基于多帧间的差的视频对象提取方法及其在DSP上的实现  

Video object extraction and its DSP realization based on multi-frame difference

作　　者：吴晓岚[1] 王世刚[1] 纪腾飞[1] 李强[1] 

机构地区：[1]吉林大学通信工程学院,长春130012

出　　处：《吉林大学学报（工学版）》2008年第1期219-223,共5页Journal of Jilin University:Engineering and Technology Edition

基　　金：国家自然科学基金项目(60572068);吉林省科技发展计划项目(200609)

摘　　要：针对自动提取方法及半自动提取方法的缺点,提出了一种在Blackfin-533 DSP(Digital Signal Processor)芯片上实现的视频对象提取方法。该方法是利用图像多帧差累积的高阶统计量来提取出视频对象,设计了DSP实现系统并对算法进行了测试。测试结果表明,该算法可以满足当前多种视频图像处理实时应用需求,具有良好的应用前景。With the development of computer and multimedia technology, the study of extracting of video object from video sequence is becoming a hot issue in the image processing. This paper presents a video object extraction method based on chip of Blackfin-533 DSP （Digital Signal Processor）. The high order statistics of multi-frame difference accumulation are used in the video object extraction. The algorithm makes full use of the DSP （Digital Signal Processor）high-performance in the data processing. The experiments have testified that the algorithm can meet many application needs in the field of image processing and has good application prospects.

关 键 词：通信技术 数字信号处理器 视频序列 视频对象平面 视频对象提取 

分 类 号：TN919.8[电子电信—通信与信息系统]