基于改进方向纹理谱特征的图像检索

在定义图像局部邻域纹理方向特性的基础上,提出了一种新的方向纹理谱描述符。该描述符针对局部邻域内中心像素与其相对的邻域像素,既充分考虑了它们间的灰度变化关系,又考虑了它们间灰度差异的变化关系,从而更有效地描述了局部纹理特征。为证明新描述符的分辨能力,采用4种不同图像库进行图像检索对比实验,结果表明,本文的新纹理谱描述符取得了最好的检索效果。     

HOS运动目标分割算法在视频监控中的应用

为了提高视频监控中运动目标分割的速度和准确度,研究并实现了一种基于高阶统计量HOS(HigherOrder Statistics)的分割算法.首先根据HOS假设检验处理帧差图,判定像素点是否属于运动区域,阈值通过灰度共生矩阵获得,考虑了背景纹理的慢变化.然后,用矩形框聚类法大致确定运动目标的范围,在该范围内使用形态运算法和首尾扫描法去除空洞.最后,使用模板相与法获得帧图像的运动目标模板,从原图像中分割运动区域.算法采用了由粗到精的分析策略,实验表明,是一种快速稳健的算法.     

结合手势二进制编码和类-Hausdorff距离的手势识别

针对目前动态手势识别方法受手势旋转、平移、缩放的影响,并解决手势识别的实时性问题,提出一种基于手势二进制编码和类-Hausdorff距离模板匹配的手势识别方法.首先,把分割好的手势图像进行标准化处理,求出标准化图像中的手势主方向,建立二维手势直角坐标系,提取空间手势特征;其次,根据前五帧手势图像中手势像素点个数的变化量识别出动态手势类型;然后,用手势二进制描述子从动态手势类型中再筛选出可能的候选手势集合;最后,用类-Hausdorff距离模板匹配方法从候选手势集合中识别出最终手势.主要创新点在于：提出的动态手势类型识别和手势二进制描述子匹配的方法,大大缩短了动态手势识别的时间;提出的结合手势主方向的类-Hausdorff距离方法,不仅对旋转、平移和缩放手势具有不变性,而且对区分度较小的手势也具有较高的识别准确率.实验结果表明,在光照相对稳定的条件下,该方法能够实时准确的实现动态手势识别,总体识别率达到95%以上,对发生缩放的手势识别率能达到92%以上,对发生旋转的手势识别率能达到87%以上.本文算法已经在一个基于手势的人机交互界面中得到应用.     

Automatic facial expression recognition: feature extraction and selection

In this paper, we investigate feature extraction and feature selection methods as well as classification methods for automatic facial expression recognition (FER) system. The FER system is fully automatic and consists of the following modules: face detection, facial detection, feature extraction, selection of optimal features, and classification. Face detection is based on AdaBoost algorithm and is followed by the extraction of frame with the maximum intensity of emotion using the inter-frame mutual information criterion. The selected frames are then processed to generate characteristic features using different methods including: Gabor filters, log Gabor filter, local binary pattern (LBP) operator, higher-order local autocorrelation (HLAC) and a recent proposed method called HLAC-like features (HLACLF). The most informative features are selected based on both wrapper and filter feature selection methods. Experiments on several facial expression databases show comparisons of different methods.     

对帧间差分算法的脉冲光干扰研究

在对帧间差分算法进行分析的基础上,建立了基于帧间差分算法的目标探测识别系统,提出了针对该算法的脉冲光干扰方法.在红外干扰光源条件下,针对帧间差分算法特点选择合适的干扰信号脉冲频率实现了对该算法的干扰.实验中,帧间差分算法选取的差频为5Hz,当干扰源的脉冲信号频率与帧间差分算法的差频相比拟(近似相等)时,相干光源和非相干光源实现了对帧间差分算法的干扰.     

Error-resilient coding of H.264 based on periodic macroblock

For the compressed video, since an inter-frame depends on the previously encoded frame, the error in one inter-frame may propagate to the following inter-frames. In this paper, we present a new error-resilient coding scheme to alleviate the effect of error propagation for the new coding standard H.264. In this new coding standard, multiple reference frame is adopted to improve the coding efficiency. By making use of the reference frame buffer in the encoder, we can reference some macroblocks in every n/sup th/ inter-frame to the frame that is n frames interval away, and these macroblocks are named as periodic macroblocks. The periodic macroblock can efficiently alleviate the error propagation between two frames that contain periodic macroblocks. We prove it in theory that encoding selected periodic macroblocks will reduce the loss probability of pixel. The selection of periodic macroblock is based on the distortion expectation of each macroblock in every n/sup th/ frame. The number of periodic macroblocks in every n/sup th/ frame can be adjusted according to the available transmission bandwidth, as the periodic macroblock will consume little more bits. The simulation results prove that the periodic macroblocks can obviously improve the quality of video at different macroblock loss rates. When the macroblock lost rate is 15% in every frame, the PSNR of video sequence can be improved about 3dB with 5% bitrate increase.     

Robust video watermarking based on affine invariant regions in the compressed domain

This paper proposes a novel robust video watermarking scheme based on local affine invariant features in the compressed domain. This scheme is resilient to geometric distortions and quite suitable for DCT-encoded compressed video data because it performs directly in the block DCTs domain. In order to synchronize the watermark, we use local invariant feature points obtained through the Harris-Affine detector which is invariant to affine distortions. To decode the frames from DCT domain to the spatial domain as fast as possible, a fast inter-transformation between block DCTs and sub-block DCTs is employed and down-sampling frames in the spatial domain are obtained by replacing each sub-blocks DCT of 2×2 pixels with half of the corresponding DC coefficient. The above-mentioned strategy can significantly save computational cost in comparison with the conventional method which accomplishes the same task via inverse DCT (IDCT). The watermark detection is performed in spatial domain along with the decoded video playing. So it is not sensitive to the video format conversion. Experimental results demonstrate that the proposed scheme is transparent and robust to signal-processing attacks, geometric distortions including rotation, scaling, aspect ratio changes, linear geometric transforms, cropping and combinations of several attacks, frame dropping, and frame rate conversion.