Abstract: A blackboard-based system for human detection in an indoor environment is designed and implemented, in order to accomplish human detection in wide angle camera images. The algorithm of this system excludes other nonhuman objects to detect human. Experimental results show that the blackboard-based method for human detection is feasible. It is applicable even in the presence of partial occlusion of humans, rotation of the head and variation of skin colors. It gives a feasible method for human detection in wide angle camera in indoor environments.

Key words: blackboard, software architecture, wide angle lens, human detection

摘要： 为了实现广角镜视频图像中的人体检测，设计并实现了基于黑板模式的室内人体检测系统。该系统算法通过排除非人体目标实现人体检测。实验结果表明基于黑板模式的人体检测方法有效，即使在人体局部遮挡、头部旋转和皮肤颜色差异的情况下也同样适用。为使用广角镜摄像机进行室内人体检测提供了一种可行的方法。

关键词: 黑板, 软件体系结构, 广角镜, 人体检测

CLC Number: 

• TP18