Visual security and protection monitoring management system for buildings in park

Abstract

The invention relates to the technical field of security and protection monitoring, in particular to a park building visual security and protection monitoring management system which comprises a plurality of monitoring devices and a plurality of display devices, laser emitters with freely adjustable irradiation angles are arranged below the monitoring devices, the park building visual security and protection monitoring management system further comprises a smoke recognition unit, independent analysis is carried out on each monitoring device, and the smoke recognition unit is connected with the monitoring devices. Monitoring images are collected regularly, a plurality of continuously collected images are compared and analyzed, and smoke generation, a smoke area and a smoke diffusion direction are identified; the smoke analysis unit is used for analyzing and calculating a smoke concentration index based on the brightness of the visible light path, analyzing a smoke thickness index based on the space coordinate of each end point of the visible light path, and calculating a smoke diffusion index based on the smoke concentration index and the smoke thickness index; and the area evacuation unit is provided with a plurality of escape channel maps in advance, and the display colors and the color depths of different areas in the escape channel maps are adjusted based on the smoke diffusion indexes.

Description

Technical Field

[0001] The present invention relates to the technical field of security monitoring, and particularly relates to a visual security monitoring and management system for park buildings. Background Art

[0002] Industrial parks are usually important areas that gather high-tech enterprises and production facilities, with a large number of production materials and technical assets. Security is of utmost importance. Effective security measures can prevent security incidents such as theft, damage, and fires, protecting the property safety of the park and the lives of employees. By setting up an intelligent security system, real-time monitoring, rapid response, and precise management can be achieved, improving the security prevention ability of the park.

[0003] When the existing security management system responds to a sudden fire, the main preventive measure is to identify a fire through a smoke sensor, and then send an alarm signal through a smoke alarm device to evacuate the crowd. However, this method cannot identify the direction of smoke diffusion and guide the crowd to evacuate, resulting in the crowd being easily lost in the smoke range and getting lost, which has certain defects. Summary of the Invention

[0004] Aiming at the above-mentioned shortcomings of the existing technology, the present invention provides a visual security monitoring and management system for park buildings, which can effectively solve the problem in the existing technology that it is difficult to identify and analyze fire smoke and guide evacuation.

[0005] To achieve the above objectives, the present invention is realized through the following technical solutions: The present invention provides a visual security monitoring and management system for park buildings, including a plurality of monitoring devices and a plurality of display devices. A laser emitter with a freely adjustable irradiation angle is provided below the monitoring device, and further includes: A smoke recognition unit that independently analyzes each monitoring device, regularly collects monitoring images, and compares and analyzes multiple continuously collected images to identify the generation of smoke and generate a smoke alarm command; While generating the smoke alarm command, identify and analyze the smoke area and the smoke diffusion direction; A smoke analysis unit that presets a plurality of fixed angles, constructs a mapping relationship f based on the laser paths when irradiating at different fixed angles. The mapping relationship f is composed of multiple groups of corresponding two-dimensional coordinates and three-dimensional coordinates. Any group of two-dimensional coordinates and three-dimensional coordinates in the mapping relationship f respectively correspond to a point in the monitoring image and a point in the actual space position; When a smoke alarm instruction is obtained, the laser irradiation angle is adjusted multiple times, and the visible light path formed by the laser emitter in the monitored image after each adjustment is identified. Based on the brightness analysis of the visible light path, the smoke concentration index is calculated. Based on the mapping relationship f, the spatial coordinates of each endpoint of the visible light path are obtained. Based on the spatial coordinates of each endpoint, the visible light path is divided into a complete light path or an incomplete light path, and the smoke thickness index is further analyzed: When the visible light path is an incomplete light path, the smoke thickness index is assigned a preset value; When the visible light path is a complete light path, the smoke thickness index is calculated based on the total number of endpoints of the visible light path and the spatial coordinates of each endpoint; Based on the smoke concentration index and the smoke thickness index, the smoke diffusion index is calculated; The area evacuation unit is preset with multiple escape route maps. Based on the positions of different monitoring devices and the corresponding smoke diffusion indices, the display colors and color depths of different areas in the escape route maps are adjusted.

[0006] Furthermore, the process of generating the smoke alarm instruction is as follows: Each monitoring device is analyzed independently, and the analyzed monitoring device is denoted as the target device. The monitored image collected by the target device is regularly obtained and denoted as the target image. The two most recently collected target images are compared, and the similarity between them is calculated. When the similarity is less than or equal to the preset similarity threshold, the current moment is denoted as the starting moment. A preset analysis period is set, and the collected target images are continuously analyzed within the analysis period after the starting moment; The target images collected within the analysis period are sequentially denoted as comparison images in chronological order , where r is the serial number of the comparison image, r = 1, 2,..., h, and h represents the total number of comparison images. For all target images They are grayscaled, and the grayscale value corresponding to each pixel point is obtained; The grayscale values of the pixel points at the same position on any two comparison images 、 are obtained and compared. When the two grayscale values are not equal, the pixel point is denoted as a different pixel point, and all different pixel points are marked on the comparison image and the number is counted and denoted as the different area , g = r + 1; For the different area and the image serial number g, a linear regression model is constructed by linear fitting, and the determination coefficient of the linear regression model is calculated . When the determination coefficient is greater than or equal to the preset test threshold, a smoke alarm instruction is generated.

[0007] Furthermore, the process of identifying the smoke area and the smoke diffusion direction is as follows: Recognition based on contour recognition technology The coverage contour of all difference pixels in is recorded as the smoke area; After generating the smoke alarm command, obtain the comparison image corresponding to the maximum difference area and record it as the first anchor image , the previous comparison image of the first anchor image is recorded as the second anchor image , respectively obtain the center points of the difference pixels in the first anchor point image and the second anchor point image as the first center point and the second center point, respectively, and construct a direction vector pointing from the second center point to the first center point, which is recorded as the smoke diffusion direction; When the second anchor image is When , get the first center point position and Bid out, calculate The distance between the midpoint of each edge line and the first center point, select the midpoint with the smallest distance as the second center point.

[0008] Furthermore, the process of obtaining the mapping relationship f is as follows: There are multiple fixed angles preset, each fixed angle corresponds to a laser indication point, and multiple laser indication points are evenly distributed in the monitoring image; A monitoring image is obtained and recorded as an analysis image, each laser indication point is marked in the analysis image, and a laser starting point is set outside the analysis image, the laser starting point corresponds to the laser emitting end of the laser emitter, and a plurality of line segments corresponding to the laser path are obtained by connecting the laser starting point and each laser indication point and recorded as a connecting line segment, and a portion of the connecting line segment in the analysis image is recorded as a path reference line; Based on the analysis image and the actual spatial position, a two-dimensional coordinate system and a three-dimensional coordinate system are constructed respectively to obtain the two-dimensional coordinates of any point on the path baseline. And the three-dimensional coordinates corresponding to the point , construct the mapping relationship f between the two-dimensional coordinates and the three-dimensional coordinates of the same point on the path baseline.

[0009] Furthermore, the visible light path differentiation process is as follows: Mark the line segment corresponding to the visible light path on the path baseline and record it as the light path segment. Obtain the plane coordinates of each endpoint of the light path segment. Based on the mapping relationship f, obtain the spatial coordinates of each endpoint and record them as the endpoint coordinates. , where i is the serial number of the endpoint coordinates, the serial number is proportional to the distance from the endpoint to the laser transmitter, i=1,2,…,j, j represents the total number of endpoint coordinates, and the endpoint coordinate set is constructed; When the endpoint coordinate set includes the spatial coordinates of the laser indicated point, the visible light path is recorded as a complete light path. When the endpoint coordinate set does not include the spatial coordinates of the laser indicated point, the visible light path is recorded as an incomplete light path.

[0010] Furthermore, the calculation process of the smoke thickness index is as follows: Obtain the coordinates of each end point The corresponding spatial coordinates are denoted as , when the visible light path is a complete light path and j is odd, through the formula Calculate the smoke thickness index, where: represents the smoke thickness index; represents the length of each light path segment; p = 1, 2,..., (j - 1) / 2; is a preset weight coefficient; When the visible light path is a complete light path and j is even, through the formula Calculate the smoke thickness index, where: p = 1, 2,..., j / 2.

[0011] Furthermore, the calculation process of the smoke diffusion index is as follows: Denote the smoke concentration index and the smoke thickness index obtained by corresponding analysis for each irradiation as , where n is the adjusted serial number, substitute into the formula for calculation to obtain the smoke diffusion index , where is a preset normalization coefficient, and m represents the number of irradiations after obtaining the smoke alarm instruction.

[0012] Furthermore, the adjustment process of the area display in the escape route map is as follows: Divide the escape route map into multiple independent display areas, and each independent display area corresponds to a monitoring device; Denote the independent display area corresponding to the monitoring device that generates the smoke alarm instruction as the smoke - affected area, obtain the smoke diffusion index corresponding to the smoke - affected area. When the smoke diffusion index is greater than the preset threshold, display the smoke - affected area as black in the escape route map. When the smoke diffusion index is less than or equal to the preset threshold, multiply the smoke diffusion index by the preset display coefficient to obtain the area brightness value, display the smoke - affected area as red in the escape route map, and adjust the red channel value of each pixel in the smoke - affected area to the area brightness value; Display other independent display areas as green in the escape route map.

[0013] A computer device, including a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, it implements the system according to any one of claims 1 to 8.

[0014] A computer-readable storage medium has a computer program stored thereon, and when the computer program is executed by a processor, the system described in any one of claims 1 to 8 is implemented.

[0015] The technical solution provided by the present invention has the following beneficial effects compared with the known prior art: 1. The present invention utilizes the change characteristics of the smoke coverage area and determines the generation of smoke through linear fitting, thereby realizing smoke recognition based on image recognition, improving the application effect of image recognition technology in the security field, and at the same time determining the smoke coverage range based on the differential area caused by the distribution of smoke particles in the image. Then, the smoke diffusion direction is obtained based on the vector change of the center point of the differential area, which is beneficial to obtaining more smoke-related data through surveillance images.

[0016] 2. The present invention analyzes the formed light path by using the Tyndall effect formed by laser irradiating smoke. It can not only analyze the concentration of smoke based on the light intensity of the visible light path from the side, but also analyze the thickness and coverage of smoke by analyzing the spatial positions of multiple endpoints corresponding to the visible light path, thereby obtaining a smoke thickness index. The larger the smoke thickness index, the larger the volume of smoke filling the space on the side. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts.

[0018] Figure 1 It is the overall module block diagram of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0020] The following further describes the present invention with reference to the embodiments.

[0021] Refer to Figure 1 , a visual security monitoring and management system for campus buildings, at least including: Multiple monitoring devices set at different locations, used to obtain monitoring images of multiple areas within the park buildings, further including: A smoke recognition unit, which analyzes and recognizes the generation of smoke in the monitoring images and generates a smoke alarm instruction, where: Each monitoring device is analyzed independently, and the analyzed monitoring device is denoted as the target device. The monitoring images collected by the target device are regularly obtained and denoted as target images. In a specific embodiment, the acquisition interval of the target images is 1 second. The two most recently acquired target images are compared and the similarity between them is calculated. When the similarity is less than or equal to the preset similarity threshold, the current moment is denoted as the starting moment; There is a preset analysis period, and the acquired target images are continuously analyzed within the analysis period after the starting moment. The specific analysis process is as follows: The target images acquired within the analysis period are sequentially denoted as comparison images in chronological order , where r is the serial number of the comparison image ( indicating the first comparison image acquired within the analysis period), r = 1, 2,..., h, and h represents the total number of comparison images. All target images are grayscaled and the grayscale value corresponding to each pixel point is obtained; The grayscale values of the pixel points at the same position on any two comparison images 、 are obtained and compared. When the two grayscale values are not equal, the pixel point is denoted as a differential pixel point. All differential pixel points are marked on the comparison image and the quantity is counted and denoted as the differential area , g = r + 1; It should be noted that generally, the continuously acquired comparison images will maintain a high similarity, and there will be no grayscale value differences. When smoke spreads in the monitoring images, due to the coverage of the smoke, the brightness of the pixel points in the covered area will change, resulting in grayscale value differences and the appearance of differential pixel points. In this case, the size of the differential area reflects the size of the smoke coverage area on the side.

[0022] For the differential area and the image serial number g, a linear regression model is constructed by linear fitting and the determination coefficient of this linear regression model is calculated . When the determination coefficient is greater than or equal to the preset test threshold, a smoke alarm instruction is generated, and the coverage contour of all differential pixel points in is recognized based on the contour recognition technology and denoted as the smoke area; It should be noted that the diffusion of smoke will significantly cause a linear increase in the smoke coverage area in the surveillance image, and this obvious linear change in the image is difficult to be caused by other situations. For example, when a person or an object passes by, the change in the size of the different area will stop after reaching a certain value (i.e., the maximum value of the human contour or object contour), and it cannot form a similar linear change feature. Through linear fitting, the generation of smoke can be determined by using the change feature of the smoke coverage area, so as to realize smoke recognition based on image recognition and improve the application effect of image recognition technology in the security field.

[0023] After generating the smoke alarm instruction, obtain the comparison image corresponding to the maximum difference area and record it as the first anchor image , record the comparison image before the first anchor point image as the second anchor point image , respectively obtain the center points of the different pixel points in the first anchor point image and the second anchor point image (i.e., the pixel point with the smallest sum of distances to other different pixel points), and record them as the first center point and the second center point respectively, and construct a direction vector pointing from the second center point to the first center point, and record it as the smoke diffusion direction; When the second anchor point image is , obtain the position of the first center point and mark it in , calculate the distances between the midpoints of each side and the first center point, and select the midpoint with the smallest distance as the second center point.

[0024] It should be noted that by determining the first anchor point image, the surveillance image collected when the smoke coverage changes rapidly can be located. Then, according to the construction of the direction vector corresponding to the first center point and the second center point of the second anchor point image and the first anchor point image, the diffusion direction of the smoke can be recognized. Because the area change of the first anchor point image is larger than that of the second anchor point image (i.e., the difference area is the maximum value), it indicates that the smoke diffuses to a higher degree during the shooting of these two images. Therefore, the smoke center point in the first anchor point image will be displaced along the diffusion direction compared with that in the second anchor point image.

[0025] The smoke analysis unit is used to analyze the concentration and thickness of the smoke in the surveillance image, where: A laser emitter is arranged below the monitoring device. The laser emitter emits visible light laser, and the laser emitter is connected to an angle adjustment mechanism for adjusting the laser irradiation angle. The angle adjustment mechanism adopts electric control adjustment, and accurately adjusts the laser irradiation angle of the laser emitter according to the input of different electric signals (i.e., the laser irradiation at a specified point can be adjusted by inputting a signal); It should be noted that adjusting the laser irradiation at a specified position through the angle adjustment mechanism with electric control is an existing technology, and will not be elaborated here too much.

[0026] Independently analyze each monitoring device and the laser emitter below it: There are multiple preset fixed angles (i.e., the irradiation angles of several fixed laser emitters), and each fixed angle corresponds to a laser indication point (i.e., the position of the light spot formed by the laser emitted by the laser emitter on the object surface). The multiple laser indication points are evenly distributed in the monitoring image, so that the laser path (invisible under normal conditions) is as evenly distributed as possible within the display space of the monitoring image; Obtain the monitoring image and record it as the analysis image (the analysis image is a two-dimensional plan). Mark each laser indication point in the analysis image, and set a laser starting point outside the analysis image. The laser starting point corresponds to the laser emission end of the laser emitter. Connect the laser starting point with each laser indication point to obtain multiple line segments corresponding to the laser paths, which are recorded as connection line segments. The part of the connection line segment in the analysis image is recorded as the path reference line, and each path reference line corresponds to a laser indication point (and a fixed angle); Based on the analysis image and the actual space position, construct a two-dimensional coordinate system and a three-dimensional coordinate system respectively, and obtain the two-dimensional coordinates of any point on the path reference line and the corresponding three-dimensional coordinates of this point , construct the mapping relationship f between the two-dimensional coordinates and the three-dimensional coordinates corresponding to the same point on the path reference line (that is, through the mapping relationship f, the three-dimensional coordinate position in the actual space corresponding to any point on the path reference line in the analysis image can be known. The mapping relationship f depends on the installation position and fixed irradiation angle of the laser emitter and will not change with the change of the picture in the figure); It should be noted that the construction of the mapping relationship f can be calculated by the analytic geometry method in mathematics. For example: by determining the emission point and the end point position coordinates , obtain the laser path equation of any point on the laser path in three dimensions, expressed as , where , traverse different t values on the path, calculate the corresponding two-dimensional projection coordinates, and obtain the mapping relationship f. Through the mapping relationship f, the actual position of the light spot on the laser path in the two-dimensional image can be determined, and then the corresponding actual optical path length and position can be calculated according to some light path segments in the two-dimensional image.

[0027] When a smoke alarm instruction is obtained, obtain the smoke area and the diffusion direction in the current monitoring image, adjust the irradiation angle of the laser emitter, and make it irradiate the multiple laser indication points within the coverage range of the smoke area along the intrusion direction in sequence. Conduct path analysis on the visible optical path formed by each irradiation, where: Based on image recognition technology, identify the visible light path formed by the laser emitter in the surveillance image (the visible light path appears as one or more discontinuous straight lines extending along the path reference line in the surveillance image), obtain the light intensities at multiple points on the visible light path except the laser indication points (light intensity, which can be calculated based on the laser display brightness of each point on the visible light path in the surveillance image), and record the maximum value as the light intensity index. , substitute it into the formula for calculation to obtain the smoke concentration index , where is the laser light intensity emitted by the laser emitter, and k is a preset constant coefficient (taking the value of 1 in a specific embodiment); It should be noted that the concentration of smoke is usually directly related to the density and distribution of smoke particles, and the light scattering intensity is proportional to the concentration of particulate matter in the smoke. Therefore, the concentration of smoke can be analyzed indirectly by analyzing the light intensity of the visible light path.

[0028] Mark the line segment corresponding to the visible light path on the path reference line as the light path segment (there may be one or more), obtain the planar coordinates (corresponding to two-dimensional coordinates) of each endpoint of the light path segment, and obtain the spatial coordinates (corresponding to three-dimensional coordinates) of each endpoint based on the mapping relationship f, which are recorded as endpoint coordinates in sequence , where i is the serial number of the endpoint coordinates, and the size of the serial number is proportional to the distance from the endpoint to the emitting end of the laser emitter. i = 1, 2,..., j, and j represents the total number of endpoint coordinates, and an endpoint coordinate set is constructed; When the spatial coordinates of the laser indication point are included in the endpoint coordinate set, this visible light path is recorded as a complete light path (that is, the laser emitted by the laser emitter can completely penetrate the smoke and irradiate the laser indication point, and this visible light path is clearly visible in the surveillance image). Analyze and calculate the smoke thickness index based on the endpoint coordinates in the endpoint coordinate set. When the spatial coordinates of the laser indication point are not included in the endpoint coordinate set, this visible light path is recorded as an incomplete light path, and the smoke thickness index is assigned a value of , which is a preset value; It should be noted that when there is no endpoint corresponding to the laser indication point in the endpoint coordinate set, it means that the light spot (light patch) formed by the laser irradiation at the laser indication point is not visible in the surveillance image. The occurrence of this situation may be due to the fact that the smoke is too thick for the laser to penetrate, or the smoke covers the image acquisition lens of the surveillance device. In either case, it means that the smoke thickness is large, the diffusion degree is serious, and the risk is higher.

[0029] Furthermore, the calculation process of the smoke thickness index is as follows: Obtain the spatial coordinates corresponding to each endpoint coordinate and record them as . When the visible light path is a complete light path and j is odd, through the formula Calculate the smoke thickness index where: Indicates the smoke thickness index; Indicates the length of each optical path; p = 1, 2, ..., (j-1) / 2; is the preset weight coefficient; When the visible light path is a complete light path and j is an even number, the formula Calculate the smoke thickness index where: p=1,2,…,j / 2.

[0030] It should be noted that when the light spot (light spot) formed by laser irradiation on the laser indication point is visible in the monitoring image, it means that the laser can penetrate the smoke. When the laser penetrates the smoke, it will produce an obvious and visible light path due to the Tyndall effect (laser irradiation of suspended particles in the smoke to form a scattering phenomenon). By analyzing the length of the visible light path, the thickness and coverage of the smoke can be analyzed. The larger the smoke thickness index, the larger the volume of smoke filling the space.

[0031] The laser indication points irradiated in sequence are recorded in order as , n is the adjustment number, and the smoke concentration index and smoke thickness index obtained by each irradiation analysis are recorded as , , substitute into the formula The smoke index is calculated in ,in is a preset normalization coefficient, m represents the number of irradiation after obtaining the smoke alarm instruction, and when the smoke diffusion index is greater than the preset threshold, the monitoring area corresponding to the monitoring device is marked as a dangerous area and a regional danger signal is generated.

[0032] The regional evacuation unit has multiple preset escape route maps. Based on the locations of different monitoring devices and the corresponding smoke diffusion index, the display colors and color depths of different areas in the escape route map are adjusted to provide evacuation instructions for evacuees.

[0033] Display devices are installed at the safety exits of each floor in the park building. The display devices display the escape route map of the corresponding floor. The escape route map is divided into multiple independent display areas, and each independent display area corresponds to a monitoring device (when a monitoring device is in an independent display area, it is recorded as the monitoring device corresponding to the independent display area); The independent display area corresponding to the monitoring device that generates the smoke alarm instruction is denoted as the smoke-affected area. Obtain the smoke diffusion index corresponding to the smoke-affected area. When the smoke diffusion index is greater than the preset threshold, display the smoke-affected area as black in the escape route map. When the smoke diffusion index is less than or equal to the preset threshold, multiply the smoke diffusion index by the preset display coefficient to obtain the area brightness value, display the smoke-affected area as red in the escape route map, and adjust the red channel value of each pixel in the smoke-affected area to the area brightness value; Display other independent display areas as green in the escape route map.

[0034] It should be noted that through the display device set at the safety exit, it can provide instructions for the evacuating personnel during the evacuation process. Based on the smoke diffusion index, part of the area in the escape route map is displayed as red with different depths, which can provide a visual path prompt for the escape personnel, so as to avoid the escape personnel straying into the area with serious smoke diffusion and getting lost in direction, and help improve the escape efficiency after a fire occurs.

[0035] A computer device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the above system is implemented.

[0036] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above system is implemented.

[0037] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements will not make the essence of the corresponding technical solutions deviate from the protection scope of the technical solutions of the embodiments of the present invention.

Claims

1. A visual security monitoring and management system for park buildings, including a plurality of monitoring devices and a plurality of display devices. A laser emitter with a freely adjustable irradiation angle is provided below the monitoring device, and it is characterized in that, It also includes: A smoke recognition unit that independently analyzes each monitoring device, regularly collects monitoring images, compares and analyzes multiple continuously collected images, recognizes the generation of smoke, and generates a smoke alarm instruction; While generating the smoke alarm instruction, it recognizes and analyzes the smoke area and the smoke diffusion direction; A smoke analysis unit that presets multiple fixed angles, constructs a mapping relationship f based on the laser paths when irradiated at different fixed angles. The mapping relationship f is composed of multiple groups of corresponding two-dimensional coordinates and three-dimensional coordinates. Any group of two-dimensional coordinates and three-dimensional coordinates in the mapping relationship f respectively correspond to a point in the monitoring image and a point in the actual space position; When a smoke alarm instruction is obtained, the laser irradiation angle is adjusted multiple times, and the visible light path formed by the laser emitter in the monitoring image after each adjustment is recognized. The smoke concentration index is calculated based on the brightness analysis of the visible light path; Based on the mapping relationship f, the spatial coordinates of each endpoint of the visible light path are analyzed. Based on the spatial coordinates of each endpoint, the visible light path is divided into a complete light path or an incomplete light path, and the smoke thickness index is further analyzed: When the visible light path is an incomplete light path, the smoke thickness index is assigned a preset value; When the visible light path is a complete light path, the smoke thickness index is calculated based on the total number of endpoints of the visible light path and the spatial coordinates of each endpoint; The smoke diffusion index is calculated based on the smoke concentration index and the smoke thickness index; A regional evacuation unit that presets multiple escape route maps, and adjusts the display color and color depth of different regions in the escape route map based on the positions of different monitoring devices and the corresponding smoke diffusion indices.

2. The visual security monitoring and management system for park buildings according to claim 1, characterized in that the smoke The specific process of generating the alarm instruction is as follows: Independently analyze each monitoring device, and record the analyzed monitoring device as the target device. Regularly obtain the monitoring images collected by the target device and record them as target images. Compare the two most recently collected target images and calculate the similarity between them. When the similarity is less than or equal to the preset similarity threshold, record the current moment as the starting moment. There is a preset analysis period, and the collected target images are continuously analyzed within the analysis period after the starting moment; The target images collected during the analysis period are sequentially recorded as comparison images in chronological order , where r is the serial number of the comparison image, r = 1, 2, …, h, and h represents the total number of comparison images. For all target images perform grayscale processing and obtain the grayscale value corresponding to each pixel point; Obtain any two comparison images , The gray values of the pixel points at the same position are obtained and compared. When the two gray values are not equal, the pixel point is recorded as a differential pixel point. On the comparison image All differential pixel points are marked and the quantity is counted and recorded as the differential area , g = r + 1; For the differential area Perform a linear fit with the image sequence number g to construct a linear regression model and calculate the coefficient of determination of the linear regression model When the coefficient of determination is greater than or equal to a preset test threshold, generate a smoke alarm instruction.

3. The visual security monitoring and management system for park buildings according to claim 2, characterized in that the smoke The process of identifying the area and the smoke diffusion direction is as follows: Identify the coverage contour of all differential pixel points based on the contour recognition technology and denote it as the smoke area; After generating the smoke alarm instruction, obtain the comparison image corresponding to the maximum difference area and record it as the first anchor image , record the comparison image before the first anchor point image as the second anchor point image , respectively obtain the central points of the different pixel points in the first anchor point image and the second anchor point image, which are recorded as the first central point and the second central point respectively, and construct a direction vector from the second central point to the first central point, which is recorded as the smoke diffusion direction; When the second anchor image is When , get the first center point position and The bid is calculated The distance between the midpoint of each edge line and the first center point, select the midpoint with the smallest distance as the second center point.

4. A visualized security monitoring and management system for park buildings according to claim 1, characterized in that, The process of obtaining the mapping relationship f is as follows: Preset multiple fixed angles, each fixed angle corresponds to a laser indication point position, and multiple laser indication point positions are evenly distributed in the monitoring image; Obtain the monitoring image and record it as the analysis image. Mark each laser indication point position in the analysis image, and set a laser starting point outside the analysis image. The laser starting point corresponds to the laser emission end of the laser emitter. Connect the laser starting point with each laser indication point position to obtain multiple line segments corresponding to the laser paths and record them as connection line segments. Record the part of the connection line segment in the analysis image as the path reference line; Construct a two-dimensional coordinate system and a three-dimensional coordinate system based on the analyzed image and the actual spatial position respectively, and obtain the two-dimensional coordinates of any point on the path baseline. And the three-dimensional coordinates corresponding to this point , and construct the mapping relationship f between the two-dimensional coordinates and the three-dimensional coordinates corresponding to the same point on the path baseline.

5. The visual security monitoring and management system for park buildings according to claim 1, wherein The process of distinguishing the visible light path is as follows: Mark the line segment corresponding to the visible optical path on the path baseline as the optical path segment, obtain the plane coordinates of each endpoint of the optical path segment, and obtain the spatial coordinates of each endpoint based on the mapping relationship f, which are successively denoted as endpoint coordinates , where i is the serial number of the endpoint coordinates, and the size of the serial number is proportional to the distance from the endpoint to the emission end of the laser emitter. i = 1, 2,..., j, and j represents the total number of endpoint coordinates, and construct an endpoint coordinate set; When the set of endpoint coordinates includes the spatial coordinates of the laser indication point position, record the visible light path as a complete light path. When the set of endpoint coordinates does not include the spatial coordinates of the laser indication point position, record the visible light path as an incomplete light path.

6. The visual security monitoring and management system for park buildings according to claim 5, wherein the smoke The process of calculating the thickness index is as follows: Obtain the coordinates of each endpoint The corresponding space coordinates are denoted as , when the visible optical path is a complete optical path and j is odd, through the formula Calculate the smoke thickness index, where: Indicates the smoke thickness index; Indicates the length of each optical path; p = 1, 2,..., (j - 1) / 2; is a preset weight coefficient; When the visible light path is a complete light path and j is an even number, the smoke thickness index is calculated through the formula where: p = 1, 2,..., j / 2.

7. The visual security monitoring and management system for park buildings according to claim 6, wherein The process of calculating the smoke diffusion index is as follows: The smoke concentration index and the smoke thickness index obtained by corresponding analysis for each irradiation are respectively denoted as , where n is the adjusted serial number, and substitute it into the formula for calculation to obtain the smoke diffusion index , where is the preset normalization coefficient, and m represents the number of irradiations after obtaining the smoke alarm instruction.

8. The visual security monitoring and management system for park buildings according to claim 1, wherein The adjustment process of the area display in the escape route map is as follows: The escape route map is divided into multiple independent display areas, and each independent display area corresponds to a monitoring device; The independent display area corresponding to the monitoring device that generates the smoke alarm instruction is recorded as the smoke-affected area. Obtain the smoke diffusion index corresponding to the smoke-affected area. When the smoke diffusion index is greater than the preset threshold, the smoke-affected area is displayed as black in the escape route map. When the smoke diffusion index is less than or equal to the preset threshold, multiply the smoke diffusion index by the preset display coefficient to obtain the area brightness value, display the smoke-affected area as red in the escape route map, and adjust the red channel value of each pixel in the smoke-affected area to the area brightness value; In the escape route map, display other independent display areas as green.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the system described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the system described in any one of claims 1 to 8.

Images