Image quality evaluation method used in video monitoring system

Abstract

The invention relates to the technical field of image detection, in particular to an image quality evaluation method used in a video monitoring system, and solves the problems that in the prior art, dynamic image recognition cannot be carried out on an obtained monitoring video, and the feature quality of an image cannot be accurately judged through an algorithm in the image recognition process. Comprising the following steps that S1, real-time monitoring and image acquisition are conducted on pictures of multiple monitoring areas through a distributed image acquisition module, then monitoring videos are transmitted to a video storage module and a video display module through a signal transmission module, and the video storage module can conduct synchronous storage of local storage and cloud server storage. According to the method, the monitoring video is dynamically captured and intercepted, the image can be automatically acquired according to the sequence of the time periods, and the image quality can be accurately calculated and judged through an algorithm according to the characteristics of the image, so that the video retention quality can be automatically evaluated after the monitoring video is acquired.

Description

Technical Field

[0001] The present invention relates to the technical field of image detection, and in particular to an image quality evaluation method used in a video monitoring system. Background Art

[0002] Video image quality analysis is a crucial technology in security monitoring systems. By analyzing the quality of video images obtained in the monitoring system, the reliability and effectiveness of video monitoring can be improved. Images are inevitably affected by various distortions during acquisition and transmission. The introduction of distortion changes the correlation between pixels and loses the original image information to a certain extent, which will affect the accuracy of subsequent target detection and defect recognition. In order to enable the detection and recognition algorithms to identify the target more accurately and reduce errors, it is necessary to evaluate the quality of the image to be detected.

[0003] However, the existing surveillance videos cannot perform dynamic image recognition, and it is not convenient to accurately judge the feature quality of the image through algorithms during the image recognition process; therefore, it does not meet the existing needs. In this regard, we propose an image quality evaluation method for video surveillance systems. Summary of the invention

[0004] The object of the present invention is to provide an image quality evaluation method for a video monitoring system to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solution: an image quality evaluation method for a video surveillance system, comprising the following steps:

[0006] S1: Use the distributed image acquisition module to perform real-time monitoring and image acquisition of multiple monitoring areas, and then transmit the monitoring video to the video storage module and the video display module through the signal transmission module. The video storage module can synchronize local storage and cloud server storage, and give priority to video storage and backup to avoid data loss and damage during subsequent video extraction and detection;

[0007] S2: The video display module can be used to centrally display images of monitoring ports at various points. Each module of the video monitoring system is driven and controlled by the system management module, so that the system management module can adjust the monitoring angle through the distributed image acquisition module, and use the video storage module to automatically segment the stored backup video, and keep the stored video sorted according to the monitoring time;

[0008] S3: When evaluating the image quality, multiple test images are distorted, and the image distortion includes compression, defocus blur, motion blur, noise, abnormal exposure, etc., and then the test images and the distorted images are input into the feature comparison model module, and the standard image corresponding to the test image is input synchronously, and then the feature comparison model module is used to compare and identify the test images and the distorted images with the standard image;

[0009] S4: When the feature contrast model module can accurately extract the distorted image after the three images are mixed, the feature contrast model module is applied to the image quality evaluation. Specifically, the dynamic video capture module is connected to the system management module, and the dynamic video capture module is used to capture multiple segments of the video of the corresponding time period, and then the captured multiple images are input into the image quality detection module, so that the image quality detection module performs distortion detection on the image based on the feature contrast model module, and the judgment threshold of whether the image is distorted is manually input and adjusted by the feature contrast model module, so as to control the number and quality of images that meet the requirements;

[0010] S5: Distorted images are automatically eliminated and images that meet the requirements are imported into the feature extraction and recognition module, so that the feature extraction and recognition module extracts and recognizes the features in the image for easy classification. The image after the identified features is further screened and stored through the feature test screening module, thereby maintaining the image quality obtained in the video surveillance stable.

[0011] Preferably, the video surveillance system includes a distributed image acquisition module, a signal transmission module, a video storage module and a monitoring platform, the signal transmission module is composed of a wireless transmission component and a wired optical fiber, and the distributed image acquisition module is connected to the video storage module via a wireless transmission component or a wired optical fiber.

[0012] Preferably, the distributed image acquisition module is composed of a camera, a monitoring bracket and a power supply line. A wireless transmission component is provided inside the camera. The camera is fixedly connected to the monitoring bracket, and the camera is electrically connected to the power supply line.

[0013] Preferably, the video storage module includes a local memory and a cloud server, the video display module and the system management module constitute a monitoring platform, the video display module is a display, and the system management module is provided with a processor and a circuit mainboard.

[0014] Preferably, the image quality evaluation method includes a dynamic video capture module, an image quality detection module, a feature comparison model module, a feature extraction and recognition module and a feature test screening module.

[0015] Preferably, the algorithms used by the feature contrast model module to evaluate image quality are SSIM and PSNR, wherein the SSIM is structural similarity and the PSNR is peak signal-to-noise ratio.

[0016] Preferably, the SSIM is:

[0017] SSIM(x,y)=[l(x,y)*c(x,y)*s(x,y)]^a;

[0018] l(x,y), c(x,y), s(x,y) represent brightness, contrast and structural similarity respectively, and a is a weight parameter.

[0019] Preferably, the PSNR is:

[0020] PSNR(dB)=20*log10(MAX / MSE);

[0021] The judgment threshold of the feature comparison model module is that the PSNR is between 40dB and 50dB;

[0022] MSE = 1 / N*∑(i=1toN)(xi-yi)^2;

[0023] Where N represents the number of pixels, xi and yi represent the pixel values ​​in the original image and the processed image respectively;

[0024] The MSE is the square value of the difference between each pixel value in the image and the corresponding pixel value of the original signal, and the average value of the square values ​​is taken. The MAX^2 is the total signal power.

[0025] Compared with the prior art, the present invention has the following beneficial effects:

[0026] The present invention can dynamically capture and intercept surveillance videos, automatically acquire images according to the order of time periods, centrally evaluate images using a feature comparison model, and accurately calculate and judge image quality through an algorithm based on image features, thereby ensuring automatic evaluation of video retention quality after surveillance videos are acquired. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a structural schematic diagram of the video monitoring system of the present invention;

[0028] Figure 2 It is a structural schematic diagram of the image quality evaluation method of the present invention;

[0029] Figure 3 Schematic diagram of the algorithm structure of image quality evaluation of the present invention;

[0030] Figure 4 It is a schematic diagram of the overall process of the present invention. DETAILED DESCRIPTION

[0031] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0032] See also Figures 1 to 4 , an embodiment of the present invention provides: an image quality evaluation method for a video monitoring system, comprising the following steps:

[0033] S1: Use the distributed image acquisition module to perform real-time monitoring and image acquisition of multiple monitoring areas, and then transmit the monitoring video to the video storage module and the video display module through the signal transmission module. The video storage module can synchronize local storage and cloud server storage, and give priority to video storage and backup to avoid data loss and damage during subsequent video extraction and detection;

[0034] S2: The video display module can be used to centrally display images of monitoring ports at various points. Each module of the video monitoring system is driven and controlled by the system management module, so that the system management module can adjust the monitoring angle through the distributed image acquisition module, and use the video storage module to automatically segment the stored backup video, and keep the stored video sorted according to the monitoring time;

[0035] S3: When evaluating the image quality, multiple test images are distorted, and the image distortion includes compression, defocus blur, motion blur, noise, abnormal exposure, etc., and then the test images and the distorted images are input into the feature comparison model module, and the standard image corresponding to the test image is input synchronously, and then the feature comparison model module is used to compare and identify the test images and the distorted images with the standard image;

[0036] S4: When the feature contrast model module can accurately extract the distorted image after the three images are mixed, the feature contrast model module is applied to the image quality evaluation. Specifically, the dynamic video capture module is connected to the system management module, and the dynamic video capture module is used to capture multiple segments of the video of the corresponding time period, and then the captured multiple images are input into the image quality detection module, so that the image quality detection module performs distortion detection on the image based on the feature contrast model module, and the judgment threshold of whether the image is distorted is manually input and adjusted by the feature contrast model module, so as to control the number and quality of images that meet the requirements;

[0037] S5: Distorted images are automatically eliminated and images that meet the requirements are imported into the feature extraction and recognition module, so that the feature extraction and recognition module extracts and recognizes the features in the image for easy classification. The image after the identified features is further screened and stored through the feature test screening module, thereby maintaining the image quality obtained in the video surveillance stable.

[0038] See also Figure 1 The video surveillance system includes a distributed image acquisition module, a signal transmission module, a video storage module and a monitoring platform. The signal transmission module is composed of a wireless transmission component and a wired optical fiber. The distributed image acquisition module is connected to the video storage module through a wireless transmission component or a wired optical fiber. The distributed image acquisition module is composed of a camera, a monitoring bracket and a power supply line. The camera is provided with a wireless transmission component inside. The camera is fixedly connected to the monitoring bracket. The camera is electrically connected to the power supply line. The video storage module includes a local storage and a cloud server. The video display module and the system management module constitute the monitoring platform. The video display module is a display. The system management module is provided with a processor and a circuit motherboard.

[0039] See also Figure 2 and Figure 3 The image quality evaluation method includes a dynamic video capture module, an image quality detection module, a feature comparison model module, a feature extraction and recognition module, and a feature test and screening module. The feature comparison model module uses SSIM and PSNR as algorithms for image quality evaluation. SSIM stands for structural similarity, and PSNR stands for peak signal-to-noise ratio. SSIM is: SSIM(x,y)=[l(x,y)*c(x,y)*s(x,y)]^a;

[0040] l(x,y), c(x,y), s(x,y) represent brightness, contrast and structural similarity respectively, and a is a weight parameter. By using the SSIM algorithm to calculate the features of the image, it can better reflect the perception results of the human eye, and compared with other evaluation methods, the results are more objective and repeatable.

[0041] PSNR is: PSNR (dB) = 20 * log10 (MAX / MSE);

[0042] The judgment threshold of the feature comparison model module is that the PSNR is between 40dB-50dB;

[0043] MSE = 1 / N*∑(i=1toN)(xi-yi)^2;

[0044] Where N represents the number of pixels, xi and yi represent the pixel values ​​in the original image and the processed image respectively, MSE is the square value of the difference between each pixel value in the image and the corresponding pixel value of the original signal, and the average of the square values ​​is taken, MAX^2 is the total signal power, and the quality of the evaluated image can be controlled by adjusting the judgment threshold of the feature contrast model module.

[0045] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A method for evaluating image quality in a video surveillance system, characterized in that: The steps include: S1: Use the distributed image acquisition module to perform real-time monitoring and image acquisition of multiple monitoring areas, and then transmit the monitoring video to the video storage module and the video display module through the signal transmission module. The video storage module can synchronize local storage and cloud server storage, and give priority to video storage and backup to avoid data loss and damage during subsequent video extraction and detection; S2: The video display module can be used to centrally display images of monitoring ports at various points. Each module of the video monitoring system is driven and controlled by the system management module, so that the system management module can adjust the monitoring angle through the distributed image acquisition module, and use the video storage module to automatically segment the stored backup video, and keep the stored video sorted according to the monitoring time; S3: When evaluating the image quality, multiple test images are distorted, and the image distortion includes compression, defocus blur, motion blur, noise, abnormal exposure, etc., and then the test images and the distorted images are input into the feature comparison model module, and the standard image corresponding to the test image is input synchronously, and then the feature comparison model module is used to compare and identify the test images and the distorted images with the standard image; S4: When the feature contrast model module can accurately extract the distorted image after the three images are mixed, the feature contrast model module is applied to the image quality evaluation. Specifically, the dynamic video capture module is connected to the system management module, and the dynamic video capture module is used to capture multiple segments of the video of the corresponding time period, and then the captured multiple images are input into the image quality detection module, so that the image quality detection module performs distortion detection on the image based on the feature contrast model module, and the judgment threshold of whether the image is distorted is manually input and adjusted by the feature contrast model module, so as to control the number and quality of images that meet the requirements; S5: Distorted images are automatically eliminated and images that meet the requirements are imported into the feature extraction and recognition module, so that the feature extraction and recognition module extracts and recognizes the features in the image for easy classification. The image after the identified features is further screened and stored through the feature test screening module, thereby maintaining the image quality obtained in the video surveillance stable.

2. The image quality evaluation method for a video surveillance system according to claim 1, characterized in that: The video monitoring system includes a distributed image acquisition module, a signal transmission module, a video storage module and a monitoring platform. The signal transmission module is composed of a wireless transmission component and a wired optical fiber. The distributed image acquisition module is connected to the video storage module via a wireless transmission component or a wired optical fiber.

3. The image quality evaluation method for a video surveillance system according to claim 2, characterized in that: The distributed image acquisition module is composed of a camera, a monitoring bracket and a power supply line. A wireless transmission component is provided inside the camera. The camera is fixedly connected to the monitoring bracket, and the camera is electrically connected to the power supply line.

4. The image quality evaluation method for a video surveillance system according to claim 3, characterized in that: The video storage module includes a local memory and a cloud server. The video display module and the system management module constitute a monitoring platform. The video display module is a display. The system management module is provided with a processor and a circuit mainboard.

5. The image quality evaluation method for a video surveillance system according to claim 4, characterized in that: The image quality evaluation method comprises a dynamic video interception module, an image quality detection module, a feature comparison model module, a feature extraction and recognition module and a feature test and screening module.

6. The image quality evaluation method for a video surveillance system according to claim 5, characterized in that: The algorithms used by the feature contrast model module to evaluate image quality are SSIM and PSNR, where SSIM stands for structural similarity and PSNR stands for peak signal-to-noise ratio.

7. The image quality evaluation method for a video surveillance system according to claim 6, characterized in that: The SSIM is: SSIM(x,y)=[l(x,y)*c(x,y)*s(x,y)]^a; l(x,y), c(x,y), s(x,y) represent brightness, contrast and structural similarity respectively, and a is a weight parameter.

8. The image quality evaluation method for a video surveillance system according to claim 7, characterized in that: The PSNR is: PSNR(dB)=20*log10(MAX / MSE); The judgment threshold of the feature comparison model module is that the PSNR is between 40dB and 50dB; MSE = 1 / N*∑(i=1toN)(xi-yi)^2; Where N represents the number of pixels, xi and yi represent the pixel values ​​in the original image and the processed image respectively; The MSE is the square value of the difference between each pixel value in the image and the corresponding pixel value of the original signal, and the average value of the square values ​​is taken. The MAX^2 is the total signal power.

Images