| 摘要 |
An object detecting method and device for detecting the position and posture of an object with low computational complexity and small storage area and a robot mounted with the device are provided. Flat parts and edge parts of an object are extracted from distance data (101) on the object (102), a given flat part is selected from the extracted flat parts and its equation is determined by calculation (103), edge parts are projected onto the calculated flat part to create an edge image (104), a reference edge image preliminarily stored in a database (105) is sequentially rotated (106), edge points having gray-scale values approximate to one another in the edge images are associated, and the difference between the positions are voted into a position space (107), the highest part of the votes is determined (108), the position and posture of the object are identified (109), and thus the position and posture are determined as those of the object. |
