基于纹理感知的地面目标检测方法

he texture difference between man-made objects and natural scene is a salient feature for biological visual system to recognize special object from background. A proposed method to detect man-made target based on texture perception is discussed in detail. Firstly, the feature parameters about texture are extracted by grey level co-occurrence matrix, and then the texture feature is estimated by combing the above extracted parameters. Secondly, texture, brightness and direction are used to build spatial salient map by nonlinear feature fusion, and it could enhance the man-made object with weak texture roughness and suppress the background with strong texture roughness. The method fuses the texture roughness and its similarity to focus on object. Some experimental results are included, and it shows that the method can accurately detect target in complex background.