Improved balanced binary tree based action recognition

• Published in: 2016 Chinese Control and Decision Conference (CCDC) pp. 113 - 118
• Main Authors: Cheng, Yanyun, Zhu, Songhao, Liang, Zhiwei, Xu, Guozheng
• Format: Conference Proceeding Journal Article
• Language: English
• Published: IEEE 01.05.2016
• Subjects: Adaptive optics; Balanced Binary Tree; Behavior Recognition; Classification algorithms; Computer vision; Conferences; Frequency-domain analysis; Human behavior; Image motion analysis; Image Segmentation; Mathematical analysis; Mathematical model; Monitoring; Moving object recognition; Optical Flow; Optical imaging; Recognition; Support Vector Machine; Tasks
• DOI: 10.1109/CCDC.2016.7530964

Action recognition is one of the core content of intelligent monitoring, and also the basis of video content analysis and understanding. A novel method is here proposed to enhance the accuracy of human behavior recognition. First, each video image is divided into five sub-regions based on the motion mechanism; then, the frequency information of optical flow within each sub-region is extracted to describe the motion characteristics of each sub-region; finally, an improved balanced binary decision tree-support vector machine is utilized to complete the task of behavior recognition. Experimental results conducted on KTH database demonstrate the proposed algorithm can improve the accuracy of behavior recognition.