Research On Digital Correction Technology For Monitoring Video Image

Digital correction technology for distorted video images is one of the important ways in video monitoring field. The digital correction technology for geometrical distorted video images to fulfill the demand of CAPF (Chinese Armed Police Force) guard missions is investigated detailed in this thesis. At first, the geometrical distortion characteristic of the vidicon lens is introduced. Then, the digital correction principle and the traditional correction methods for geometrical distorted video images are expounded. Finally, two methods for correcting geometrical distorted video images are proposed.According to the realistic demand of image accuracy in the CAPF guard missions, a correction method for geometrical distorted monitoring video image based on the grid templates is proposed after analyzing the correction methods which are based on the hexangular lattice template and the grid template. The geometrical distorted monitoring video image can be corrected by means of low order interpolation in the proposed method, which is based on segmental correction and segmental interpolation. The simulation results show, compared to the traditional methods, although the image accuracy is lower, the proposed method has advantages of fast computing and easy implementation which can fulfill the demand of CAPF guard missions.A correction method for distorted monitoring video image is also proposed in this paper based on cumulative-sum and computation of distortion parameters to solve the problem of high computing complexity in the classic correcting methods. After analyzing the distortion characteristic of the vidicon lens, a mathematical model of geometrical distortion, which is suitable for computing and cumulative-sum, is designed to realize the geometrical correction by means of calculating the distortion parameters rapidly. The simulation results show, compared to traditional Least Squares Method, the proposed method has the same accuracy and the advantages of fast computing, easy implementation and high applicability.