Camera correction method, device, electronic device and storage medium

Abstract

Embodiments of the present disclosure provide a camera correction method, device, electronic equipment, and storage medium, which belong to the technical field of image processing. The method includes: obtaining the overlap rate of the first identification frame area in the first camera and the second identification frame area in the second camera, the overlap rate includes the intersection area between the first identification frame area and the second identification frame area and the ratio of the union area; determine whether the coincidence rate is greater than or equal to the preset threshold; if the coincidence rate is less than the preset threshold, then adjust the first camera and the second camera according to the preset rules, so that the first camera and the second camera The coincidence rate corresponding to the camera is greater than the preset threshold. The solution of the present invention displays the alignment effect of binocular or even multi-eye cameras in real time in the form of data, and debugs each camera to achieve the optimal alignment effect, thus ensuring that only one camera can obtain the face recognition frame in practical applications, saving other cameras Detection time, thereby improving the passing speed of face recognition.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to a camera correction method, device, electronic equipment and storage medium. Background technique [0002] At present, face recognition technology is widely used in various electronic display products. In face recognition technology, face recognition may be attacked by counterfeit faces such as mobile phones, printed paper, computer screens, face masks, 3D models, etc., and in order to prevent these effects, liveness detection technology is currently used to defend against these attacks. Currently, binocular cameras are used in face recognition devices, one for face recognition and one for liveness detection. The camera used for live detection can be an IR camera, a 3D structured light camera, a TOF depth camera, etc. [0003] Currently, a device using a binocular camera needs to detect faces in the binocular image separately, and find the position of the face ...

AI Technical Summary

Problems solved by technology

Generally, the face detection algorithm takes about 10 times more time than the face key point regression algorithm.

In the current binocular cameras, it is generally necessary to use both cameras to detect faces, which increases the detection time in the face recognition process. The time-consuming and inefficient use of binocular cameras for face detection is the current binocular The main problems of the camera in face recognition

[0004] It can be seen that the existing binocular cameras have the technical problems of long detection time and low efficiency

Images