Wide-angle camera irregular distortion global correction method

Abstract

The invention relates to a wide-angle camera irregular distortion global correction method, and belongs to the field of visual measurement. The method comprises the following steps: S1, picture collection: using related software to make an auxiliary line for a collected image, and displaying the auxiliary line in real time; Aligning the auxiliary lines to the central area of the checkerboard, andshooting a picture; S2, corner detection; S3, checkerboard detection: for the detected corner, detecting a checkerboard through a growth method; S4, data initializing; S5, curved surface fitting: fitting an interpolation curved surface by using an improved thin plate spline interpolation TPS method; S6, generation of an interpolation map: reversely solving the interpolation map by using the obtained interpolation curved surface; And S7, real-time distortion correction: loading an interpolation map, collecting pictures in real time, and carrying out distortion correction in real time. Accordingto the method, some irregular (symmetrical) distortion can be corrected, and compared with an original TPS interpolation method, the edge precision and the observability outside a control point are improved.

Description

technical field [0001] The invention belongs to the field of visual measurement, and relates to a global correction method for irregular distortion of a wide-angle camera. Background technique [0002] With the rapid development and continuous updating of the computer industry and the chip industry, technologies such as intelligent driving and intelligent monitoring continue to mature. The function and application of the camera have been greatly developed. The camera is not only to obtain images, but we hope to process the obtained pictures through the computer and extract the required information from them to complete subsequent calculations and applications. We know that the points in the three-dimensional world have unique corresponding points on the two-dimensional image. In order to obtain the information in the image more accurately, we need to establish a camera imaging model. The parameters of these models are camera parameters. The process of solving these camera pa...

AI Technical Summary

Problems solved by technology

[0006]Secondly, for irregular image distortion or asymmetrical distortion, using Zhang Zhengyou’s calibration method cannot be solved

[0007] Finally, the original TPS (thin plate spline interpolation) algorithm performs distortion correction. Due to the influence of control points, for edges (outside the outermost control points) area), there will be large unprocessed distortions, which will affect the visibility

Images