A novel circular points-based self-calibration method for a camera’s intrinsic parameters using RANSAC

Abstract

Intrinsic parameters in camera calibration are commonly solved with high precision using the homography matrix of space coordinates and image coordinates of a template. However, in practice the accuracy in the extraction of corner points of the calibration plate is usually affected by the lighting environment, leading to obvious fluctuation in the calibration results. This study proposes a novel self-calibration method to improve the calibration accuracy and reduce the instability of calibration results by using circular points and the RANdom SAmple consensus (RANSAC) method. The circular points for the intrinsic parameters are calculated using the cross-ratio method with a calibration plate containing nine corner points. The distance between the circular points and image of the absolute conic is defined. The threshold value of the RANSAC model is simulated by a computer. The intrinsic parameters are initially estimated using the unreliable calibrating images excluded by the RANSAC method. The definition of the threshold is based on the Sampson estimation. The maximum likelihood estimation method is performed to reestimate the intrinsic parameters and optimize the calibration result. The findings of the numerical simulations and experiments on wing-fuselage docking based on monocular vision demonstrate that the proposed method is more robust and efficient at improving the calibration accuracy than the traditional methods. The measurement error is reduced to less than 0.013 mm when the calibration algorithm is applied to actual applications such as wing-fuselage docking.

Conic tangents based high precision extraction method of concentric circle centers and its application in camera parameters calibration

“…Most calibration approaches are performed with high-precision or special structure targets that are difficult to manufacture. According to the employed targets, the calibration methods can be divided into 1D methods 5 – 9 , 2D method 10 – 26 and 3D method 2 , 3 , 27 – 31 and approximate 3D method 32 – 40 . The targets in 1D methods are usually calibration rods composed of several collinear points.…”

“…However, there are few collinear points, so the calibration accuracy is usually not high. The targets in 2D methods are usually calibration boards with given patterns, such as circles 16 – 25 , checkerboard 10 – 15 or other star-shaped pattern 26 . The targets in 3D methods are usually composed of two or more planes with definite position relationships, which may be helpful to obtain higher calibration accuracy but also leads to the problems of high manufacturing cost and the complex calibration process.…”

“…Meng and Hu 19 were the first to calibrate the camera intrinsic parameters based on the technique of circular points with a hub-and-spoke plane target including six spokes. However, Li et al 26 pointed out that Meng and Hu’s method has two major shortcomings. A numerical method should be employed in these methods to solve the conjugate complex points, such as Newton iteration 25 .…”

Dynamic Granularity Matrix Space Model Based High Robust Multi-Ellipse Center Extraction Method for Camera Calibration

“…The task of camera calibration is to determine the relationship between the two-dimensional (2D) coordinates of image points and the 3D world coordinates of the object, which takes into account the optical geometry of cameras and the positional relat ionship between cameras. In order to improve the robustness and calibration accuracy, various calibrat ion boards are used for camera calibrat ion, such as planar chessboard [11], Charuco board [12], circu lar arrays calibrat ion board [13][14], and so on. In practice, the extraction accuracy of corner points with planar chessboard is usually affected by the lighting environment and noise, leading to obvious fluctuation of calibrat ion results [14].…”

“…In order to improve the robustness and calibration accuracy, various calibrat ion boards are used for camera calibrat ion, such as planar chessboard [11], Charuco board [12], circu lar arrays calibrat ion board [13][14], and so on. In practice, the extraction accuracy of corner points with planar chessboard is usually affected by the lighting environment and noise, leading to obvious fluctuation of calibrat ion results [14]. Although Charuco board can achieve higher calibration accuracy than planar chessboard, Charuco board has higher production complexity and is also affected by extraction accuracy of corner points under uneven illu mination.…”