Projective rectification of infrared image based on projective geometry

Projective distortion caused by tilted camera brings great inconvenience to subsequent image processing and pattern recognition. In this paper, a method based on projective geometry is introduced to correct the projective distortion of infrared images. The dual conic in distortion image, which contains all the information of projective rectification, is identified firstly, and then the projective transformation matrix is derived by decomposing the dual conic. Finally, optimization is carried out to get the optimal result. Experimental results show that the method used in this paper can remove the projective distortion of infrared image effectively.