Calibration and ranging method and device, electronic device, and computer-readable storage medium

Abstract

The embodiment of the disclosure discloses a calibration and ranging method and device, an electronic device, and a computer-readable storage medium. The ranging method includes: acquiring calibrationdata of an image collection device; carrying out calculation based on the calibration data to obtain a calibration distance between a to-be-measured target and the image collection device, and carrying out estimation based on the calibration data to obtain an estimated distance between the to-be-measured target and the image collection device; and calculating a distance between the distance between the to-be-measured target and the image collection device based on the calibration distance and the estimated distance. Therefore, the calibration and ranging accuracy is improved while the viewingangle is extended and the calibration and ranging errors are reduced; and the hardware resources are saved and the application range is wide.

Description

technical field [0001] The present disclosure belongs to the technical field of computer vision, in particular to a computer vision-based calibration and ranging method, device, electronic equipment, and computer-readable storage medium. Background technique [0002] With the development of society, people have higher and higher requirements for non-contact distance or length measurement. In the prior art, monocular distance measurement is a relatively commonly used non-contact distance measurement method. Monocular ranging, also known as single-camera ranging, calculates the physical distance between the camera or camera and the target to be measured based on the image coordinates in the given photo based on simple geometric relationships. When using monocular ranging , first calculate the vertical distance between the cross-section of the target to be measured and the camera lens on the horizontal plane, then calculate the distance between the target to be measured and th...

AI Technical Summary

Problems solved by technology

The monocular distance measurement technology is based on a simple geometric relationship and relies on lens internal parameters for calculation. The calculation process is relatively simple, involves fewer parameters, and has low requirements for hardware. However, due to its strong dependence on lens parameters, in practice In the application, there is usually the problem of inaccurate lens parameters, so the error of monocular distance measurement is relatively large, and the existing monocular distance measurement methods can only be applied to general standard lenses instead of wide-angle lenses because they do not consider lens distortion. The range is small. If it is applied to a wide-angle lens, it will cause an error of about 1 meter because the lens distortion is not considered.

The basic principle of binocular or multi-lens ranging technology is similar to that of monocular ranging technology, but binocular or multi-lens ranging technology uses two or more cameras to measure distance, and its visual range is larger, but its The hardware resources used are relatively large, and the requirements for hardware are relatively high. It is not suitable for some scenarios that require high cost performance.

Images