| 摘要 |
<P>PROBLEM TO BE SOLVED: To make fast processing for correcting an image of a photographed object. <P>SOLUTION: A digital camera 1 is equipped with a CPU and an image processor. The image processor acquires an outline from an image obtained by photographing a white plate 2 as the photographed object by using a Roberts filter. The image processor generates a reduced image by selecting pixels from the acquired contour to reduce the amount of data. Further, the image processor selects straight lines as candidates forming the image of the white plate 2 according to length and angle information of them to acquire a quadrangular shape of the white plate 2 which is formed of the selected straight lines. The image processor finds projection parameters showing the relation between the image of the white plate 2 and the actual white plate 2 from vertex positions of the quadrangle to perform projection conversion of the image of the white plate 2. <P>COPYRIGHT: (C)2005,JPO&NCIPI |
