A distortion image correction method based on machine vision. (January 2020)

• Record Type: Journal Article
• Authors: Tan, Guangxing; Ding, Ying; Fu, Dandan; Wang, Yuchen
• Is Part Of: Journal of physics. Volume 1453(2020)
• Publication Date: 2020-01
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/1453/1/012136
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: As autopilot becomes a hot spot in the field of automobile, machine vision is widely used in autopilot technology, and calibration of camera parameters is the basic work of obtaining driving environment by using vision. In this paper, based on camera parameters and machine vision technology, the monocular camera image distortion correction algorithm is studied. Based on the perspective transformation principle, the method of controlling transformation point is used to correct the image. The black-and-white checkerboard is taken as the object of image acquisition, and the results obtained in this paper are compared with those obtained by the calibration toolbox camera calibrator of matlab camera. The algorithm proposed in this paper is used to correct the images collected in traffic environment, which lays a foundation for the subsequent recognition of traffic signs. The experimental results show that the distorted image correction results are good, the accuracy is higher, and it is practical and effective.