基于MHI与LBP表情识别的辅助异常行为识别

Auxiliary Abnormal Behavior Recognition Based on MHI and LBP Expression Recognition

为了进一步提高家居环境中异常行为的在线识别能力,将运动历史图像与表情识别相结合,提出了一种基于MHI(Motion History Image)与LBP(Local Binary Pattern)表情识别的辅助异常行为识别方法。该方法通过运动历史图像辅助识别运动轨迹,同时利用LBP提取人脸特征,进行面部表情识别,进而辅助识别异常行为状态。实验结果表明,本文方法对异常行为的识别率有一定的提升。

In order to further improve the online recognition ability of abnormal behavior in the home environment,a method of auxiliary abnormal behavior recognition based on MHI(Motion History Image)and LBP(Local Binary Pattern)expression recognition is proposed by combining motion history images with expression recognition.This method uses motion history images to assist in recognizing motion tracks,and LBP to extract facial features for facial expression recognition,thus assisting in identifying abnormal behavior states.The experimental results show that this method can improve the recognition rate of abnormal behavior.