Binocular camera and self-calibration method for binocular camera

Abstract

The invention discloses a binocular camera comprising first and second imaging lenses and first and second image sensors, wherein the binocular camera further comprises first and second optical path conversion devices arranged on object sides of the first and second imaging lenses, a calibration mark fixed relative to the first imaging lens and a calibration lens arranged between the calibration mark and the second optical path conversion device, and the calibration mark can be imaged on the second image sensor by the calibration lens and the second imaging lens. The invention further discloses a self-calibration method for the binocular camera. The self-calibration method comprises the following steps: collecting an image of the calibration mark through the second image sensor; extracting feature parameters in the image; comparing the feature parameters with a pre-stored reference benchmark; and when the difference of the feature parameters and the pre-stored reference benchmark exceeds a predetermined threshold, judging that the binocular camera needs to be marked again. According to the self-calibration method disclosed by the invention, the state in which the binocular camera needs to be marked again can be detected conveniently and effectively.

Description

technical field [0001] The present invention generally relates to a stereo camera device, in particular to a binocular camera device and a self-calibration method for the binocular camera. Background technique [0002] Binocular cameras are an increasingly attractive device capable of providing stereoscopic vision. Based on the images obtained by the binocular camera, the three-dimensional space position of the object captured by the binocular camera relative to the camera can be calculated through the principle of binocular parallax. [0003] In the three-dimensional space position calculation, the positional relationship between the two imaging lenses of the binocular camera has a great influence, so the positional relationship needs to be calibrated and calibrated. After the initial calibration and calibration, since the connection between the two imaging lenses in the binocular camera cannot be an ideal rigid body connection, the mechanical connection will slowly deform...

AI Technical Summary

Problems solved by technology

Returning to the factory for recalibration will affect the use of binocular cameras

Dynamic calibration usually extracts the feature points in the binocular image in real time for recalibration, but in the case of inconspicuous features, it often leads to dynamic calibration failure, and even deteriorates the calibration results

Images