Human path classifier architecture

• Main Authors: Khan, Imran Moez, Htike@Muhammad Yusof, Zaw Zaw, Khalifa, Othman Omran, Lai, Weng Kin
• Format: Book Chapter
• Language: English
• Published: IIUM Press 2011
• Subjects: TK7800 Electronics. Computer engineering. Computer hardware. Photoelectronic devices
• Online Access: http://irep.iium.edu.my/21651/1/Chapter_18.pdf; http://irep.iium.edu.my/21651/; http://rms.research.iium.edu.my/bookstore/default.aspx
• Institution: Universiti Islam Antarabangsa Malaysia

Many surveillance systems today provide only a passive form of site monitoring. Extensive video records may be kept to help find the instigator of criminal activities after the crime has been committed but preventive measures require human involvement. In addition, there is a need for large amounts of data storage to keep up to several large volumes of video streams that may be needed for later analysis. However, monitoring and storage space are not the only concerns. Behavioral analysis itself can be applied to numerous features extracted from video sequences including path detection and other aspects of human behaviour. Up till now, path classification has been carried out mainly using Boolean logic and allows only the identification of unusual paths, and not the extent to which they are deviant from usual paths. This chapter reports on the results to solve this problem with a fuzzy inference approach to classify paths into different categories