CN-121982115-A - Multi-camera correction method, system and medium

CN121982115ACN 121982115 ACN121982115 ACN 121982115ACN-121982115-A

Abstract

The invention discloses a correction method, a system and a medium for multiple cameras, wherein the correction method comprises the steps of controlling sampling conditions, analyzing system gains of the cameras, determining correction coefficients of the cameras based on the ratio of standard gains to the system gains, correcting the cameras to enable the responsivity of the corrected cameras to be consistent, extracting multi-frame images collected by the same camera, calculating time domain noise values of the cameras, determining noise correction coefficients of the cameras by adopting the ratio of the standard time domain noise values to the time domain noise values of the cameras, and carrying out noise filtering on gray values of pixel points in the images collected by the cameras according to the noise correction coefficients of the cameras so as to realize noise consistency adjustment of the cameras. The invention realizes the effect that the noise level is consistent on the premise of the consistency of the responsivity of each camera, and improves the universality of each camera under the same model or different models.

Inventors

LU ZHIPENG
Yin Yaxiang
YUAN JIANHUI
SHAO YUNFENG
CAO GUIPING
DONG NING

Assignees

合肥埃科光电科技股份有限公司

Dates

Publication Date: 20260505
Application Date: 20260113

Claims (10)

1. A method for correcting a plurality of cameras, for correcting noise of the plurality of cameras after correction of responsivity, so as to realize consistency adjustment of image quality collected by each camera, the method being characterized by comprising: Controlling sampling conditions, and analyzing the system gain of each camera to realize the system gain of each camera detected in the process of adjusting the light source from dark field to bright field; Based on the ratio between the standard gain and the system gain, determining the correction coefficient of each camera to correct each camera, so that the responsivity of each corrected camera is consistent; Extracting multi-frame images acquired by the same camera, calculating the time domain noise value of each camera, determining the noise correction coefficient of each camera by adopting the ratio of the standard time domain noise value to the time domain noise value of each camera, and carrying out noise filtration on the gray value of each pixel point in the images acquired by the cameras according to the noise correction coefficient of each camera so as to realize noise consistency adjustment of each camera; The time domain noise value is an average value of standard deviations corresponding to gray value variation of each pixel point in the whole image in a multi-frame image acquired by the same camera.
2. The method for correcting a camera according to claim 1, wherein two frames of images under the same sampling condition are extracted respectively, the average gray value of the two frames of images under the same sampling condition and the variance corresponding to the gray value difference of the two frames of images are calculated, and the ratio between the variance and the average gray value under the sampling condition corresponding to the maximum value of the variance is screened to determine the system gain of the camera.
3. The method for correcting a multi-camera according to claim 2, wherein the ratio between the average gray value of the image under each sampling condition and the average gray value of the image under the sampling condition with the maximum variance is selected based on the average gray value of the image under the sampling condition with the maximum variance, and the gray value of the image under each sampling condition is smaller than the threshold value of the ratio coefficient.
4. The method for correcting a multi-camera according to claim 1, wherein the method for determining the standard gain includes screening a maximum value of system gains of each camera, and taking a product of a preset coefficient and the maximum system gain as the standard gain, wherein the preset coefficient is greater than 1.
5. The method of calibrating a multi-camera according to claim 4, wherein each camera is calibrated to control the light source to obtain a stable light source, the method comprising: Judging whether a target gray value is positioned in a gray value range corresponding to an exposure parameter upper limit and an exposure parameter lower limit according to an initial exposure parameter range, wherein the exposure parameter range consists of an exposure parameter lower limit and an exposure parameter upper limit, and the exposure parameter lower limit is smaller than the exposure parameter upper limit; If the target gray value is not in the gray value range corresponding to the upper limit of the exposure parameter and the lower limit of the exposure parameter, adjusting the light source current to ensure that the target gray value is in the gray value range corresponding to the upper limit of the exposure parameter and the lower limit of the exposure parameter, and determining the light source brightness under the current light source current; based on the determined light source brightness under the current light source current, adopting gray values corresponding to an upper limit of an exposure parameter and a lower limit of the exposure parameter to construct a linear relation between the exposure time and the gray values; and judging whether the fitting times are larger than the set fitting times, and if so, performing dichotomy adjustment on the exposure parameter range, and updating the exposure parameter range to judge whether the gray value of the image is adjusted to the target gray value.
6. The method according to claim 5, wherein if the gray level of the image is not adjusted to the target gray level after the exposure parameter range is adjusted by the dichotomy, the determined brightness of the light source is halved or the updated exposure parameter range is adjusted by the dichotomy to obtain the exposure parameter and the brightness of the light source, and the gray level of the light source image is adjusted to the target gray level.
7. A method for correcting a camera according to claim 1, wherein the method for determining the standard time domain noise value of the same model of camera comprises determining the average value of the time domain noise values of the cameras of the same model.
8. The method for correcting a plurality of cameras according to claim 7, wherein a sliding window is adopted, gray values of the same pixel position in a plurality of continuous frames of images are extracted, an average gray value of the pixel position is calculated, and a noise correction coefficient of a camera is combined to calculate a gray value corresponding to the pixel position after filtering in each frame of images; the calculation formula of the gray value yi corresponding to the filtered image is that yi=xi/K σ +u(1-K σ ), xi is the gray value of a pixel point in the ith frame image, K σ is the noise correction coefficient corresponding to the camera for collecting a plurality of frames of images, and u is the average gray value of the pixel position in a plurality of frames of images.
9. A multi-camera correction system for use in the multi-camera correction method of any one of claims 1-8, comprising: The system data analysis module is used for controlling sampling conditions and analyzing the system gain of each camera so as to realize the system gain of each camera detected in the process of adjusting the light source from a dark field to a bright field; The response adjusting module is used for determining correction coefficients of all cameras based on the ratio of the standard gain to the system gain so as to correct all cameras, and the responsivity of all corrected cameras is consistent; The noise adjusting module is used for extracting multi-frame images acquired by the same camera, calculating the time domain noise value of each camera, determining the noise correction coefficient of each camera by adopting the ratio of the standard time domain noise value to the time domain noise value of each camera, and carrying out noise filtering on the gray value of each pixel point in the images acquired by the cameras according to the noise correction coefficient of each camera so as to realize noise consistency adjustment of each camera; The time domain noise value is an average value of standard deviations corresponding to gray value variation of each pixel point in the whole image in a multi-frame image acquired by the same camera.
10. A computer readable storage medium comprising a computer program, characterized in that the computer program, when executed by a processor, implements a method of correcting a multi-camera according to any of claims 1-8.

Description

Multi-camera correction method, system and medium Technical Field The invention belongs to the field of machine vision, and particularly relates to a multi-camera correction method, a multi-camera correction system and a multi-camera correction medium. Background When the responsiveness consistency correction is performed through the gain of the camera system, the collection of the gain of the camera system needs to be completed. In the existing scheme, the camera is usually subjected to the factory test and then is additionally subjected to the system gain collection test, the step is separated from the factory test step of the camera, additional test working hours are required, and test data are required to be manually arranged. The degree of universality of cameras of different types is different, on one hand, the fact that the cameras of different types have different name descriptions for the same feature in the process of correcting the gain of a camera system is reflected, on the other hand, the cameras of different types have larger response differences for the same light source, and the factors lead to a single fixed method strategy, so that the cameras of different types are difficult to flexibly adapt. In practical application, especially in a multi-camera system or a camera device for mass production, even if each camera is subjected to flat field correction, due to factors such as inherent manufacturing process differences among different sensors, and the like, the output gray values of the cameras possibly have larger response deviation under the same light source condition, so that the comparability and harmony of image data among the multiple cameras are affected, difficulties are brought to application links such as subsequent image splicing, contrast analysis, measurement judgment and the like of customers, the overall reliability of the multiple cameras is reduced, noise is increased in the process of correcting the response gray values of the cameras, and in order to realize the consistency of the response of the cameras while additionally increasing the filtering of time domain noise, the method, the system and the medium for correcting the multiple cameras are provided. Disclosure of Invention The invention aims to overcome the problems in the prior art and provide a multi-camera correction method, a multi-camera correction system and a multi-camera correction medium. In order to achieve the technical purpose and the technical effect, the invention is realized by the following technical scheme: A method for correcting a plurality of cameras, for correcting noise of the plurality of cameras after correction of responsivity, so as to implement consistency adjustment of image quality collected by each camera, the method comprising: Controlling sampling conditions, and analyzing the system gain of each camera to realize the system gain of each camera detected in the process of adjusting the light source from dark field to bright field; Based on the ratio between the standard gain and the system gain, determining the correction coefficient of each camera to correct each camera, so that the responsivity of each corrected camera is consistent; Extracting multi-frame images acquired by the same camera, calculating the time domain noise value of each camera, determining the noise correction coefficient of each camera by adopting the ratio of the standard time domain noise value to the time domain noise value of each camera, and carrying out noise filtration on the gray value of each pixel point in the images acquired by the cameras according to the noise correction coefficient of each camera so as to realize noise consistency adjustment of each camera; The time domain noise value is an average value of standard deviations corresponding to gray value variation of each pixel point in the whole image in a multi-frame image acquired by the same camera. Further, two frames of images under the same sampling condition are respectively extracted, the average gray value of the two frames of images under the same sampling condition and the variance corresponding to the gray value difference value of the two frames of images are calculated, and the ratio between the variance and the average gray value under the sampling condition corresponding to the maximum value of the variance is screened to determine the system gain of the camera. Further, based on the average gray value of the image under the sampling condition with the maximum variance, the ratio between the average gray value of the image under each sampling condition and the average gray value of the image under the sampling condition with the maximum variance is screened out, and the gray value of the image under each sampling condition is smaller than the threshold value of the ratio coefficient. Further, the method for determining the standard gain comprises the steps of screening the maximum value of the system gains of all cameras, and taking