New feature descriptor: Extended Symmetrical-Diagonal Hexadecimal Pattern for efficient background subtraction and object tracking. (February 2018)

• Record Type: Journal Article
• Authors: Jeyabharathi, D.; Dejey,
• Is Part Of: Computers & electrical engineering. Volume 66(2018)
• Page Start: 454
• Page End: 473
• Publication Date: 2018-02
• Subjects: Object tracking -- Background subtraction -- Spatio-temporal video patch -- Local ternary pattern -- Extended Symmetrical-Diagonal Hexadecimal Pattern; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2017.11.001
• Languages: English
• ISSNs: 0045-7906

Abstract

Highlights: New feature descriptor ES-DHP for both background subtraction and tracking. Symmetrical patterns are used to give an exact background model. It reduce the length of the feature vector. Abstract: Now-a-days the biggest challenge of computer vision technology is perfect detection of an object and tracking in an outdoor environment. Many approaches have been proposed to address some of the challenges, but still some problems persist, for instance high frequency motion of dynamic texture, motion of camera, abrupt changes in illumination, etc. To cope with these challenging scenarios effectively, a novel Extended Symmetrical-Diagonal Hexadecimal Pattern (ES-DHP) is proposed for both background subtraction and object tracking. Each object is represented by its corresponding ES-DHP. Metrics like F-score, FNR, FPR, MMR and elapsed time are used to measure the performance of the proposed system. Extensive experimental evaluations on a wide range of benchmark data sets validate the efficiency of ES-DHP when compared with other existing methods for unconstrained video analytics. From the outcomes, it is viewed that the proposed framework provides great outcomes for background subtraction and tracking. Graphical abstract: