Human faces detection method using genetic algorithm
Yuuki Yokoo, Masafumi Hagiwara · 2002
A novel method to detect human facial regions from an image with a complex background has been proposed. To detect human facial regions precisely, without imposing any constraints on the face size, position or features (such as glasses, beard and so on), the method employs a genetic algorithm (GA). First, the proposed method detects the edges in an input image. Second, it searches elliptical regions using a GA, based on the idea that a human face can be approximated by an ellipsoid. Finally, it judges whether each region is a human face or not. We confirmed by computer simulations that the proposed method could detect human facial regions in various images.