Global Binary Continuity for Color Face Detection With Complex Background

International Conference on Materials, Alloys and Experimental Mechanics 2017

材料科学;金属学;机械制造

Belavadi, Bhaskar^1 ; Prashanth, K. V. Mahendra^1 ; Joshi, Sujay S.^2 ; Suprathik, N.^2

Department of ECE, SJBIT, VTU, Bengaluru

560060, India^1

Department of ECE, RVCE, Bengaluru

560059, India^2

关键词: Background environment;    Color images;    Color space;    Complex background;    discontinuous points;    Global Binary Continuity;    Human faces;    Segmented images;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/225/1/012256/pdf
DOI  :  10.1088/1757-899X/225/1/012256

学科分类：材料科学（综合）

来源: IOP

【 摘 要 】

In this paper, we propose a method to detect human faces in color images, with complex background. The proposed algorithm makes use of basically two color space models, specifically HSV and YCgCr. The color segmented image is filled uniformly with a single color (binary) and then all unwanted discontinuous lines are removed to get the final image. Experimental results on Caltech database manifests that the purported model is able to accomplish far better segmentation for faces of varying orientations, skin color and background environment.