| 摘要 |
PROBLEM TO BE SOLVED: To provide an image processing device capable of improving accuracy in object recognition based on images picked up by a camera and an image processing method.SOLUTION: An object recognition section of the image processing device separately extracts a first feature point and a second feature point of a first segment and a second segment each of whose variation amount is not less than a variation amount threshold, and a first feature point and a second feature point of a first segment and a second segment each of whose variation amount is less than the variation amount threshold, and determines a corresponding point of the first feature point of the first segment whose variation amount is not less than the variation amount threshold in a second picked-up image to be the second feature point of the second segment corresponding to the first segment whose variation amount is not less than the variation amount threshold in order to recognize an object. Through this, the first feature point and the second feature point only by an object to be recognized are discriminated from the first feature point and the second feature point by a distance view which is not to be recognized, and the first feature point and the second feature point only by the object to be recognized are associated with each other, which can improve the accuracy in object recognition.SELECTED DRAWING: Figure 2 |
