A new texture representation approach based on local feature saliency

Although it has been studied in some depth, texture characterization is still a challenging issue for real-life applications. In this study, we propose a multiresolution salient-point-based approach in the wavelet domain. This incorporates a two-phase feature extraction scheme. In the first phase, each wavelet subband (LH, HL, or HH) is used to compute local features by using multidisciplined (statistical, geometrical, or fractal) existing texture measures. These features are converted into binary images, called salient point images (SPIs), via threshold operation. This operation is the key step in our approach because it provides an opportunity for better segmentation and combination of multiple features. In the final phase, we propose a set of new texture features, namely, salient-point density (SPD), non-salient-point density (NSPD), salient-point residual (SPR), saliency and non-saliency product (SNP), and salient-point distribution non-uniformity (SPDN). These features characterize various aspects of image texture such as fineness/coarseness, primitive distribution, internal structures, etc. These features are then applied to the well-known K-means algorithm for unsupervised segmentation of texture images. Experimental results with the standard texture (Brodatz) and natural images demonstrate the robustness and potential of the proposed features compared to the wavelet energy (WE) and local extrema density feature (LED). The text was submitted by the authors in English. Md. Khayrul Bashar was born in Chittagong, Bangladesh in 1969. He received his B.E. (1993), M.Tech. (1998), and PhD (2004) degrees from Bangladesh University of Engineering and Technology (BUET), Indian Institute of Technology (IIT) Bombay, and Nagoya University, respectively. He was a research engineer from 1995 to 1999 at Bangladesh Space Research and Remote Sensing Organization (SPARRSO) and assistant professor from 1999 to 2000 at the department of Electrical and Electronic Engineering, Chittagong University of Engineering and Technology (CUET), Bangladesh. Since 2004, he has been a research fellow in the department of Information Engineering, Nagoya University, Japan. Dr. Bashar is a member of IEEE, IEICE, BCS, and IEB. His research interest includes developing algorithms for image understanding, content-based image retrieval and web-application design, analysis and testing. Noboru Ohnishi was born in Aichi, Japan in 1951. He received his B.E., M.E., and PhD degrees in Electrical Engineering from Nagoya University in 1973, 1975, and 1984, respectively. From 1975–1986, he worked as a researcher in the Rehabilitation Engineering centre under the Ministry of Labor, Japan. In 1986, he joined as an Assistant Professor in the dept. of Electrical Engineering of Nagoya University. Currently, he is a professor of the dept. of Information Engineering at the same university. During his long professional life, he has also served as a visiting researcher (1992–1993) in the laboratory of artificial intelligence at Michigan University, and team leader (1993–2001) at the Bio-mimetic Control Research Center, RIKEN, Nagoya, Japan. He also holds many respectable positions at various professional bodies in Japan and he has published many research papers (more than 140) in various international journals. For his technical creativity and ingenuity, he was awarded SICE society prizes in 1996 and 1999. His research interest includes brain analysis, modeling, and brain support, computer vision, and audition. He is a member of IEEE, IPSJ, IEICE, IEEJ, IIITE, JNNS, SICE and RSJ. Kiyoshi Agusa received his PhD degree in computer science from Kyoto University in 1982. Currently, he is a professor of the department of Information Systems, Graduate School of Information Science, Nagoya University. His research area includes software engineering, program repository, and software reuse. Since 2003, he has been working as a team leader of a university-industry collaboration project entitled “e-Society,” which is a part of the “e-Japan” project, and doing research on reliability issues for web-based applications. He is a member of IPSJ, ISSST, IEICE, ACM and IEEE.     

稀疏表达的运动数据压缩

随着运动数据越来越多地被应用于动画制作和科研领域,高效的运动数据压缩技术也逐渐成为一个热门的研究课题.基于稀疏表达提出一种新的运动数据有损压缩方法.首先对输入的运动数据进行分析生成稀疏表达字典;然后基于稀疏表达字典对运动数据中的每一帧进行稀疏线性表达;最后用K-SVD算法对字典和稀疏表示进行迭代优化.实验结果表明,本文方法可以达到较高的压缩比(50倍左右),同时保持原始运动数据的完整性,还原后可控制重建误差在肉眼不易分辨的范围内(平均RMS误差2.0以下),并且本文方法特别适用于对较短运动数据的压缩.     

融合LBP和局部稀疏表示的人脸表情识别

为更好获取人脸局部表情特征,提出了一种融合局部二值模式（Local Binary Pattern,LBP）和局部稀疏表示的人脸表情特征与识别方法。为深入分析表情对人脸子区域的影响,根据五官特征对人脸进行非均匀分区,并提取局部LBP特征;为精细刻画人脸局部纹理,整合人脸局部特征,设计了人脸局部稀疏重构表示方法,并根据表情对各局部子区域的影响因子,加权融合局部重构残差进行人脸表情识别。在JAFFE2表情人脸库上的对比实验,验证了该方法的可行性和鲁棒性。     

局部联合结构化稀疏表示的单样本人脸识别

针对单样本问题,基于相同类别的人脸变化信息应有相似的稀疏编码这一事实,提出结构化稀疏变化字典学习方法,以得到较好的共享类内变化字典。同时鉴于同一人脸的所有区域应有相同的类标签,通过训练样本与变化字典按坐标分块联合表示查询人脸区域,然后给稀疏系数引入导致结构化稀疏效果的约束条件,实现对应类别字典的自动选择,从而更好地表示查询人脸。提出的人脸表示方法可以在局部识别方法的优势上整合全局信息,使得在AR、Extended Yale B、CMU-PIE人脸库上的表现超过其他单样本识别相关的方法,取得了较好的识别效果。     

混合表情的稀疏表达分析

提出了一种混合表情的定量描述方法。基于压缩感知的理论框架,以面部特征点的Gabor小波系数为表情特征对混合表情进行了分析;利用隶属度函数定量表示混合表情中的不同组成。实验结果表明,该方法可以简单有效地表示混合表情中各基本表情的组成。     

基于稀疏表示的脑电（EEG）情感分类

计算机对人类情绪与情感的识别研究已经成为了脑机接口领域的研究热点。通过分析人类在生活中的各种情感状态,提取脑电信号的特征并对情感状态进行识别、分类是情感智能化领域的重要方向。针对基于音乐视频诱导的情感数据集DEAP进行了研究,提取脑电信号的频域特征后,提出了采用加速近邻梯度算法（APG)和正交匹配算法（OMP）求解稀疏编码的稀疏表示分类模型进行情感分类,并与支持向量机算法（SVM）做效果比较。实验结果表明,APG算法通过L1范数正则近似求解以其快速的收敛速度在情感数据集上有着较好的分类表现,而OMP算法与SVM算法的分类效果相差无几,实现了情感脑电信号的分类。     

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.     

仿射变换虚实配准的空间信息增强表达

伴随着移动GIS和可穿戴计算的出现,GIS作业空间被拓展到户外,户外增强现实空间信息表达致力于将虚拟空间信息叠加到真实地理场景上,增强、拓展用户对地理场景的认知。为了有效地降低空间信息增强表达的虚实配准误差,汲取计算机视觉中的算法思想,提出通过建立全局仿射坐标系,解算虚拟物体到真实场景图像的投影坐标,找到获取像素相对深度参数的方法,进而避开相机定标这个复杂过程,最终为户外增强现实系统提供位置相关的信息源,也为空间信息可视化表达提供了新界面。     

基于局部特征和稀疏表示的图像目标检测算法

传统的基于局部特征的图像目标检测算法具有对遮挡和旋转敏感、检测精度不高以及运算速度慢的特点,为了改进该算法的性能,提出了一种将图像局部特征应用于稀疏表示理论的图像目标检测算法。该算法利用随机树的方式有监督地学习样本图像的局部特征形成字典,通过学习好的字典和测试图像的子块来预测图像中目标的中心位置,以此寻求待检测图像稀疏的表示,从而实现对图像中感兴趣目标的检测。实验结果表明,该算法对目标的遮挡、旋转和复杂背景有很好的鲁棒性,而且检测精度和运算速度相对于同类经典算法均有提高。     

近邻局部OMP稀疏表示图像去噪

目的 基于分类的稀疏字典去噪算法改善了字典训练阶段的效率问题,但稀疏分解阶段仍是全字典匹配,影响算法运行速度。为了解决稀疏去噪算法在稀疏分解阶段因复杂矩阵运算及字典全局搜索导致的算法效率低,以及冗余的稀疏字典因无法描述图像具体特征而影响图像去噪效果的问题,提出改进算法。方法 首先稀疏分解阶段,在原正交匹配追踪算法基础上引入字典原子聚类思想,提出局部正交匹配追踪算法,将全局搜索优化为局部搜索;为保证局部搜索仍能保持良好的匹配结果,提出近邻择优策略,计算聚类中心与信号原子的距离,从而按照某一阈值自适应地选择最优的n个子字典作为稀疏分解的匹配空间;最后将图像分解为内容簇和背景簇,对内容簇采用基于近邻的局部K奇异值分解（K-SVD）算法去噪,背景簇采用均值滤波方法去噪。结果 对USC标准数据库中大量图像进行去噪实验,本文算法去噪结果的峰值信噪比值比K-SVD算法平均提高了1.53 dB,比2维块匹配（BM3D）算法平均提高了0.72 dB,比聚类的稀疏表示去噪（CSR）算法平均提高了0.5 dB;运行时间比原算法提高了23.2%。结论 本文算法针对灰度图像去噪,在去噪效果及去噪效率方面均有改善,尤其对细节纹理较丰富的灰度图像去噪具有一定的应用价值。     

Palmprint identification using sparse and dense hybrid representation

Multimedia Tools and Applications - Among various palmprint identification methods proposed in the literature, Sparse Representation for Classification (SRC) is very attractive, offering high...     

Face hallucination with imprecise-alignment using iterative sparse representation

Existing face hallucination methods assume that the face images are well-aligned. However, in practice, given a low-resolution face image, it is very difficult to perform precise alignment. As a result, the quality of the super-resolved image is degraded dramatically. In this paper, we propose a near frontal-view face hallucination method which is robust to face image mis-alignment. Based on the discriminative nature of sparse representation, we propose a global face sparse representation model that can reconstruct images with mis-alignment variations. We further propose an iterative method combining the global sparse representation and the local linear regression using the Expectation Maximization (EM) algorithm, in which the face hallucination is converted into a parameter estimation problem with incomplete data. Since the proposed algorithm is independent of the face similarity resulting from precise alignment, the proposed algorithm is robust to mis-alignment. In addition, the proposed iterative manner not only combines the merits of the global and local face hallucination, but also provides a convenient way to integrate different strategies to handle the mis-alignment problem. Experimental results show that the proposed method achieves better performance than existing methods, especially for mis-aligned face images.     

Remote-sensing image fusion using sparse representation with sub-dictionaries

Remote-sensing image fusion aims to obtain a multispectral (MS) image with a high spatial resolution, which integrates spatial information from the panchromatic (Pan) image and with spectral information from the MS image. Sparse representation (SR) has been recently used in remote-sensing image fusion method, and can obtain superior results to many traditional methods. However, the main obstacle is that the dictionary is generated from high resolution MS images (HRMS), which are difficult to acquire. In this article, a new SR-based remote-sensing image fusion method with sub-dictionaries is proposed. The image fusion problem is transformed into a restoration problem under the observation model with the sparsity constraint, so the fused HRMS image can then be reconstructed by a trained dictionary. The proposed dictionary for image fusion is composed of several sub-dictionaries, each of which is constructed from a source Pan image and its corresponding MS images. Therefore, the dictionary can be constructed without other HRMS images. The fusion results from QuickBird and IKONOS remote-sensing images demonstrate that the proposed method gives higher spatial resolution and less spectral distortion compared with other widely used and the state-of-the-art remote-sensing image fusion methods.     

局部模板更新逆向联合稀疏表示目标跟踪算法

逆向联合稀疏表示算法可充分利用跟踪过程中的时间相似性和空间连续性,但由于遮挡、光照变化等的影响,易出现跟踪漂移.为解决上述问题,提出一种基于局部模板更新逆向联合稀疏表示目标跟踪算法,其通过逆向局部重构目标模板集完成逆向联合稀疏表示.首先,在首帧初始化目标模板集,利用粒子滤波获取候选图像,并对其分块处理,构建逆向联合稀疏...     

使用BSP和遗传算法的图像稀疏化技术

图像稀疏化技术是利用图像中稀少的且与具体应用相关的数据来表示原始图像的技术.使用BSP和遗传算法的方法在图像中生成能够近似图像的自适应的网格,即用较少的包含重要信息的像素来表示图像,实现图像的稀疏化,达到压缩之目的.该自适应网格能够以很高的质量重构出原始图像,在图像处理和计算机视觉领域有很好的应用前景.     

基于一致性稀疏表示的地震信号补全算法

针对由于记录仪器的故障和保存媒介的缺陷导致的地震信号中一部分波形出现的缺失,提出一种基于一致性稀疏表示(CSR)的地震波补全算法。首先,通过稀疏表示(SR)模型表示每一帧地震信号;随后,采用主分量分析(PCA)的方法提取在频谱的分布上的帧间一致性信息;最后,利用地震信号的每帧的稀疏性和用帧与帧之间频谱分布的一致性,对缺损的信号进行恢复。在对历史地震信号进行的仿真实验中,当缺损的部分的跨度达到帧数据量的50%时,传统的稀疏表示恢复算法的误差已经很大,而一致性稀疏表示模型仍然能够得到很好的结果。仿真实验结果表明,一致性稀疏表示模型的补全效果要远好于传统稀疏表示模型的补全效果。