This paper presents a new method for skin region segmentation based on a zero-sum game theory model which exploits the opposite classifications of an image region by different skin detectors. In fact, these regions are considered conflict areas between two players (skin and non-skin) and skin detectors are considered strategies. An appropriate utility function is then defined. The computation of the saddle point (The Nash equilibrium) in the mixed extension of the proposed zero-sum game allows classifying effectively the conflict areas and so reducing the false positive skin detection. Experiments were conducted on three publically available databases using four selected skin detectors based on skin color information, skin-texture cues and employ rule-based or neural networks. The results show that the proposed method outperforms the existing skin segmentation approaches in reducing the false positive rates and obtains promising results in the skin segmentation performance.

本文提出了一种基于零和博弈理论模型的皮肤区域分割新方法，该模型利用不同的皮肤检测器对图像区域进行相反的分类。实际上，这些区域被认为是两个参与者（皮肤和非皮肤）之间的冲突区域，并且皮肤检测器被认为是策略。然后定义适当的效用函数。所提出的零和游戏的混合扩展中的鞍点（纳什均衡）的计算可以有效地对冲突区域进行分类，从而减少错误的阳性皮肤检测。实验是在三个可公开获得的数据库上进行的，这些数据库基于皮肤颜色信息，皮肤纹理提示并使用基于规则或神经网络的四个选定的皮肤检测器。