Towards Integrated Image Contrast Models in Segmentation of Trees

摘要: Computer vision is an area in high demand which bringing new trends for urban and rural applications. Some examples can be found autonomous navigation projects, monitoring services, fruits/grain harvesting, pest control, so forth. However, drastic or even unperceptive changes the image acquisition process limit development of these applications, especially problems that require solutions uncontrolled environments such as outdoor areas. Thus, definition what a machine looking at challenging task. In this study, we dealt with segmentation problem order to develop method delineate tree trunks, their branches, foliage. As detection crucial topic mobile robotics, investigated it give initial interpretation external scenes. We prepared dataset validate proposal two classes were defined, non-tree. The pixels each classified based on proposed method, results show our obtained positive result 91% accuracy.