Object recognition has a wide application in robotic vision. Recognizing the objects from the workspace is the most basic step in robotic technology which aims at achieving the unmanned operation. The recognition of objects obtained from the scene will be affected by the deformation of the target object, the occlusion by the background and other factors. This paper puts forward an algorithm of object recognition and fitting based on the shape information, with the intention to improve the effects of recognition and fitting the contour at the same time. At first, we make the use of the morphological edge detection and get rid of the most affects of the background. Then, we adopt an edge-tracing method and extract the feature points which match the feature data in the model library as the following step and determine the contour of the object to be fitted. At last, we fit the contour according to an improved fitting algorithm. |