A face detection method based on kernel probability map. (August 2015)

• Record Type: Journal Article
• Authors: Mahmoodi, Mohammad Reza; Sayedi, Sayed Masoud
• Is Part Of: Computers & electrical engineering. Volume 46(2015)
• Page Start: 205
• Page End: 216
• Publication Date: 2015-08
• Subjects: Face detection -- Skin detection -- Face segmentation -- Face extraction -- Image processing -- Kernel probability map; 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.2015.02.005
• Languages: English
• ISSNs: 0045-7906

Abstract

Graphical abstract: Highlights: A multi-stage face detection method is proposed using color and texture information. Using skin detection, candidate windows are extracted. Candidate windows are verified based on combination of KPM and LHP. Both qualitative and quantitative results confirm the merit of the algorithm. Abstract: Face detection is one of the most important parts of biometrics and face analysis science. In this paper, a novel multi-stage face detection method is proposed which can remarkably detect faces in different images with different illumination conditions, variety of poses and disparate sizes. The idea is to utilize a preprocessing step to filter many non-face windows by means of a skin segmentation procedure in order to boost the speed of the detection and also utilize the color information as much as possible. Subsequently, candidate windows are fed to a Local Hierarchical Pattern (LHP) generator unit where a new texture pattern is produced. Based on this pattern, a kernel probability map is calculated for each window, and by summing probabilities of all kernels and comparing it with a predefined threshold, decision is made about content of the window. Not only does this algorithm effectively eliminate many non-face regions, but it is also capable of detecting faces with relatively acceptable rate in different conditions.