基于自适应背景模型的步态检测与识别

Gait Detection and Recognition Based on Adaptive Background Model

提出了一种基于自适应背景模型的快速背景提取及步态检测的方法,并且利用傅里叶描绘子对步态轮廓图像进行描述,进行数据维数压缩,得到匹配模板,用最近邻法进行分类识别。该算法在NLPR步态数据库上取得了较高的识别率。

This paper proposes a new method of fast extracting the background and detecting the gait based on an adaptive background model, at the same time, we adopt fourier descriptor to describe gait contour, to reduce data dimension, and to create templates. The recognition is achieved by nearest algorithm. By utilizing the algorithm,the experiments made on NLPR database have achieved comparatively high correction identification ratio.