Image Based Automatic Detection Quadrilateral Algorithm
HU Li-hu · Journal of Chinese Computer Systems · 2014
The rectangle is an important shape of mechanical components. However,due to the diversity of the image,few people made studies the automatic detection algorithm of rectangle. As the projective transformation of the rectangle is quadrilateral,rectangle extraction can be converted to the extraction of quadrilateral in the image. So,an image based automatic detection quadrilateral algorithm was introduced in the paper. The algorithm pretreated the combination and deletion of the lines detected by canny edge detector,and then the initial quadrilateral can be extracted by the geometric characteristics of the quadrilateral. Finally,based on the high.level semantic information,error quadrilateral can be found and deleted. At last,taking VB as the development tool,the automatic extraction system based on quadrilateral has been designed and implemented,and the effectiveness of the proposed algorithm was verified by image single object angles for objects. The proposed algorithm can be used in the matching,classification and detection of the image.