The Research Of Distortion Correction Algorithm For Wide-angle Image And FPGA Implementation

With the development of computer technology,and the application of all kinds of intelligent products,the camera was widely used in road traffic,medical,banking and other fields of the machine vision.The camera uses a wide-angle lens with a larger field angle than average lens.But the wide-angle lens also has a significant flaw,when it is in the shooting image,it will inevitably produce barrel distortion,which is a non-linear distortion,the screen in addition to the straight line through the center is still straight,bending the other part of the straight.The farther away from the center,the greater the amount of deformation.This distortion image could not meet the normal visual requirements of people.In order to enhance the visual effect,this paper improves the algorithm of wide-angle image distortion correction.In this paper,the correction algorithm of image distortion caused by the use of wide-angle lens has been studied,so that the corrected image can match the actual image to improve the accuracy of distorted image.In this paper,we will analyze the existing distortion correction algorithm.On this basis,we propose a new distortion correction algorithm and implement it with MATLAB.The algorithm is based on the algorithm of variation concentric circular template correction.The coordinate mapping relationship between the distorted image and the ideal image is calculated by the software based on the algorithm of variation concentric circular template correction.So the coordinate mapping from the ideal to distortion point can be searched in the 32×32 grid lookup table(the grid size is dependent on the resolution)of the mapping relationship,while the rest of coordinate have to be calculated by bilinear interpolation.Finally,the linear interpolation method is used to recover the pixels of the distorted image,and the correction of the wide-angle image is completed.With the limitation of the operation speed of computers,the software can not correct the real-time distorted images.In other words,the software can only correct the static but not dynamic distortion of the image.But the real-time distortion of the image correction can be implemented if the hardware is involved.In order to obtain a high-speed and stable calibration system,the distortion correction algorithm is implemented in hardware.The accuracy of the hardware correction depends on the accuracy of the correction of thesoftware algorithm.In this paper,the coordinates and gray levels need to be interpolated,and the bilinear interpolation can not only meet the certain precision requirements,but also the calculation and complexity can be controlled within a certain range.In this paper,with the use of verilog hardware description language for hardware implementation,the use of nlint for code style checking,and the use of VCS for hardware design module-level/IP verification,in order to ensure the design function to achieve the completeness and correctness.Finally,the design of the hardware simulation,that is,FPGA prototype verification.The verification results show that the design can quickly and effectively correct the wide-angle distortion image.The algorithm relies on a solid theoretical basis and a large number of simulation experiments,continuous optimization.Used in the video YCb Cr422 on the wide-angle image distortion correction processing to support the maximum resolution is 1920×1080,the proposed lens field angle is less than or equal to 120°.