CN-121998885-A - Video monitoring-oriented low-illumination image quality self-adaptive enhancement method and system

CN121998885ACN 121998885 ACN121998885 ACN 121998885ACN-121998885-A

Abstract

The embodiment of the invention provides a low-illumination image quality self-adaptive enhancement method and system for video monitoring, and belongs to the field of image processing. The method comprises the steps of determining an initial monitoring video, obtaining a key frame image corresponding to the initial monitoring video, performing deblurring treatment on the key frame image by utilizing entropy value information corresponding to the key frame image and introducing illumination evaluation parameters to obtain a deblurring image corresponding to the key frame image, performing uncertainty analysis on the deblurring image to obtain a target fuzzy value corresponding to the deblurring image, performing image adjustment on the deblurring image according to the target fuzzy value to obtain a corresponding uncertainty image, performing low-light image enhancement treatment on the uncertainty image to obtain a target enhanced image, performing image edge monitoring on the target enhanced image to obtain target edge points and target accuracy corresponding to the target edge points, and performing image enhancement adjustment on the target enhanced image according to the target accuracy until the target enhanced image meeting preset accuracy is obtained.

Inventors

WANG YANFENG
XIAO SHIFAN
CAI DAFU
LIU XIN
CHEN YINING
HUANG SHENGQIANG

Assignees

广东唯康教育科技股份有限公司

Dates

Publication Date: 20260508
Application Date: 20260320

Claims (10)

1. The low-illumination image quality self-adaptive enhancement method for video monitoring is characterized by comprising the following steps of: Determining an initial monitoring video and obtaining a key frame image corresponding to the initial monitoring video; performing deblurring processing on the key frame image by utilizing entropy value information corresponding to the key frame image and introducing illumination evaluation parameters to obtain a deblurred image corresponding to the key frame image; Performing uncertainty analysis on the deblurred image to obtain a target fuzzy value corresponding to the deblurred image, and performing image adjustment on the deblurred image according to the target fuzzy value to obtain a corresponding uncertainty image; Performing low-light image enhancement processing on the uncertain image to obtain a target enhanced image, and performing image edge monitoring on the target enhanced image to obtain a target edge point and target accuracy corresponding to the target edge point; And carrying out image enhancement adjustment on the target enhanced image according to the target accuracy until the target enhanced image meeting the preset accuracy is obtained.
2. The method according to claim 1, wherein the performing deblurring processing on the key frame image by using entropy information corresponding to the key frame image and introducing illumination evaluation parameters to obtain a deblurred image corresponding to the key frame image includes: performing entropy information analysis on the key frame image to obtain the target brightness region corresponding to the key frame image; mapping the key frame image into a target coordinate system taking the target brightness area as a center to obtain target coordinate information of the key frame image in the target coordinate system; calculating distance information between each first sub-pixel in the key frame image and the target brightness area, and determining initial irradiance corresponding to the first sub-pixel according to the distance information; carrying out data clustering on the key frame images according to the initial irradiance and the target coordinate information to obtain a target cluster; performing reliability parameter calculation on each sub-class cluster in the target cluster to obtain the illumination evaluation parameters corresponding to the sub-class clusters; Optimizing the initial irradiance in the sub-cluster according to the illumination evaluation parameter to obtain target irradiance corresponding to the sub-cluster; and performing deblurring processing on the key frame image according to the target irradiance and the target brightness region to obtain the deblurred image corresponding to the key frame image.
3. The method according to claim 2, wherein the performing entropy information analysis on the key frame image to obtain the target luminance area corresponding to the key frame image includes: Performing target identification on the key frame image to obtain target structure information corresponding to the key frame image; Performing brightness analysis on the target structure information to obtain row brightness information and column brightness information corresponding to the target structure information; carrying out neighborhood analysis according to the row brightness information and the column brightness information to obtain an initial brightness region; and carrying out entropy information analysis on the initial brightness region to obtain the target brightness region.
4. The method of claim 2, wherein optimizing the initial irradiance in the sub-cluster according to the illumination evaluation parameter to obtain a target irradiance corresponding to the sub-cluster comprises: Determining a preset evaluation parameter, and determining irradiance to be optimized and first irradiance which does not need to be optimized in the sub-cluster according to the illumination evaluation parameter and the preset evaluation parameter; extracting edge characteristics of the key frame image to obtain an edge characteristic image corresponding to the key frame image; generating a target guide image according to the edge feature image, and performing filtering processing on the irradiance to be optimized according to the target guide image to obtain second irradiance corresponding to the irradiance to be optimized; and determining the target irradiance corresponding to the sub-cluster according to the first irradiance and the second irradiance.
5. The method according to claim 1, wherein the performing uncertainty analysis on the deblurred image to obtain a target blur value corresponding to the deblurred image includes: Normalizing the deblurred image to obtain a normalized image; Determining a fuzzy control parameter, and performing membership function calculation on each second sub-pixel in the normalized image according to the fuzzy control parameter to obtain a first data value corresponding to the second sub-pixel; Performing non-membership function calculation on each second sub-pixel in the normalized image according to the fuzzy control parameters to obtain a second data value corresponding to the second sub-pixel; Fusing the first data value and the second data value to determine the target fuzzy value corresponding to the second sub-pixel; wherein the first data value and the second data value are obtained according to the following formula: ; ; Wherein, the Representing the first data value corresponding to the second sub-pixel when the horizontal position is i and the vertical position is j; C, d, f represent constants; representing a pixel value corresponding to the second sub-pixel under the normalized image when the horizontal position is i and the vertical position is j; and representing the second data value corresponding to the second sub-pixel when the horizontal position is i and the vertical position is j.
6. The method of claim 5, wherein the fusing the first data value and the second data value to determine the target blur value corresponding to the second subpixel comprises: performing cubic calculation on the first data value to obtain a first numerical value, and performing cubic calculation on the second data value to obtain a second numerical value; and calculating the sum value between the first numerical value and the second numerical value to obtain a target value, and determining the target fuzzy value corresponding to the second sub-pixel according to the target value.
7. The method of claim 1, wherein performing low-light image enhancement processing on the uncertain image to obtain a target enhanced image comprises: Performing wavelet transformation processing on the uncertain image to obtain an initial low-frequency wavelet coefficient corresponding to the uncertain image; Determining a screening threshold value, and performing reduced range processing on the initial high-frequency wavelet coefficient according to the screening threshold value to obtain a target high-frequency wavelet coefficient; performing image enhancement processing on the uncertain image according to the target high-frequency wavelet coefficient to obtain a first image; performing blurring processing on the initial low-frequency wavelet coefficient to obtain a target low-frequency wavelet coefficient; performing image enhancement processing on the uncertain image according to the target low-frequency wavelet coefficient to obtain a second image; and carrying out image fusion according to the first image and the second image to obtain the target enhanced image.
8. The method of claim 7, wherein said performing image enhancement processing on said uncertain image according to said target low frequency wavelet coefficients to obtain a second image comprises: Determining a pixel minimum value and a pixel maximum value corresponding to the uncertain image, and determining a pixel membership degree corresponding to each pixel position in the uncertain image according to the pixel minimum value and the pixel maximum value; Obtaining relevant pixels corresponding to each pixel position under a target window from the uncertain image, and calculating the average membership corresponding to the pixel position according to the relevant pixels; determining an adjusting factor, and determining the pixel contrast corresponding to the pixel position by using the pixel membership degree and the average membership degree according to the adjusting factor; Performing fuzzy enhancement on the pixel position according to the pixel contrast and the average membership to obtain an enhanced membership corresponding to the pixel position; determining an enhanced pixel value corresponding to the pixel position according to the pixel maximum value and the pixel minimum value and the enhanced membership degree; The second image is obtained from the enhanced pixel values and the pixel locations.
9. The method of claim 8, wherein the blurring enhancement is performed on the pixel location according to the pixel contrast and the mean membership to obtain an enhanced membership corresponding to the pixel location, comprising: comparing the average membership with the pixel membership to obtain a target comparison result; when the target comparison result meets a first preset result, calculating a difference value result between the pixel contrast and a preset constant to obtain a first difference value, and calculating a summation result between the pixel contrast and the preset constant to obtain a target sum value; Calculating a first product between the mean membership and the first difference value, and calculating a division result between the first product and the target sum value to obtain the enhanced membership corresponding to the pixel position; When the target comparison result meets a second preset result, calculating a difference value result between the pixel contrast and the preset constant to obtain the first difference value, and calculating a summation result between the pixel contrast and the preset constant to obtain the target sum value; calculating a difference result between the average membership and the preset constant to obtain a second difference; Calculating a product result between the first difference value and the second difference value to obtain a second product, and calculating a difference result between the preset constant value and the second product to obtain a third difference value; and carrying out division operation according to the third difference value and the target sum value to obtain the enhanced membership degree corresponding to the pixel position.
10. A low-light image quality self-adaptive enhancement system for video monitoring, the system comprising: the image acquisition module is used for determining an initial monitoring video and acquiring a key frame image corresponding to the initial monitoring video; the image processing module is used for performing deblurring processing on the key frame image by utilizing entropy value information corresponding to the key frame image and introducing illumination evaluation parameters to obtain a deblurred image corresponding to the key frame image; The image adjustment module is used for carrying out uncertainty analysis on the deblurred image to obtain a target fuzzy value corresponding to the deblurred image, and carrying out image adjustment on the deblurred image according to the target fuzzy value to obtain a corresponding uncertain image; The edge detection module is used for carrying out low-light image enhancement processing on the uncertain image to obtain a target enhanced image, and carrying out image edge monitoring on the target enhanced image to obtain a target edge point and target accuracy corresponding to the target edge point; And the enhancement adjusting module is used for carrying out image enhancement adjustment on the target enhancement image according to the target accuracy until the target enhancement image meeting the preset accuracy is obtained.

Description

Video monitoring-oriented low-illumination image quality self-adaptive enhancement method and system Technical Field The invention relates to the field of image processing, in particular to a low-illumination image quality self-adaptive enhancement method and system for video monitoring. Background Along with the progress of artificial intelligence technology, the research and education integration of obstetrics is deepened continuously, and artificial intelligence is also gradually applied to driving the development of vocational education. In order to ensure the teaching quality, the professional education usually records the learning process of students and the teaching activities of teachers by means of video monitoring, and evaluates the teaching effects and learning results by video analysis. However, in the video acquisition link, due to inconsistent ambient lighting conditions, the video quality is uneven, so that the accuracy of video analysis is reduced, and the reliability of vocational education evaluation and management is affected. Disclosure of Invention The embodiment of the invention mainly aims to provide a low-illumination image quality self-adaptive enhancement method and system for video monitoring, which aim to solve the problems that in the prior art, due to inconsistent illumination conditions of an acquired monitoring video environment, video quality is uneven, and further, the accuracy of video analysis is reduced, so that the evaluation and management reliability of professional education are affected. In a first aspect, an embodiment of the present invention provides a low-illumination image quality adaptive enhancement method for video monitoring, including: Determining an initial monitoring video and obtaining a key frame image corresponding to the initial monitoring video; performing deblurring processing on the key frame image by utilizing entropy value information corresponding to the key frame image and introducing illumination evaluation parameters to obtain a deblurred image corresponding to the key frame image; Performing uncertainty analysis on the deblurred image to obtain a target fuzzy value corresponding to the deblurred image, and performing image adjustment on the deblurred image according to the target fuzzy value to obtain a corresponding uncertainty image; Performing low-light image enhancement processing on the uncertain image to obtain a target enhanced image, and performing image edge monitoring on the target enhanced image to obtain a target edge point and target accuracy corresponding to the target edge point; And carrying out image enhancement adjustment on the target enhanced image according to the target accuracy until the target enhanced image meeting the preset accuracy is obtained. In a second aspect, an embodiment of the present invention provides a low-illumination image quality adaptive enhancement system for video monitoring, including: the image acquisition module is used for determining an initial monitoring video and acquiring a key frame image corresponding to the initial monitoring video; the image processing module is used for performing deblurring processing on the key frame image by utilizing entropy value information corresponding to the key frame image and introducing illumination evaluation parameters to obtain a deblurred image corresponding to the key frame image; The image adjustment module is used for carrying out uncertainty analysis on the deblurred image to obtain a target fuzzy value corresponding to the deblurred image, and carrying out image adjustment on the deblurred image according to the target fuzzy value to obtain a corresponding uncertain image; The edge detection module is used for carrying out low-light image enhancement processing on the uncertain image to obtain a target enhanced image, and carrying out image edge monitoring on the target enhanced image to obtain a target edge point and target accuracy corresponding to the target edge point; And the enhancement adjusting module is used for carrying out image enhancement adjustment on the target enhancement image according to the target accuracy until the target enhancement image meeting the preset accuracy is obtained. The embodiment of the invention provides a low-illumination image quality self-adaptive enhancement method and system for video monitoring, wherein the method comprises the steps of determining an initial monitoring video, obtaining a key frame image corresponding to the initial monitoring video, performing deblurring treatment on the key frame image by utilizing entropy value information corresponding to the key frame image and introducing illumination evaluation parameters, obtaining a deblurring image corresponding to the key frame image, further improving blur retention key information caused by uneven illumination or motion in a pertinence manner, performing low-light image enhancement treatment introduced, directly coping with challenges of une