ABSTRACT
Visual navigation of the mobile robot mainly through the vision to perceive a scene, it extracts the image feature and describes the environmental information. Generally, image features are the description of image properties and initial features obtained by mobile robot vision[1]. Image features can be divided into global and local features. The global features are extracted from the whole frame image, they are sensitive to the viewangle and illumination, andmeanwhile, they need a large amount of calculation[2]. The local feature focuses on the description of the local features in the image. It can effectively resist geometric transformation and its redundancy is low[3,4].
