基于Hilbert-Huang描述子的步态识别方法

基于Hilbert-Huang描述子的步态识别方法 Gait Recognition Based on Hilbert-Huang Descriptors 程琼 CHENG Qiong 1 2 first-author 付波 FU Bo 1 2 湖北工业大学电气与电子工程学院 湖北工业大学电气与电子工程学院 School of Electrical and Electronic Engineering, Hubei University of Technology School of Electrical and Electronic Engineering, Hubei University of Technology 湖北工业大学电气与电子工程学院,湖北,武汉,430068 湖北工业大学电气与电子工程学院,湖北,武汉,430068 提出了一种基于Hilbert-Huang描述子和K-近邻法的步态识别方法.对于每个步态序列,首先通过背景减差法检测运动目标的轮廓,然后使用Hilbert-Huang描述子对轮廓线进行描述,获取步态特征.运用K-近邻法分析方法,对步态进行自动识别.实验结果表明所提出的步态识别方法识别率达到了78%. This paper proposes a gait recognition method based on Hilbert-Huang Descriptors （HHDs） and K-Nearest Neighbors method （KNN）. For each image sequence, background subtraction algorithm is firstly used to segment the silhouettes of moving object from the background. When analyzing the body contours by HHDs, we obtained the gait feathers. K-Nearest Neighbors method is finally applied to analysis and recognition automatically. The experimental results demonstrated that the approach reached 78% correctly rate. 步态识别 Hilbert-Huang描述子 K-近邻法 gait recognition Hilbert-Huang descriptors K-Nearest Neighbors TP301 国家自然科学基金（60702079）;湖北省自然基金（2006ABA027）资助项目 2008-01-01 2021-04-01 1