摘要
目标运动的视觉分析在虚拟现实、视觉监控、感知接口中均有广泛的应用前景,文章介绍了运动目标视觉分析中基于轮廓的分析识别算法,即对每个序列而言,图象二值化后,用背景剪除法检测出人的运动轮廓,然后将这些时变的二维轮廓形状转化为对应的一维距离信号,并归一化处理,以便对有待分类的序列进行识别。
Visual analysis of human motion is driven by a wide spectrum of promising applications in many areas such as virtual reality, smart surveillance, perceptual user interface. This paper proposes a motion-based algorithm by silhouette analysis. For each image sequence, an improved background subtraction algorithm and a simple correspondence procedure are first used to segment and track the moving silhouettes of a walking figure from the background. Then eigenspace transformation is applied to reduce the dimensionality of the input feature space for the supervised pattern classification in the lower-dimensional eigenspace for recognition.
出处
《四川理工学院学报（自然科学版）》
CAS
2006年第4期30-33,共4页
Journal of Sichuan University of Science & Engineering(Natural Science Edition)
关键词
运动视觉分析
背景减除
阀值分割
轮廓提取
motion-visual analysis
background subtraction
threshold partition
figure pick-up
作者简介
唐挺（1976-），男，讲师，硕士，主要从事图像处理及分析模式的研究.
