Lu, JShen, JYan, W-QBacic, B2019-03-112019-03-112017-062017-06International Journal of Digital Crime and Forensics (IJDCF), 9(3), 11-27. doi:10.4018/IJDCF.20170701021941-62101941-6229https://hdl.handle.net/10292/12346This paper presents an empirical study for human behavior analysis based on three distinct feature extraction techniques: Histograms of Oriented Gradients (HOG), Local Binary Pattern (LBP) and Scale Invariant Local Ternary Pattern (SILTP). The utilized public videos representing spatio-temporal problem area of investigation include INRIA person detection and Weizmann pedestrian activity datasets. For INRIA dataset, both LBP and HOG were able to eliminate redundant video data and show human-intelligible feature visualization of extracted features required for classification tasks. However, for Weizmann dataset only HOG feature extraction was found to work well with classifying five selected activities/exercises (walking, running, skipping, jumping and jacking).IGI GLOBAL AUTHORS, UNDER FAIR USE CAN: Post the final typeset PDF (which includes the title page, table of contents and other front materials, and the copyright statement) of their chapter or article (NOT THE ENTIRE BOOK OR JOURNAL ISSUE), on the author or editor's secure personal website and/or their university repository site.Histograms of Oriented Gradients (HOG); Human Behavior Recognition; Local Binary Pattern (LBP)Journal ArticleOpenAccess10.4018/IJDCF.2017070102