Search

In this paper, a stereo calibration method for binocular ultra-wide angle long-wave infrared camera is proposed on the basis of an equivalent small field of view camera. Extrinsic parameters are calibrated through the corrected images from the left and right cameras. They can be viewed as images taken by a small field of view camera. The calibration procedure consists of three steps: monocular calibration, distortion correction and extrinsic parameters calibration. In order to evaluate the accuracy of the method, stereo vision of the camera is modelled and a 3D reconstruction approach is presented. A series of experiments, including intrinsic parameters, extrinsic parameters and 3D reconstruction, are conducted to validate the proposed method. The results show that the baseline length error decreases to 0.67%, and the relative error for the 3D reconstruction of corners is smaller than 8.11%. In contrast to the common stereo calibration method, it improves calibration accuracy. [ABSTRACT FROM AUTHOR]

In this paper a approach of corners positioning for calibrating binocular ultra-wide angle long-wave infrared camera is presented. Firstly, the generalized camera model which suits for its imaging relationship is introduced. Secondly, a method for subpixel corners positioning is proposed, containing edge extraction, gray centroid and curve fitting. Thirdly, aiming at calibrating extrinsic parameters of the camera, we raise a method with least squares optimization. The calibration experiments are conducted on the basis of an active infrared radiation calibration checkerboard. The performance of the approach is discussed by using other data sets. The results reveal that: MRE (mean reprojection error) and RMS (root mean square) respectively reduce to 0.23pixels and 0.33pixels, and the error of the baseline length declines from 1.00 % to 0.23 % in comparison with common methods. [ABSTRACT FROM AUTHOR]

• An automatic calibration method is proposed for the ultra-field infrared and visible cameras. • A thermal radiation checkerboard is designed, which can be used in infrared and visible bands at the same time. • The corners and center can be automatically located at subpixel level. • RBF based nonlinear refinement is proposed to improve the calibration accuracy. Camera calibration is a vital and important work for machine vision. In the paper, an optimized calibration method for ultra-field dual bands cameras is proposed, which is based on the designed thermal radiation checkerboard. The optimizations are achieved in subpixel location, center location and mapping relationship, respectively. Specifically, the subpixel corner is located by performing the designed variable quadrilateral template on edge gradients and calculating their minimum orthogonal value. Then, the center location is accomplished through Gaussian fitting, in which the bottom point is taken as the potential center in iterations. Moreover, the RBF network is employed to replace the polynomial fitting and fit the mapping relationship between the camera coordinates and sensor coordinates. Thus, the accuracy of the mapping is increased, which contributes to the optimization of the intrinsic and extrinsic parameters. The experiments results show that the average root mean square error of the reprojection is reduced by nearly 50%. The calibrated parameters are accurate and have physical significance in the proposed calibration scheme. [ABSTRACT FROM AUTHOR]