运动人体轮廓检测算法的分析与改进

Analysis and improvement of human movement detection algorithm

在视频序列的人体运动分析中,实时分割出运动的人体,是研究的关键步骤。为了克服不均匀光照、前景运动缓慢、背景中存在摇摆的树叶等因素对检测带来的影响,提出了一种背景减除法与帧间差分相结合的运动目标检测方法。该方法首先通过基于帧差法的背景模型建立方法建立背景图像,再结合背景减除与带有权值的帧间差分检测运动目标,降低目标物体对速度和环境干扰的敏感性。最后通过形态学梯度运算操作消除外界噪声的影响。实验结果表明,本文提出的算法计算简单,对环境适应能力较强,是一种有效的运动人体检测方法。

In human motion analysis of video sequences, segmenting the motion human body in real-time is the key step. In order to overcome the influence of uneven illumination, low movement of objects in the foreground, swing leaves in the background and other factors to detection, a moving object detection algorithm based on background and the weights multiple interframe difference is presented. At first, background model is obtained by the method based on frame difference. Then, the moving object is extracted with background subtraction and multi-frame-differencing, which is insensitivity to the target object＇s speed and environment disturbance. Finally, morphological gradient operation is applied to remove the influence of outside noise. Experiment results show that the proposed algorithm has a strong ability to adapt the environment and is an effective detection method for moving the human body.