Conference Paper

Enhanced Skin Detection Technique Using Block Matching

Coll. of Comput. Studies, Amman Arab Univ. for Grad. Studies, Amman
DOI: 10.1109/ICSPC.2007.4728245 Conference: Signal Processing and Communications, 2007. ICSPC 2007. IEEE International Conference on
Source: IEEE Xplore


Skin detection represents a challenge in recent years, due to the fact that imaging conditions and illumination factor affect the resulted image. In addition, human skin color has a wide range making it a tedious task to target and extract. This paper addresses a novel method to detect skin in still images. The detection approach is based on five face detection models with integral modifications to increase detection rate in those models. Our hybrid skin detection method is used to detect skin regions in images using a block matching technique integrated over the resulted map of each model to pick the most frequent block as the final result. The hybrid model produced a better result in reference to the overall skin recall and precision values when tested on a single and multiple subjects in an image.

3 Reads
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Accurate hand segmentation is a challenging task in computer vision applications. We propose a new method to segment hand based on free-form skin color model. The pixel value of a person's hand is captured and represented in YCbCr color model. The CbCr color space is mapped to a CbCr plane in order to produce a clustered region of skin color. Then, instead of using ellipse to model the skin color, edge detection is performed on the clustered region to construct a free-form skin color model. The hand segmentation result, tested on various complex backgrounds gives promising results.
    Full-text · Conference Paper · Sep 2010
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Hand segmentation is often the first step in applications such as gesture recognition, hand tracking and recognition. We propose a new technique for hand segmentation of color images using adaptive skin color model. Our method captures pixel values of a person's hand and converts them into YCbCr color space. The technique will then map the CbCr color space to CbCr plane to construct a clustered region of skin color for the person. Edge detection is applied to the cluster in order to create an adaptive skin color boundaries for classification. Experimental results demonstrate successful detection over a variety of hand variations in color, position, scale, rotation and pose.
    Full-text · Conference Paper · Dec 2010