人体行为识别的改进Radon变换方法

Human activity recognition based on improved Radon transforms

提出一种新的人体行为识别特征提取方法。针对Radon变换对缩放敏感的问题,采用改进的Radon变换提取运动人体区域最小外接矩形的Radon变换特征,并采用隐马尔可夫模型进行行为识别。该方法提取特征时不再需要进行规范化处理,提高了特征的鲁棒性。实验结果表明,该方法对噪声不敏感、计算简单、识别效率高。

A new method of feature extraction for human behavior identification is proposed.In order to solve the scaling sensitivity of Radon transformation,an improved Radon transformation is used to extract Radon features of minimum enclosing rectangle for motion human and hidden markov Model is used to recognize human activities.The normalization processing is no needed when proposed method is used to extract the features.The robustness of features is improved.The experimental results show that the proposed method is insensitive to noise,low computation and higher recognition rate.