Research Of The Photogrammetry Based On Digital Photograph Technology

The camera calibration not only is one of the basic problems of computer vision, but also is the foundation of photogrammetry. The purpose of camera calibration is to determine the camera’s internal and external parameters, and the relationship between the three-dimensional coordinate and the image coordinate of the projection on the camera imaging plane about the point is obtained by these parameters. And these parameters of the camera can be obtained through extracting the internal corner from the checkerboard pattern. Firstly, all corners is detected from the image, and then the internal corners are recognized from all of corners on the basis of the local gray features of the checkerboard pattern’s internal corners and the structural characteristics that consist of the grid line of the checkerboard pattern, finally the connection is determined between the internal corner and the point of three-dimensional space, and the origin of reference frame is also determined in order to achieve the extraction the image’s internal corner of the the checkerboard pattern, as well as to realize the calibration of the camera.The stereo matching is the key technology in photogrammetry, and also is the precondition of the three-dimensional information recovery. The parallax of the corresponding points is attained according to the stereo images, and then the three-dimensional images are reconstructed accordance to the value of parallax.The image matching is the reconstructing process of three-dimensional in essence, first of all, the parallax of the corresponding points is attained according to the stereoscopic image, and then the scene of three-dimensional images is reconstructed in accordance with the value of parallax. The binocular stereo vision technology is the foundation and the mainly way to extract the three-dimensional information from the objects in the close-range photogrammetry. At first, the corresponding relationship is seeked among the points of scenery between the right image and the left image, and then the value of parallax is solved for every point. If the stereoscopic image satisfies a geometry constraint conditions, the reconstruction of three-dimensional object will be achieved, and it can accurate obtain the corresponding points and the imaging models after the camera calibration, so this is the key to get the accurate data. The distance information of original scene is recoveried based on the parallax information as well as the geometric relation between the projection models in the case of the imaging model as the known quantity, and then the three-dimensional description is obtained directly. Whether or not the three-dimensional data model will be established quickly, the precondition is the auto matching problem can resolve accurately and rapidly among the corresponding points.