machine (SVM) based compression artifact‐reduction technique

Abstract— A compression artifact‐reduction algorithm based on support vector regression is proposed. The algorithm belongs to a broad family of standard reconstruction methods, but a standardization model is determined from a set of training samples of original images and the corresponding noise‐corrupted version. As opposed to artifact‐reduction methods specific to each type of compression artifact (e.g., blocking, ringing, etc.), we treat such artifacts as a manifestation of the same problem, which is the quantization of DCT coefficients. In the testing step, the algorithm tries to undo the effect of quantization by using the relationship between the original and artifact‐corrupted image, determined during the training step. Experimental results exhibit significant reduction in all types of compression artifacts.     

基于DCT域的高压缩图像去块效应算法

块离散余弦变换(B lock D iscrete Cosine Transform)的主要缺点是在低比特率时其恢复图像的块边界上会出现明显可见的方块效应,降低了图像的视觉质量.为了尽量消除块效应并保护图像的边缘信息提出了一种基于DCT域的块效应消除算法.该算法充分利用了人类视觉系统(Hum an V isual System)特性,建立了块效应模型并给出一个简便的检测边缘标准,对于平滑区,对影响块效应的参数进行修正,然后用线性函数块代替阶跃函数块去除块效应,最后再对更新块和纹理区在DCT域中进行后滤波.仿真结果验证了本文算法的有效性.     

利用块效应特征的JPEG图像盲取证研究进展

数字图像的盲取证技术由于不依赖任何预嵌入的信息来鉴别图像真实性和完整性的优势,正逐步成为数字媒体安全领域新的研究热点。由于JPEG图像是目前最流行的图像格式,并且块效应是JPEG图像与生俱来的本质特征,因此如何更加有效地利用块效应特征对JPEG图像的真伪进行盲取证研究具有非常重要的现实意义和应用价值。首先对目前国内外利用JPEG图像编码特性的盲取证方法进行归类分析;然后重点针对利用块效应特征的JPEG图像盲取证技术展开讨论,详细介绍并总结了基于块效应测度和基于块效应网格提取的两类盲取证算法的核心思想和局限性;最后提出了存在的问题及未来的研究方向。     

基于对象的多级图像增强法

针对图像局域增强时出现的噪声过增强和环块状伪轮廓问题,提出了基于对象的多级对比度拉伸图像增强法。首先采用形态学分水岭及区域合并法对图像进行分割,得到图像的构成对象;然后在对象之间采用相邻极点间拉伸法增大对象间灰度动态范围,在对象内部采用线形拉伸法增强对象纹理并保持对象形态。实验结果表明,该方法在增强图像结构的同时,能够有效避免环块状伪轮廓,抑制平滑区域噪声过增强,保持图像原始整体亮度,使增强后的图像具有自然的外观。     

Deep learning-based compressed image artifacts reduction based on multi-scale image fusion

One of the visually noticeable compression artifacts in block-based image/video compression platforms is called blocking artifact. Several post-processing methods were presented to reduce such kind of artifacts. However, most methods in the literature often induce visibly blurring artifacts. The paper presents a deep network to eliminate image compression artifacts (usually denoted by image deblocking) based on image fusion in multi-scale manner. Recent deep learning-based related methods usually learn deep models using a loss function in per-pixel manner based on explicit image priors in order to directly produce clean images. In place of existing deep learning-guided approaches, the problem is reformulated in this paper to the learning of the residuals (or artifacts) between the received images and their corresponding clean images (ground truths). In the presented deep framework, an input image is first down-sampled while naturally reducing the blocking artifacts. Then, our multi-scale image fusion model is used for fusing the different down-scaled versions (of less artifacts) with the input image (with severer artifacts) to estimate the blocking artifacts. Then, by deducting the estimated artifacts from the input image, the blocking artifacts can be significantly eliminated and most original image details are preserved simultaneously. The presented method is well applicable to any vision-based computer systems with digital visual codec embedded.     

基于复数小波阈值和w-TV的去块效应算法

块离散余弦变换在低比特率时其恢复图像的块边界上会出现明显可见的方块效应,降低了图像的视觉质量。针对该问题提出一种基于复数小波阈值和加权全变差（w-TV）的块效应消除算法,利用复数小波阈值和加权全变差的特性对图像进行处理。实验证明,该算法较好地去除了块效应,尽可能保留了图像的特征结构,具有良好的视觉效果。     

Block Truncation Coding based effective watermarking scheme for image authentication with recovery capability

Due to the advancement of photo-editing software, powerful computers and high resolution capturing devices, it has become tough to prevent the digital image from tampering. So, in these days just by looking a digital image we cannot say whether it is a genuine or not. This is why digital image authentication, as well as restoration, has become the essential issues, especially when it is utilized in medical science, evidence of court, and forensic science. This paper proposes an effective self-embedding fragile watermarking technique for the digital image authentication as well as recovery. The watermark is generated by quantization, and block truncation coding (BTC) of each 2 × 2 non-overlapping block and embedded in three least significant bits (LSBs) of the corresponding mapped block. The recovery bits are derived from most significant bits (MSBs) of the host image, and the authentication bits are derived from recovery bits, the spatial location of pixels and watermark keys. Even if tempering rate is 50%, the reconstruction of tampered image is achieved with high peak signal-to-noise ratio (PSNR) and normalized correlation coefficient (NCC). The experimental results demonstrate that the proposed scheme not only outperforms high-quality recovery fidelity but also negotiate the blocking artifacts additionally it improves the accuracy of tamper localization due to the use of very small size blocks.

     

基于JPEG块效应频域特性的合成图像检测

JPEG图像在压缩过程中所产生的块效应在功率谱曲线上体现为周期性波峰,而篡改JPEG图像所造成块效应不一致将导致周期性波峰的衰弱或消除。利用上述原理,提出了一种基于JPEG块效应频域特性的合成图像检测算法。算法对待测图像进行去噪,提取包含块效应的噪声,对其进行重叠分块并求得每块的块效应度量值,依据该度量值检测并定位篡改区域。实验结果表明,相对于传统的基于块效应不一致的算法,能够更好地检测多种不同图像格式的合成和篡改区域较小等情况。     

Multi-focus image fusion using dictionary-based sparse representation

Multi-focus image fusion has emerged as a major topic in image processing to generate all-focus images with increased depth-of-field from multi-focus photographs. Different approaches have been used in spatial or transform domain for this purpose. But most of them are subject to one or more of image fusion quality degradations such as blocking artifacts, ringing effects, artificial edges, halo artifacts, contrast decrease, sharpness reduction, and misalignment of decision map with object boundaries. In this paper we present a novel multi-focus image fusion method in spatial domain that utilizes a dictionary which is learned from local patches of source images. Sparse representation of relative sharpness measure over this trained dictionary are pooled together to get the corresponding pooled features. Correlation of the pooled features with sparse representations of input images produces a pixel level score for decision map of fusion. Final regularized decision map is obtained using Markov Random Field (MRF) optimization. We also gathered a new color multi-focus image dataset which has more variety than traditional multi-focus image sets. Experimental results demonstrate that our proposed method outperforms existing state-of-the-art methods, in terms of visual and quantitative evaluations.     

基于小波变换的方块效应去除法

低比特率JPEG压缩图像会产生严重方块效应，即马赛克现象，后处理技术可以有效地减少该效应。针对压缩图像在小波域中有结构信息和块不连续性的不同表现，提出了一种基于小波变换的分析方法，对图像在小波域中的不同响应分别加以处理。该算法根据小波系数在不同范围内的不同特性对图像进行区域分类，进而对不同区域做出不同的处理。试验结果表明，该方法能有效地减少方块效应，提高解码端图像的视觉质量。     

边导向的双三次彩色图像插值

提出了一个新的边导向的双三次卷积(Cubic convolution, CC)彩色图像插值算法. 对于待插值的像素, 首先在其邻域检测两个正交方向边的强度. 如果该 像素在一个强边上, 则沿着强边的方向执行CC插值估计该像素;否则, 该像素 在弱边或纹理区域, 通过加权平均两个正交方向的CC插值估计该像素. 本文方法也考虑了彩色平面之间的相关性. 实验结果显示, 本文方法显著优于经 典的CC插值和其他一些插值方法.     

基于分块的小波多聚焦图像融合方法

提出了一种基于分块的小波多聚焦图像融合方法。并采用均方根误差对该方法进行了评价,实验结果表明,该疗法其钉非常好的融合效果。其融合性能优于对图像不作小波分解而直接进行分块融合的方法和仅作小波分解而不进行分块的融合方法。与已有融合疗法相比,能消除块痕迹,且能节约运算量。     

Geo-parcel-based crop classification in very-high-resolution images via hierarchical perception

ABSTRACT

The basic application of remote sensing is classifying surface objects in images. Traditional pixel-based or object-based classification methods are poorly suited to very high-resolution (VHR) images captured by remote sensors with high spatial resolutions. In the field of computer vision, deep learning has recently achieved great advances in natural image processing. Inspired by this, we propose a methodology guided by hierarchical perception to classify crops in VHR images based on geo-parcels. Geo-parcel-based crop classification is used in agriculture and in refined farmland management. The proposed methodology can be divided into three steps: zoning, location and quality. In the first step, the image is divided into blocks based on the road network. In the second step, geographical entities are extracted from every block defined in the zoning step. In the last step, the geographical entity types are identified based on the texture information. These steps provide mutual constraints. In each step, the information is extracted by neural networks that have been adapted to the VHR images. The experimental results indicate that our methodology performs well, with a precision greater than 90%. Furthermore, our methodology combines deep learning techniques and theory regarding image perception by humans, providing a valuable method for processing remote sensing information.     

基于JPEG双量化效应的图像篡改盲检测

JPEG图像的双量化效应是检测和发现JPEG图像篡改的重要线索。针对现有检测算法大多数是基于DCT块效应而较少利用双量化效应的情况,提出了一种利用JPEG双量化效应的图像篡改盲检测新方案。该方案对比用DCT系数直方图计算的未篡改区域后验概率和用区间长度计算的未篡改区域后验概率之间的差异性,提取能有效区分篡改块和未篡改块的特征,计算每个DCT块的特征值,然后设置阈值,将特征值大于阈值的图像块判定为篡改块,最后通过选取连通区域,标定篡改区域。实验结果表明,与已有的类似方案相比,该方案能够较精确地检测和定位篡改区域,且对颜色、纹理丰富的图像具有明显优势。