Smart community public safety monitoring system and method

Abstract

The invention relates to the technical field of community public safety monitoring, and particularly discloses an intelligent community public safety monitoring system and method, the total number of monitoring cameras in a community is acquired, an image acquisition command is sent, image frames are continuously acquired by each monitoring camera, and an image sequence of each monitoring camera is formed; performing image processing on each frame in the image sequence of each monitoring camera, calculating a definition evaluation index and a sharpness evaluation index of each frame image corresponding to each monitoring camera, and screening out each abnormal frame image of each monitoring camera; performing deep analysis on each abnormal frame image marked as each monitoring camera, extracting a light parameter of each monitoring camera, and further comprehensively judging a lens smudginess index of each monitoring camera; according to the method and the device, the lens state of the camera can be effectively monitored, the cleaning early warning can be timely sent out, the quality and the safety of a monitored image are ensured, and the risk of a monitoring blind area caused by lens smudginess is effectively reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of community public safety monitoring, and relates to a smart community public safety monitoring system and method. Background Art

[0002] In modern community management, assessing the degree of camera lens contamination is a crucial step in ensuring the effective operation of security surveillance systems. First, cameras are critical to community security, and their image quality directly impacts monitoring effectiveness and timely response to security incidents. If the lens is covered with dust, dirt, or other substances, image clarity and contrast will be significantly reduced, leading to blind spots and increasing safety hazards. Furthermore, dirty lenses can cause false alarms, disrupting normal community management. Therefore, regularly assessing and maintaining camera lens cleanliness not only improves the reliability of the surveillance system but also protects the safety and property of community residents.

[0003] However, current technologies for assessing camera lens contamination still have some flaws and drawbacks. First, many assessment methods rely solely on image integrity to determine the degree of camera lens contamination. This can lead to inaccuracies, as it is difficult to distinguish between contamination and the effects of weather or lighting changes. These methods are also susceptible to interference from environmental conditions and lack the ability to detect subtle contamination. Light analysis methods, on the other hand, can improve recognition accuracy through multi-dimensional analysis, such as contrast, clarity, and color changes, providing more consistent and accurate assessment results, effectively addressing the limitations of traditional methods. Summary of the Invention

[0004] In view of the above problems existing in the prior art, the present invention provides a smart community public safety monitoring system and method for solving the above technical problems.

[0005] In order to achieve the above-mentioned and other purposes, the technical solutions adopted by the present invention are as follows: In one aspect, the present invention provides a method for monitoring public safety in a smart community, the method comprising the following steps: Obtain the total number P of surveillance cameras within the community, send an image capture command, and have each surveillance camera continuously capture image frames to form an image sequence for each surveillance camera; Perform image processing on each frame in the image sequence of each surveillance camera and calculate the clarity evaluation index of each frame of the image corresponding to each surveillance camera and sharpness evaluation index , thereby filtering out each abnormal frame image of each surveillance camera; Perform in-depth analysis on each abnormal frame image marked as each surveillance camera, extract the light parameters of each surveillance camera, and then comprehensively determine the lens dirtiness index of each surveillance camera; Based on the lens dirtiness index of each surveillance camera, a corresponding cleaning warning signal is generated, and maintenance personnel are notified through the smart community management platform to clean the lens.

[0006] Calculate the clarity evaluation index and sharpness evaluation index of each surveillance camera corresponding to each frame of image. The specific calculation process is as follows: Convert each frame image corresponding to each surveillance camera into a grayscale image, and amplify it until the frame is displayed, and obtain the total number of row frame frames and column frame frames corresponding to each frame image of each surveillance camera; The pixel brightness value of each surveillance camera corresponding to each frame image at the coordinate (x, y) is calculated , x is the number of each row frame, x=1,2,...M, y is the number of each column frame, y=1,2,...N, M and N represent the total number of row frames and column frames respectively; is the grayscale image pixel value of the k-th frame image corresponding to the i-th surveillance camera at the coordinate (x, y), i is the number of the surveillance camera, i=1,2,...P,k is the number of each frame image; The clarity evaluation index of each frame image corresponding to each surveillance camera is calculated from this ; Use the Sobel operator to calculate the gradient amplitude of each frame image corresponding to each surveillance camera at the coordinate (x, y) ,in They represent the horizontal gradient and vertical gradient of the k-th frame image corresponding to the i-th surveillance camera at the coordinate (x, y); Then calculate the sharpness evaluation index of each frame image corresponding to each surveillance camera .

[0007] The filtering logic for filtering out abnormal frame images of each surveillance camera is as follows: Set the clarity evaluation index threshold and sharpness evaluation index threshold, respectively, and record them as QX and RD; If the clarity evaluation index of a surveillance camera corresponding to a frame of image If the value is less than the set clarity evaluation index threshold QX, the frame image is judged as a frame image with abnormal clarity; In this screening method, the abnormal frame images of each surveillance camera with each definition are obtained; If the sharpness evaluation index of a surveillance camera corresponding to a frame of image If the value is less than the set sharpness evaluation index threshold RD, the frame image is determined to be a sharpness abnormal frame image; In this screening method, each sharpness abnormal frame image of each surveillance camera is obtained; The abnormal definition frame images and the abnormal sharpness frame images of each surveillance camera are integrated to obtain the abnormal frame images of each surveillance camera.

[0008] The light parameters of each surveillance camera include light refraction change value, light scattering degree and light transmission loss value.

[0009] The specific extraction logic for extracting the light refraction change value of each surveillance camera is as follows: Arrange the abnormal frame images corresponding to each surveillance camera in sequence according to the shooting timestamp, and match the feature points of all abnormal frame images corresponding to each surveillance camera, thereby calculating the average displacement value of the feature points corresponding to each surveillance camera in all abnormal frame images: ; is the average displacement value of the feature point corresponding to the i-th surveillance camera in all abnormal frame images, are the position coordinates of the feature point corresponding to the i-th surveillance camera in the g+1-th abnormal frame image and the g-th abnormal frame image, respectively. g is the number of each abnormal frame image, g=1,2,...Q, Q is the total number of abnormal frame images; The average displacement value of the feature points corresponding to each surveillance camera in all abnormal frame images is converted into the light refraction change value, thereby obtaining the light refraction change value of each surveillance camera , σ is the set proportional coefficient.

[0010] The specific extraction logic for extracting the light transmission loss value of each surveillance camera is as follows: Obtain the cleaning date corresponding to the last cleaning of each surveillance camera; and simultaneously obtain the shooting timestamp of each abnormal frame image corresponding to each surveillance camera, and extract from the database the shooting image corresponding to the cleaning date of the last cleaning of each surveillance camera corresponding to the same timestamp, and simply record it as the reference image of each abnormal frame image corresponding to each surveillance camera; Convert each abnormal frame image and the corresponding reference image of each surveillance camera into a grayscale image, and calculate the average brightness of each abnormal frame image and the corresponding reference image of each surveillance camera. Subtract the average brightness of the corresponding reference image from the average brightness of each abnormal frame image of each surveillance camera to obtain the brightness difference value between each abnormal frame image and the corresponding reference image of each surveillance camera; The calcHist function is used to calculate the color histogram of each abnormal frame image and the corresponding reference image of each surveillance camera, thereby calculating the color histogram difference between each abnormal frame image and the corresponding reference image of each surveillance camera. , where j is the number of each color channel in the histogram, j=1,2,...N, N is the total number of color channels, They represent the histogram values of the g-th abnormal frame image and the corresponding reference image on color channel j of the i-th surveillance camera, respectively. They represent the average values of the histograms of the g-th abnormal frame image and the corresponding reference image of the i-th surveillance camera, respectively, and are used for normalization processing; Calculate the light transmission loss value of each surveillance camera , g is the number of each abnormal frame image, g=1,2,...Q, Q is the total number of abnormal frame images, α1 and α2 represent the set weight coefficients respectively; is the brightness difference between each abnormal frame image and the corresponding reference image of each surveillance camera.

[0011] The specific extraction logic for extracting the light scattering degree of each surveillance camera is as follows: Divide each abnormal frame image corresponding to each surveillance camera into F image sub-blocks; thereby collect the maximum pixel value and the minimum pixel value of each image sub-block in each abnormal frame image corresponding to each surveillance camera; and use the character and Instead; use the contrast calculation formula , calculate the contrast of each abnormal frame image corresponding to each surveillance camera , f is the number of each image sub-block, f=1,2,...F; Use Laplace transform to calculate the clarity of each abnormal frame image corresponding to each surveillance camera ; Then estimate the light scattering degree of each surveillance camera , ω1, ω2 and ω3 represent predetermined weight factors respectively.

[0012] Determine the lens dirtiness index of each surveillance camera, including: The light refraction change value of each surveillance camera is normalized to obtain the normalized light refraction change value of each surveillance camera, which is recorded as ; Similarly, the normalized light scattering degree and light transmission loss value of each surveillance camera are obtained; they are recorded as ; The lens dirtiness index of each surveillance camera is calculated from this .

[0013] Another aspect of the present invention provides a smart community public safety monitoring system, which includes a camera image sequence acquisition module, an abnormal frame image screening module, a lens dirt index judgment module, and a cleaning warning signal generation module. The above modules are connected by wired and / or wireless connections to achieve data transmission between the modules; Camera image sequence acquisition module: obtains the total number P of surveillance cameras in the community, sends an image acquisition command, and each surveillance camera continuously acquires image frames to form an image sequence of each surveillance camera; Abnormal frame image screening module: performs image processing on each frame in the image sequence of each surveillance camera, calculates the clarity evaluation index and sharpness evaluation index of each frame image corresponding to each surveillance camera, and thus screens out the abnormal frame images of each surveillance camera; Lens dirt index judgment module: This module performs in-depth analysis on each abnormal frame image marked as belonging to each surveillance camera, extracts the light parameters of each surveillance camera, and then comprehensively judges the lens dirt index of each surveillance camera; Cleaning warning signal generation module: Generates corresponding cleaning warning signals based on the lens dirtiness index of each surveillance camera, and notifies maintenance personnel to clean the lens through the smart community management platform.

[0014] As described above, the smart community public security monitoring system and method provided by the present invention have at least the following beneficial effects: This invention provides a smart community public safety monitoring system and method. By systematically analyzing and processing image sequences from surveillance cameras, we can not only improve the quality and reliability of the monitoring system but also enable proactive maintenance of the equipment. This method has significant practical application value and necessity, and can provide strong support for the construction and development of smart communities. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0016] Figure 1 It is a schematic diagram of the connection of each step of the method of the present invention.

[0017] Figure 2 Schematic diagram of the connection of various modules of the system of the present invention. DETAILED DESCRIPTION

[0018] The above contents described below in conjunction with the implementation of the present invention are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the claims, they shall fall within the scope of protection of the present invention.

[0019] Example 1 The present invention is based on the premise that environmental correction has been fully completed to ensure the accuracy and reliability of the results. Environmental correction is an important step that aims to eliminate the interference caused by changes in external conditions on image analysis.

[0020] See also Figure 1 As shown, a smart community public safety monitoring method includes the following steps: Obtain the total number P of surveillance cameras within the community, send an image capture command, and have each surveillance camera continuously capture image frames to form an image sequence for each surveillance camera; Perform image processing on each frame in the image sequence of each surveillance camera and calculate the clarity evaluation index of each frame of the image corresponding to each surveillance camera and sharpness evaluation index , thereby filtering out each abnormal frame image of each surveillance camera; Calculate the clarity evaluation index and sharpness evaluation index of each surveillance camera corresponding to each frame of image. The specific calculation process is as follows: Convert each frame image corresponding to each surveillance camera into a grayscale image, and amplify it until the frame is displayed, and obtain the total number of row frame frames and column frame frames corresponding to each frame image of each surveillance camera; The pixel brightness value of each surveillance camera corresponding to each frame image at the coordinate (x, y) is calculated , x is the number of each row frame, x=1,2,...M, y is the number of each column frame, y=1,2,...N, M and N represent the total number of row frames and column frames respectively; is the grayscale image pixel value of the k-th frame image corresponding to the i-th surveillance camera at the coordinate (x, y), i is the number of the surveillance camera, i=1,2,...P,k is the number of each frame image; The clarity evaluation index of each frame image corresponding to each surveillance camera is calculated from this ; Use the Sobel operator to calculate the gradient amplitude of each frame image corresponding to each surveillance camera at the coordinate (x, y) ,in They represent the horizontal gradient and vertical gradient of the k-th frame image corresponding to the i-th surveillance camera at the coordinate (x, y); Then calculate the sharpness evaluation index of each frame image corresponding to each surveillance camera .

[0021] The filtering logic for filtering out abnormal frame images of each surveillance camera is as follows: Set the clarity evaluation index threshold and sharpness evaluation index threshold, respectively, and record them as QX and RD; If the clarity evaluation index of a surveillance camera corresponding to a frame of image If the value is less than the set clarity evaluation index threshold QX, the frame image is judged as a frame image with abnormal clarity; In this screening method, the abnormal frame images of each surveillance camera with each definition are obtained; If the sharpness evaluation index of a surveillance camera corresponding to a frame of image If the value is less than the set sharpness evaluation index threshold RD, the frame image is determined to be a sharpness abnormal frame image; In this screening method, each sharpness abnormal frame image of each surveillance camera is obtained; The abnormal definition frame images and the abnormal sharpness frame images of each surveillance camera are integrated to obtain the abnormal frame images of each surveillance camera.

[0022] Perform in-depth analysis on each abnormal frame image marked as each surveillance camera, extract the light parameters of each surveillance camera, and then comprehensively determine the lens dirtiness index of each surveillance camera; The light parameters of each surveillance camera include light refraction change value, light scattering degree and light transmission loss value.

[0023] The specific extraction logic for extracting the light refraction change value of each surveillance camera is as follows: Arrange the abnormal frame images corresponding to each surveillance camera in sequence according to the shooting timestamp, and match the feature points of all abnormal frame images corresponding to each surveillance camera, thereby calculating the average displacement value of the feature points corresponding to each surveillance camera in all abnormal frame images: ; is the average displacement value of the feature point corresponding to the i-th surveillance camera in all abnormal frame images, are the position coordinates of the feature point corresponding to the i-th surveillance camera in the g+1-th abnormal frame image and the g-th abnormal frame image, respectively. g is the number of each abnormal frame image, g=1,2,...Q, Q is the total number of abnormal frame images; The average displacement value of the feature points corresponding to each surveillance camera in all abnormal frame images is converted into the light refraction change value, thereby obtaining the light refraction change value of each surveillance camera , σ is the set proportional coefficient.

[0024] The process of converting average displacement values into light refraction change values involves a key assumption: the displacement of feature points is mainly caused by the refraction change of light in different media. The rationality and application of this assumption need to be analyzed from multiple perspectives: First, light refraction is the change in direction of light as it propagates between different media due to the different refractive indices of the media. This change in refractive index causes a shift in the light path, which can be seen as a shift in the position of feature points in the image. By observing the displacement of feature points in the image, we can infer that the light path has changed in the medium. The key to using the proportionality factor σ to convert the average displacement of a feature point into a refractive index change is to establish a quantitative relationship between displacement and refractive index change. This relationship can be calibrated using experimental data. In practical applications, we can measure the displacement of the feature point under experimental conditions where the refractive index change is known (for example, under different temperatures or pressures) to determine the value of σ. The benefit of this method is that it provides a non-invasive, image-analysis-based means of estimating refractive index changes. Compared to physical methods that directly measure the refractive index (such as using a refractometer), image-based analysis does not require additional equipment installation and maintenance. This is particularly useful when measurements need to be made in remote or inaccessible environments. In addition, through image sequence analysis, real-time monitoring of refractive index changes can be achieved, helping to identify dynamic changes in environmental conditions. In practical applications, this method is essential because it can provide a relatively simple means of monitoring changes in environmental conditions, especially in situations where direct measurement of the refractive index is not possible.

[0025] The specific extraction logic for extracting the light transmission loss value of each surveillance camera is as follows: Obtain the cleaning date corresponding to the last cleaning of each surveillance camera; and simultaneously obtain the shooting timestamp of each abnormal frame image corresponding to each surveillance camera, and extract from the database the shooting image corresponding to the cleaning date of the last cleaning of each surveillance camera corresponding to the same timestamp, and simply record it as the reference image of each abnormal frame image corresponding to each surveillance camera; It also includes: if the light intensity of the cleaning date corresponding to the last cleaning of each surveillance camera is not within the same intensity range as the light intensity of the day, then the light intensity of each date within five days after the cleaning date of the last cleaning is obtained; if the light intensity of each date within five days after the cleaning date of the last cleaning is not within the same intensity range as the light intensity of the day, then the light intensity of the cleaning date of the previous cleaning is screened for comparison; if the comparison is inconsistent, then the light intensity of each date within five days after the cleaning date of the previous cleaning is obtained for comparison, and so on, until the light intensity of the cleaning date within the cleaning date segment is within the same intensity range as the light intensity of the day; wherein the cleaning date segment is represented as starting with the cleaning date and ending with each date within five days after the cleaning date of the last cleaning, and the obtained date segment is recorded as the cleaning date segment.

[0026] Convert each abnormal frame image and the corresponding reference image of each surveillance camera into a grayscale image, and calculate the average brightness of each abnormal frame image and the corresponding reference image of each surveillance camera. Subtract the average brightness of the corresponding reference image from the average brightness of each abnormal frame image of each surveillance camera to obtain the brightness difference value between each abnormal frame image and the corresponding reference image of each surveillance camera; The calcHist function is used to calculate the color histogram of each abnormal frame image and the corresponding reference image of each surveillance camera, thereby calculating the color histogram difference between each abnormal frame image and the corresponding reference image of each surveillance camera. , where j is the number of each color channel in the histogram, j=1,2,...N, N is the total number of color channels, They represent the histogram values of the g-th abnormal frame image and the corresponding reference image on color channel j of the i-th surveillance camera, respectively. They represent the average values of the histograms of the g-th abnormal frame image and the corresponding reference image of the i-th surveillance camera, respectively, and are used for normalization processing. N is usually set to 365.

[0027] When evaluating transmission loss, changes in the color histogram are key indicators. Transmission loss typically results in color shifts and reduced brightness, which manifest as a shift in the histogram distribution. The above formula quantifies this shift, providing an objective measure for calculating transmission loss. Therefore, applying this method not only improves processing accuracy but also provides essential technical support for image analysis in complex environments.

[0028] In the above calculation formula represents the probability distribution of pixels on each color channel, It calculates the overlap between the two distributions and measures the similarity of the two image histograms by taking the square root of the probability product of each color channel and summing them up.

[0029] Calculate the light transmission loss value of each surveillance camera , g is the number of each abnormal frame image, g=1,2,...Q, Q is the total number of abnormal frame images, α1 and α2 represent the set weight coefficients respectively; is the brightness difference between each abnormal frame image and the corresponding reference image of each surveillance camera.

[0030] The weighted summation of brightness difference and color histogram difference to calculate light transmission loss is based on the fact that they can comprehensively reflect image changes caused by transmission loss. Brightness difference captures overall lighting changes, while color histogram difference reveals changes in color distribution due to transmission or contamination. This weighted summation combines these two pieces of information into a single metric, providing a more accurate assessment of light transmission loss. This method benefits from comprehensively considering multiple influencing factors, improving the reliability and accuracy of the assessment results. It is an essential step for effectively monitoring light transmission loss in practical applications.

[0031] The specific extraction logic for extracting the light scattering degree of each surveillance camera is as follows: Divide each abnormal frame image corresponding to each surveillance camera into F image sub-blocks; thereby collect the maximum pixel value and the minimum pixel value of each image sub-block in each abnormal frame image corresponding to each surveillance camera; and use the character and Instead; use the contrast calculation formula , calculate the contrast of each abnormal frame image corresponding to each surveillance camera , f is the number of each image sub-block, f=1,2,...F; Use Laplace transform to calculate the clarity of each abnormal frame image corresponding to each surveillance camera ; Then estimate the light scattering degree of each surveillance camera , ω1, ω2 and ω3 represent predetermined weight factors respectively.

[0032] Light scattering refers to the deflection and diffusion of light as it encounters particles or inhomogeneous media during propagation. This phenomenon can significantly affect image quality. By analyzing image contrast, clarity, and color variations, the degree of light scattering can be reasonably estimated. The rationale and necessity of this method can be explained from several perspectives: First, contrast is a measure of the difference between bright and dark areas in an image. Light scattering causes light to be deflected and diffused multiple times before entering the camera, making the bright and dark areas of the image unclear and reducing the contrast. Therefore, by measuring the change in image contrast, the impact of light scattering on the image can be effectively reflected. Clarity is related to the details and edge sharpness of the image. Scattering will blur the edges of the image because after the light is scattered, the originally focused light is dispersed, resulting in the loss of image details. Therefore, using methods such as the Laplace operator to evaluate the change in image clarity can further capture the impact of scattering on image details; Secondly, light scattering can cause changes in the color distribution of an image, such as color shift and saturation loss. The extent of this color change can be quantified by calculating the difference in the color histogram between each abnormal frame image and the corresponding reference image from each surveillance camera. This method is reasonable because it not only considers changes in the mean color but also changes in the shape of the color distribution, providing a comprehensive color difference metric. In real-world applications, the rationality of these methods is supported by extensive data. For example, in atmospheric science and environmental monitoring, researchers assess air quality and haze levels by changes in contrast and clarity. These studies have shown that there is a significant negative correlation between light scattering and contrast and clarity; at the same time, it provides a contactless, automated means of assessing light scattering that is suitable for image analysis under various lighting and environmental conditions. By comprehensively considering contrast, clarity, and color changes, a more comprehensive and accurate estimate of light scattering can be obtained. In addition, this method does not rely on complex equipment and expensive measuring instruments, and is highly operational and cost-effective.

[0033] The weighted summation of contrast, clarity, and color histogram differences to calculate the degree of light scattering is based on the close relationship between these indicators and image quality. First, contrast reflects the brightness difference between different areas in the image. Lower contrast usually means that light scattering causes the image to be blurry. Second, clarity is an important indicator for measuring the ability to distinguish image details. Light scattering will cause the edges of the image to be blurred, thereby reducing clarity. Finally, color histogram differences can reveal changes in the color distribution of the image. Light scattering often leads to color shift and reduced saturation. By weighting and summing these indicators, a comprehensive light scattering index can be obtained. The weight of each indicator in the overall calculation can be adjusted according to the specific application scenario and needs to ensure the accuracy and applicability of the evaluation results. It can comprehensively consider the impact of multiple factors on image quality, rather than relying solely on a single indicator. This can more comprehensively reflect the comprehensive impact of light scattering on the image and improve the reliability of the evaluation. Traditional techniques using a single metric may not accurately reflect the complex phenomenon of light scattering. For example, relying solely on contrast may overlook the impact of color variations, while relying solely on clarity may overlook changes in overall brightness. Therefore, a weighted summation approach allows for a more comprehensive and accurate assessment of light scattering, providing a more reliable foundation for subsequent image processing and analysis.

[0034] Determine the lens dirtiness index of each surveillance camera, including: The light refraction change value of each surveillance camera is normalized to obtain the normalized light refraction change value of each surveillance camera, which is recorded as ; Similarly, the normalized light scattering degree and light transmission loss value of each surveillance camera are obtained; they are recorded as ; The lens dirtiness index of each surveillance camera is calculated from this .

[0035] Light refraction, scattering, and transmission are all important factors affecting image quality. Together, they reflect how light changes as it passes through a lens, and a dirty lens can cause anomalies in these values. By multiplying these three factors together, we can comprehensively assess the overall condition of a lens. Taking the square root helps scale and make the results more comparable.

[0036] Based on the lens dirtiness index of each surveillance camera, a corresponding cleaning warning signal is generated, and maintenance personnel are notified through the smart community management platform to clean the lens.

[0037] The lens dirtiness index of each surveillance camera is compared with the lens dirtiness index interval corresponding to each level of the set cleaning warning signal. If the lens dirtiness index of each surveillance camera is within the lens dirtiness index interval corresponding to the corresponding level of the cleaning warning signal, a cleaning warning signal of the corresponding level will be issued to each surveillance camera. The cleaning warning signals are divided into level one, level two and level three, among which level three is greater than level two and level one.

[0038] Example 2 See also Figure 2 As shown, a smart community public safety monitoring system includes a camera image sequence acquisition module, an abnormal frame image screening module, a lens dirt index judgment module, and a cleaning warning signal generation module. The above modules are connected by wired and / or wireless connections to achieve data transmission between the modules; Camera image sequence acquisition module: obtains the total number P of surveillance cameras in the community, sends an image acquisition command, and each surveillance camera continuously acquires image frames to form an image sequence of each surveillance camera; Abnormal frame image screening module: performs image processing on each frame in the image sequence of each surveillance camera, calculates the clarity evaluation index and sharpness evaluation index of each frame image corresponding to each surveillance camera, and thus screens out the abnormal frame images of each surveillance camera; Lens dirt index judgment module: This module performs in-depth analysis on each abnormal frame image marked as belonging to each surveillance camera, extracts the light parameters of each surveillance camera, and then comprehensively judges the lens dirt index of each surveillance camera; Cleaning warning signal generation module: Generates corresponding cleaning warning signals based on the lens dirtiness index of each surveillance camera, and notifies maintenance personnel to clean the lens through the smart community management platform.

[0039] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0040] It should be understood that determining B based on A does not mean determining B based solely on A. B can also be determined based on A and / or other information.

[0041] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0042] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A smart community public safety monitoring method, characterized in that: include: Obtain the total number P of surveillance cameras within the community, send an image capture command, and have each surveillance camera continuously capture image frames to form an image sequence for each surveillance camera; Perform image processing on each frame in the image sequence of each surveillance camera and calculate the clarity evaluation index of each frame of the image corresponding to each surveillance camera and sharpness evaluation index , thereby filtering out each abnormal frame image of each surveillance camera; Perform in-depth analysis on each abnormal frame image marked as each surveillance camera, extract the light parameters of each surveillance camera, and then comprehensively judge the lens dirtiness index of each surveillance camera ; Based on the lens dirtiness index of each surveillance camera, a corresponding cleaning warning signal is generated, and maintenance personnel are notified through the smart community management platform to clean the lens.

2. A smart community public safety monitoring method according to claim 1, characterized in that: Calculate the clarity evaluation index and sharpness evaluation index of each surveillance camera corresponding to each frame of image. The specific calculation process is as follows: Convert each frame image corresponding to each surveillance camera into a grayscale image, and amplify it until the frame is displayed, and obtain the total number of row frame frames and column frame frames corresponding to each frame image of each surveillance camera; The pixel brightness value of each surveillance camera corresponding to each frame image at the coordinate (x, y) is calculated , x is the number of each row frame, x=1,2,...M, y is the number of each column frame, y=1,2,...N, M and N represent the total number of row frames and column frames respectively; is the grayscale image pixel value of the k-th frame image corresponding to the i-th surveillance camera at the coordinate (x, y), i is the number of the surveillance camera, i=1,2,...P,k is the number of each frame image; The clarity evaluation index of each frame image corresponding to each surveillance camera is calculated from this ； Use the Sobel operator to calculate the gradient amplitude of each frame image corresponding to each surveillance camera at the coordinate (x, y) ,in They represent the horizontal gradient and vertical gradient of the k-th frame image corresponding to the i-th surveillance camera at the coordinate (x, y); Then calculate the sharpness evaluation index of each frame image corresponding to each surveillance camera .

3. A smart community public safety monitoring method according to claim 2, characterized in that: The filtering logic for filtering out abnormal frame images of each surveillance camera is as follows: Set the clarity evaluation index threshold and sharpness evaluation index threshold, respectively, and record them as QX and RD; If the clarity evaluation index of a surveillance camera corresponding to a frame of image If the value is less than the set clarity evaluation index threshold QX, the frame image is judged as a frame image with abnormal clarity; In this screening method, the abnormal frame images of each surveillance camera with each definition are obtained; If the sharpness evaluation index of a surveillance camera corresponding to a frame of image If the value is less than the set sharpness evaluation index threshold RD, the frame image is determined to be a sharpness abnormal frame image; In this screening method, each sharpness abnormal frame image of each surveillance camera is obtained; The abnormal definition frame images and the abnormal sharpness frame images of each surveillance camera are integrated to obtain the abnormal frame images of each surveillance camera.

4. A smart community public safety monitoring method according to claim 1, characterized in that: The light parameters of each surveillance camera include light refraction change value, light scattering degree and light transmission loss value.

5. A smart community public safety monitoring method according to claim 4, characterized in that: The specific extraction logic for extracting the light refraction change value of each surveillance camera is as follows: Arrange the abnormal frame images corresponding to each surveillance camera in sequence according to the shooting timestamp, and match the feature points of all abnormal frame images corresponding to each surveillance camera, thereby calculating the average displacement value of the feature points corresponding to each surveillance camera in all abnormal frame images: ; is the average displacement value of the feature point corresponding to the i-th surveillance camera in all abnormal frame images, are the position coordinates of the feature point corresponding to the i-th surveillance camera in the g+1-th abnormal frame image and the g-th abnormal frame image, respectively. g is the number of each abnormal frame image, g=1,2,...Q, Q is the total number of abnormal frame images; The average displacement value of the feature points corresponding to each surveillance camera in all abnormal frame images is converted into the light refraction change value, thereby obtaining the light refraction change value of each surveillance camera , σ is the set proportional coefficient.

6. A smart community public safety monitoring method according to claim 4, characterized in that: The specific extraction logic for extracting the light transmission loss value of each surveillance camera is as follows: Obtain the cleaning date corresponding to the last cleaning of each surveillance camera; and simultaneously obtain the shooting timestamp of each abnormal frame image corresponding to each surveillance camera, and extract from the database the shooting image corresponding to the cleaning date of the last cleaning of each surveillance camera corresponding to the same timestamp, and simply record it as the reference image of each abnormal frame image corresponding to each surveillance camera; Convert each abnormal frame image and the corresponding reference image of each surveillance camera into a grayscale image, and calculate the average brightness of each abnormal frame image and the corresponding reference image of each surveillance camera. Subtract the average brightness of the corresponding reference image from the average brightness of each abnormal frame image of each surveillance camera to obtain the brightness difference value between each abnormal frame image and the corresponding reference image of each surveillance camera; The calcHist function is used to calculate the color histogram of each abnormal frame image and the corresponding reference image of each surveillance camera, thereby calculating the color histogram difference between each abnormal frame image and the corresponding reference image of each surveillance camera. , where j is the number of each color channel in the histogram, j=1,2,...N, N is the total number of color channels, They represent the histogram values of the g-th abnormal frame image and the corresponding reference image on color channel j of the i-th surveillance camera, respectively. They represent the average values of the histograms of the g-th abnormal frame image and the corresponding reference image of the i-th surveillance camera, respectively, and are used for normalization processing; Calculate the light transmission loss value of each surveillance camera , g is the number of each abnormal frame image, g=1,2,...Q, Q is the total number of abnormal frame images, α1 and α2 represent the set weight coefficients respectively; is the brightness difference between each abnormal frame image and the corresponding reference image of each surveillance camera.

7. A smart community public safety monitoring method according to claim 6, characterized in that: The specific extraction logic for extracting the light scattering degree of each surveillance camera is as follows: Divide each abnormal frame image corresponding to each surveillance camera into F image sub-blocks; thereby collecting the maximum pixel value and the minimum pixel value of each image sub-block in each abnormal frame image corresponding to each surveillance camera; Use characters in sequence and Instead; use the contrast calculation formula , calculate the contrast of each abnormal frame image corresponding to each surveillance camera , f is the number of each image sub-block, f=1,2,...F; Use Laplace transform to calculate the clarity of each abnormal frame image corresponding to each surveillance camera ; Then estimate the light scattering degree of each surveillance camera , ω1, ω2 and ω3 represent predetermined weight factors respectively.

8. A smart community public safety monitoring method according to claim 4, characterized in that: Determine the lens dirtiness index of each surveillance camera, including: The light refraction change value of each surveillance camera is normalized to obtain the normalized light refraction change value of each surveillance camera, which is recorded as ; Similarly, the normalized light scattering degree and light transmission loss value of each surveillance camera are obtained; they are recorded as ; The lens dirtiness index of each surveillance camera is calculated from this .

9. A smart community public safety monitoring system, characterized by: The method is based on the smart community public safety monitoring method described in claims 1-8. The system includes a camera image sequence acquisition module, an abnormal frame image screening module, a lens dirt index judgment module, and a cleaning warning signal generation module. The above modules are connected by wired and / or wireless connections to achieve data transmission between the modules. Camera image sequence acquisition module: obtains the total number P of surveillance cameras in the community, sends an image acquisition command, and each surveillance camera continuously acquires image frames to form an image sequence of each surveillance camera; Abnormal frame image screening module: performs image processing on each frame in the image sequence of each surveillance camera, calculates the clarity evaluation index and sharpness evaluation index of each frame image corresponding to each surveillance camera, and thus screens out the abnormal frame images of each surveillance camera; Lens dirt index judgment module: This module performs in-depth analysis on each abnormal frame image marked as belonging to each surveillance camera, extracts the light parameters of each surveillance camera, and then comprehensively judges the lens dirt index of each surveillance camera; Cleaning warning signal generation module: Generates corresponding cleaning warning signals based on the lens dirtiness index of each surveillance camera, and notifies maintenance personnel to clean the lens through the smart community management platform.

Images