Authors : Maryam Mahmood Hussein and Ammar Hussein Mutlag
Abstract: Face identification is an important factor in the field of Personal Computer (PC) vision, since, it is the initial phase in different applications like face acknowledgment, military knowledge and reconnaissance, human PC association and so on. In this study, the execution examination has been made of 2 different algorithms for face identification in color images. The 1st algorithm is the CIELAB color model. In this algorithm, the knowledge of geometrical properties of human face has been utilized to determine if the skin area is human face or not. The 2nd algorithm is Viola-Jones algorithm. This method is used to identify face from entire image. The main concepts engaged with Viola-Jones are Harr features integral image, AdaBoost learning and cascade classifier. The overall comparison for both algorithms have applied under MATLAB environment. For fair comparison, 66 images are applied for both algorithms. The success of the Viola-Jones algorithm is 87.87% meanwhile the success of the CIELAB algorithm is 46.96%. The outcome exhibits that the Viola-Jones method gives better outcomes contrasted the other method.
Maryam Mahmood Hussein and Ammar Hussein Mutlag, 2019. Face Detection Methods: A Comparative Study Between Viola-Jones and Skin Color Detection. Journal of Engineering and Applied Sciences, 14: 4754-4760.