Research on distortion correction of fisheye image based on improved linear feature model

The fisheye lens is a kind of ultra-wide angle camera, which has the characteristics of short focal length and large angle of view. It is widely used in video conference, security monitoring, intelligent transportation and other fields. However, the large viewing angle advantage of the fisheye lens is at the cost of very serious distortion. In order to obtain effective scene information in the picture, it is necessary to correct the distorted image to a distortion free linear perspective projection image. This paper studies the characteristics of nonlinear imaging models and proposes a distortion correction algorithm of fisheye image based on improved linear feature model. The algorithm uses Taylor polynomial to approximate the general model of fisheye lens, proposes the method of backward coordinate mapping for distortion correction. The experimental results show that the method proposed in this paper has better performance in terms of qualitative characterization of pixel error and re-projection error.