Fire fighting access occupation identification method and equipment based on grid line topology and medium

Abstract

The invention discloses a fire fighting access occupancy recognition method and device based on grid line topology and a medium, and the method comprises the steps: obtaining a fire fighting access image in a non-occupancy state, and extracting grid lines of the non-occupancy fire fighting access image, so as to determine a reference topological graph corresponding to the grid lines; acquiring a real-time fire fighting access image, and performing illumination recognition optimization on the real-time fire fighting access image to obtain a fire fighting access real-time detection image; extracting real-time grid lines of the fire fighting access real-time detection graph to determine a real-time topological graph corresponding to the real-time grid lines; based on the reference topological graph and the real-time topological graph, through fire fighting access occupancy judgment, obtaining occupancy area data of the fire fighting access; and determining the fire fighting access occupancy state through substantial occupancy verification according to the occupancy area data. According to the method, the technical problems that in the prior art, fire fighting access occupancy identification is limited by the types of occupancy objects and recognition of small interferents frequently alarms are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent video monitoring, in particular to a fire passage occupation recognition method based on a grid line topology, a device and a medium. BACKGROUND

[0002] A fire passage is a life passage in an emergency such as a fire, and its smoothness directly affects the rescue efficiency. In order to enable the fire passage to be in a safe state of smoothness at all times, the grid line region in the fire passage is usually monitored by a camera.

[0003] In the prior art, the monitoring of the fire passage region is mainly through target monitoring, semantic segmentation, and regional intrusion detection methods. Firstly, the target monitoring detection method identifies preset categories such as vehicles, debris, and other occupying objects by training a model, but cannot identify untrained objects. Secondly, the semantic segmentation detection method directly segments the fire passage region and the non-passage region, but is easily affected by light changes, shadows, and small interference objects, has a high false positive rate, and cannot distinguish between slight coverage and substantial occupation. Thirdly, the regional intrusion detection method judges whether an object has entered by setting a virtual fence, but does not consider the inherent structure of the fire passage, and is too sensitive to low objects close to the ground or partially intruding objects, and has poor adaptability. SUMMARY

[0004] The embodiments of the present application provide a fire passage occupation recognition method based on a grid line topology, a device and a medium, which solve the technical problems of the prior art that the fire passage occupation recognition is limited by the type of occupying objects and frequent alarm caused by the identification of small interference objects.

[0005] In a first aspect, the embodiments of the present application provide a fire passage occupation recognition method based on a grid line topology, characterized in that the method comprises: acquiring a fire passage image in a non-occupied state, and extracting a grid line of the non-occupied fire passage image to determine a reference topology graph corresponding to the grid line; acquiring a real-time fire passage image, and performing light recognition optimization on the real-time fire passage image to obtain a real-time detection graph of the fire passage; extracting a real-time grid line of the real-time detection graph of the fire passage to determine a real-time topology graph corresponding to the real-time grid line; based on the reference topology graph and the real-time topology graph, determining an occupied region data of the fire passage through fire passage occupation judgment; and determining an occupation state of the fire passage through substantial occupation verification according to the occupied region data.

[0006] In an implementation manner of the present application, the grid lines of the unoccupied fire passage image are extracted to determine the reference topological graph corresponding to the grid lines, specifically comprising: performing foreground division on the unoccupied fire passage through semantic segmentation to obtain grid region data; determining the grid lines based on the grid region data, and determining the node parameters and edge parameters corresponding to the grid lines through grid feature recognition according to the grid lines; wherein the node parameters comprise: intersection point type, node pixel coordinates; the edge parameters comprise: edge start point, edge key point, edge length; and the node parameters and edge parameters are arranged into a graph structure set to determine the reference topological graph.

[0007] In an implementation manner of the present application, the real-time fire passage image is subjected to illumination recognition optimization to obtain a real-time fire passage detection graph, specifically comprising: performing histogram equalization processing on the real-time fire passage image to determine a real-time fire passage feature enhanced image; performing four-direction edge detection on the real-time fire passage feature enhanced image through a Sobel operator to obtain a gradient amplitude graph of grid line features; and obtaining the real-time fire passage detection graph through grid feature fusion based on the gradient amplitude graph.

[0008] In an implementation manner of the present application, the real-time grid lines of the real-time fire passage detection graph are extracted to determine the real-time topological graph corresponding to the real-time grid lines, specifically comprising: performing foreground division on the real-time fire passage detection graph through semantic segmentation to obtain real-time grid region data; determining the real-time grid lines based on the real-time grid region data, and determining the node parameters and edge parameters corresponding to the real-time grid lines through grid feature recognition according to the real-time grid lines; and arranging the node parameters and edge parameters into a real-time graph structure set to determine the real-time topological graph.

[0009] In an implementation manner of the present application, based on the reference topological graph and the real-time topological graph, the occupancy area data of the fire passage is obtained through fire passage occupancy determination, specifically comprising: calculating the missing nodes of the real-time topological graph relative to the reference topological graph; calculating the missing edges of the real-time topological graph relative to the reference topological graph; performing pixel range positioning on the missing nodes to determine the intersection point missing area of the fire passage grid lines; and performing strip pixel covering on the missing edges to determine the edge missing area of the fire passage grid lines.

[0010] In an implementation manner of the present application, according to the occupancy area data, the fire passage occupancy state is determined through substantial occupancy verification, specifically comprising: based on the occupancy area data, continuously monitoring the fire passage images under a preset frame number to obtain fire passage occupancy frame data; in the case that the position coincidence degree of the missing nodes and edges of each frame of the fire passage occupancy frame data is lower than a preset percentage value, determining that the fire passage occupancy state is temporary occupancy; and in the case that the position coincidence degree of the missing nodes and edges of each frame of the fire passage occupancy frame data is not lower than the preset percentage value, judging the area difference of the missing area to determine the fire passage occupancy state.

[0011] In an implementation form of the application, after determining the fire passage occupancy state by substantial occupancy verification according to the occupancy area data, the method further comprises: triggering an alarm information in a case that the occupancy time of the fire passage occupancy state exceeds a preset configuration time; recording occupancy information corresponding to the fire passage occupancy state synchronously based on the alarm information; wherein the occupancy information comprises: occupancy start time, occupancy position, and occupancy real-time image.

[0012] In an implementation form of the application, after recording the occupancy information corresponding to the fire passage occupancy state synchronously based on the alarm information, the method further comprises: performing motion monitoring of an occupancy object in a detection area corresponding to the occupancy information to obtain three-dimensional feature data of the occupancy object; and directly triggering the alarm information in a case that the occupancy object enters the grid line area of the fire passage again based on the three-dimensional feature data.

[0013] In a second aspect, the embodiments of the application further provide a fire passage occupancy recognition device based on grid line topology, characterized in that the device comprises: at least one processor; and a memory in communication connection with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to: acquire a fire passage image in a non-occupancy state, and extract grid lines of the non-occupancy fire passage image to determine a reference topology graph corresponding to the grid lines; acquire a real-time fire passage image, and perform illumination recognition optimization on the real-time fire passage image to obtain a fire passage real-time detection graph; extract real-time grid lines of the fire passage real-time detection graph to determine a real-time topology graph corresponding to the real-time grid lines; obtain occupancy area data of the fire passage based on the reference topology graph and the real-time topology graph through fire passage occupancy determination; and determine a fire passage occupancy state by substantial occupancy verification according to the occupancy area data.

[0014] In a third aspect, the embodiments of the application further provide a non-volatile computer storage medium for fire passage occupancy recognition based on grid line topology, which stores computer executable instructions, and the computer executable instructions are configured to: acquire a fire passage image in a non-occupancy state, and extract grid lines of the non-occupancy fire passage image to determine a reference topology graph corresponding to the grid lines; acquire a real-time fire passage image, and perform illumination recognition optimization on the real-time fire passage image to obtain a fire passage real-time detection graph; extract real-time grid lines of the fire passage real-time detection graph to determine a real-time topology graph corresponding to the real-time grid lines; obtain occupancy area data of the fire passage based on the reference topology graph and the real-time topology graph through fire passage occupancy determination; and determine a fire passage occupancy state by substantial occupancy verification according to the occupancy area data.

[0015] The embodiment of the present application provides a fire passage occupation recognition method, device and medium based on a grid line topology, through comparison of grid line topology construction of a fire passage image in a non-occupation state and real-time grid line topology construction of a fire passage real-time detection image, through fire passage occupation judgment and substantial occupation verification, the technical problems that fire passage occupation recognition in the prior art is limited by the type of occupying objects and frequent alarm caused by small interference objects are solved, accurate judgment of the occupation state of the fire passage is realized, various occupation conditions can be recognized by presetting a target category, the missing report rate is reduced, the influence of small interference objects is excluded, false positives are reduced, the inherent structure of the grid line of the fire passage is utilized, the stability and reliability of recognition in a complex environment are improved, and false alarms caused by temporary shielding are avoided. BRIEF DESCRIPTION OF DRAWINGS

[0016] The drawings described herein are used to provide further understanding of the present application, constitute a part of the present application, the illustrative embodiments of the present application and the description thereof are used to explain the present application, and do not constitute improper limitations on the present application. In the drawings: Figure 1 A fire passage occupation recognition method based on a grid line topology provided by the embodiment of the present application is shown in a flowchart; Figure 2 An internal structure schematic diagram of a fire passage occupation recognition device based on a grid line topology provided by the embodiment of the present application. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical scheme and advantages of the present application clearer, the technical scheme of the present application will be clearly and completely described below in combination with specific embodiments of the present application and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0018] The embodiment of the present application provides a fire passage occupation recognition method, device and medium based on a grid line topology, through comparison of grid line topology construction of a fire passage image in a non-occupation state and real-time grid line topology construction of a fire passage real-time detection image, through fire passage occupation judgment and substantial occupation verification, the technical problems that fire passage occupation recognition in the prior art is limited by the type of occupying objects and frequent alarm caused by small interference objects are solved, accurate judgment of the occupation state of the fire passage is realized, various occupation conditions can be recognized by presetting a target category, the missing report rate is reduced, the influence of small interference objects is excluded, false positives are reduced, the inherent structure of the grid line of the fire passage is utilized, the stability and reliability of recognition in a complex environment are improved, and false alarms caused by temporary shielding are avoided.

[0019] The technical solutions of the embodiments of the present application are described in detail below with reference to the drawings.

[0020] Figure 1 A grid line topology-based fire passage occupation identification method flowchart is provided for the embodiments of the present application. As shown in Figure 1 The grid line topology-based fire passage occupation identification method provided by the embodiments of the present application specifically includes the following steps: Step 101, acquire a fire passage image in a non-occupied state, and extract grid lines of the non-occupied fire passage image to determine a reference topology graph corresponding to the grid lines.

[0021] Illustratively, the grid lines drawn on the fire passage are key areas that cannot be occupied. In order to solve the technical problem of whether the grid line part is occupied in the prior art, the embodiments of the present application extract the grid lines of the non-occupied fire passage image, construct a corresponding reference topology graph, realize the reference topology of occupation comparison using the inherent structure of the grid lines of the fire passage, and provide a data basis for fire passage occupation monitoring.

[0022] Specifically, the grid lines of the non-occupied fire passage image are extracted to determine the reference topology graph corresponding to the grid lines, specifically including: through semantic segmentation, the non-occupied fire passage is foreground divided to obtain grid region data; based on the grid region data, the grid lines are determined, and according to the grid lines, the node parameters and edge parameters corresponding to the grid lines are determined through grid feature recognition; wherein the node parameters include: intersection point type, node pixel coordinates; the edge parameters include: edge start point, edge key point, edge length; the node parameters and edge parameters are arranged into a graph structure set to determine the reference topology graph.

[0023] In one embodiment, a fixed monitoring camera is installed above the fire passage (such as a vertical rod or a wall surface), the field of view covers the complete passage area, the resolution is ≥200 million pixels, the frame rate is ≥25 fps, and all-weather (including night infrared) imaging is supported.

[0024] After the camera is installed, multiple frames (≥10 frames) of clear images of the non-occupied fire passage are collected, the stable area between frames is taken as a reference picture, and the grid lines in the reference picture are extracted through semantic segmentation. The grid lines are yellow grid separation lines on the ground of the fire passage, and the complete grid contour is obtained after filtering noise.

[0025] Based on the complete grid contour, the node parameters and edge parameters corresponding to the grid line contour are determined; wherein the node parameters include: intersection point type, node pixel coordinates; the edge parameters include: edge start point, edge key point, edge length.

[0026] Finally, the node parameters and edge parameters are sorted into a graph structure set to determine the reference topology graph, and the reference topology graph is systematically stored as the inherent configuration of the camera.

[0027] Step 102, acquiring a real-time fire passage image, and performing illumination recognition optimization on the real-time fire passage image to obtain a fire passage real-time detection image.

[0028] For example, in actual application, in the case of night or poor light conditions, it is difficult to identify the grid line features in some grid line areas. In order to cope with the above situation, the present application performs illumination recognition optimization on the real-time fire passage image to obtain a fire passage real-time detection image, realizes the identification of the fire passage grid line under weak light, and improves the system robustness.

[0029] Specifically, the illumination recognition optimization is performed on the real-time fire passage image to obtain the fire passage real-time detection image, which specifically includes: performing histogram equalization processing on the real-time fire passage image to determine a real-time fire passage feature enhancement image; performing four-direction edge detection on the real-time fire passage feature enhancement image through a Sobel operator to obtain a gradient amplitude graph of the grid line features; and based on the gradient amplitude graph, performing grid feature fusion to obtain the fire passage real-time detection image.

[0030] In one embodiment, the real-time fire passage image is subjected to illumination adaptive preprocessing to obtain a preprocessing image corresponding to the real-time fire passage image and having enhanced details.

[0031] Then, according to the design specification requirements of the existing grid lines, a convolution kernel with directions of 0° (horizontal), 45°, 90° (vertical), and 135° is determined to perform convolution operation, and gradient components in four key directions of 0° (horizontal), 45°, 90° (vertical), and 135° are accurately extracted, respectively. Based on the actual requirement that the fire passage grid lines are composed of square or diamond structures, the main line directions of the orthogonal grid or the rotated diamond grid can be effectively covered.

[0032] Finally, the gradient graphs in the four directions are linearly fused to generate a comprehensive edge intensity graph. The Hough transform can be used to detect line segments, calculate the density and spatial distribution of the grid line intersection points in the local region of the image, obtain a grid line feature heat map representing the probability of the existence of the grid, and perform grid feature fusion to obtain the fire passage real-time detection image.

[0033] Step 103, extracting the real-time grid line of the fire passage real-time detection image to determine the real-time topology graph corresponding to the real-time grid line.

[0034] Specifically, the real-time grid line of the fire passage real-time detection graph is removed to determine the real-time topology graph corresponding to the real-time grid line, specifically comprising: through semantic segmentation, foreground division is performed on the fire passage real-time detection graph to obtain real-time grid area data; based on the real-time grid area data, the real-time grid line is determined, and the node parameters and edge parameters corresponding to the real-time grid line are determined through grid feature recognition according to the real-time grid line; the node parameters and edge parameters are arranged into a real-time graph structure set to determine the real-time topology graph.

[0035] In one embodiment, first, the camera real-time captures the fire passage image, 1 frame every 30 seconds, which can be configured according to actual needs. Image cropping is performed on each frame of image to retain only the preset fire passage area, excluding non-passage backgrounds such as green belts, walls and other interference.

[0036] The grid line in the real-time detection graph is extracted through semantic segmentation, which is a yellow grid separation line on the ground of the fire passage. After filtering noise, the complete grid contour of the real-time detection graph is obtained.

[0037] Based on the complete grid contour, the node parameters and edge parameters corresponding to the grid line contour of the real-time topology graph are determined; wherein the node parameters include: intersection point type, node pixel coordinates; the edge parameters include: edge start point, edge key point, edge length.

[0038] Finally, the node parameters and edge parameters are arranged into a graph structure set to determine the real-time topology graph.

[0039] Step 104, based on the reference topology graph and the real-time topology graph, the occupancy area data of the fire passage is obtained through fire passage occupancy determination.

[0040] For example, due to the technical problem that the prior art fire passage occupancy recognition is limited by the type of the occupying object, the present application adopts a method of constructing topology based on fire passage grid lines to avoid the above problems by directly judging whether the topology graph is missing (corresponding to being blocked), thereby realizing topology recognition of fire passage occupancy determination, without the need to train a model to recognize different blocking objects, improving recognition efficiency and reducing the difficulty of later maintenance.

[0041] Specifically, based on the reference topology graph and the real-time topology graph, the occupancy area data of the fire passage is obtained through fire passage occupancy determination, specifically comprising: calculating the missing nodes of the real-time topology graph relative to the reference topology graph; calculating the missing edges of the real-time topology graph relative to the reference topology graph; pixel range positioning is performed on the missing nodes to determine the intersection point missing area of the fire passage grid line; the missing edge is covered with a strip of pixels to determine the edge missing area of the fire passage grid line.

[0042] In one embodiment, the real-time graph structure (G t) with the reference graph structure (G0), calculate the missing nodes (N_missing = G0nodes - G t edges) and the missing edges (E_missing = G0edges - G t edges).

[0043] The pixel area corresponding to the missing node is centered on the node coordinates, with a range of 5x5 pixels; the pixel area corresponding to the missing edge is the coverage range of the line segment of the corresponding edge (a 3-pixel wide strip area).

[0044] Finally, record the position of the missing area, expressed as a pixel coordinate range, and the missing area, expressed as the number of pixels, to determine the edge missing area of the fire passage grid line.

[0045] Step 105, according to the occupied area data, determine the fire passage occupancy state through substantial occupancy verification.

[0046] For example, the present application improves the stability and reliability of recognition in complex environments through substantial occupancy verification using the inherent structure of the fire passage grid line, and avoids false alarms caused by temporary obstructions.

[0047] Specifically, according to the occupied area data, determine the fire passage occupancy state through substantial occupancy verification, specifically including: based on the occupied area data, continuously monitor the fire passage image under a preset number of frames to obtain fire passage occupancy frame data; in the case where the position coincidence degree of the missing nodes and edges of each frame of the fire passage occupancy frame data is lower than the preset percentage value, determine that the fire passage occupancy state is temporary occupancy; in the case where the position coincidence degree of the missing nodes and edges of each frame of the fire passage occupancy frame data is not lower than the preset percentage value, judge the area difference of the missing area to determine the fire passage occupancy state.

[0048] Specifically, after determining the fire passage occupancy state through substantial occupancy verification according to the occupied area data, the method further comprises: in the case where the occupancy duration of the fire passage occupancy state exceeds the preset configuration time, triggering an alarm information; based on the alarm information, synchronously recording the occupancy information corresponding to the fire passage occupancy state; wherein the occupancy information includes: occupancy start time, occupancy position, occupancy real-time image.

[0049] Further, after synchronously recording the occupancy information corresponding to the fire passage occupancy state based on the alarm information, the method further comprises: detecting the motion of the occupancy object within the detection area to obtain three-dimensional feature data of the occupancy object; based on the three-dimensional feature data, in the case where the occupancy object enters the fire passage grid line area again, directly triggering an alarm information.

[0050] In one embodiment, N frames of images are continuously monitored, N≥5, which can be configured according to actual needs. If the following conditions are met at the same time, it is determined to be persistent occupation, excluding temporary occlusion, such as pedestrians and fallen leaves temporarily covering. The coincidence degree of the missing node / edge position of each frame is ≥80%; The area difference of the missing area is ≤20%.

[0051] When the duration of persistent occupation exceeds the set threshold, the alarm module triggers an alarm, records the start time, position and real-time image of the occupation as evidence, and in the case that the occupied object corresponding to the occupation information moves out of the fire passage grid line area again, it is directly regarded as occupation and the alarm information is triggered.

[0052] The above is the method embodiment of the present application. Based on the same inventive concept, the present application also provides a fire passage occupation identification device based on grid line topology, the structure of which is shown in Figure 2 .

[0053] Figure 2 A fire passage occupation identification device based on grid line topology provided by the embodiment of the present application is shown in Figure 2 . The device includes: at least one processor 201; and a memory 202 in communication connection with the at least one processor; Wherein the memory 202 stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor 201 to enable the at least one processor 201 to: acquire a fire passage image in a non-occupied state, and extract the grid lines of the non-occupied fire passage image to determine the reference topology graph corresponding to the grid lines; acquire a real-time fire passage image, and perform light recognition optimization on the real-time fire passage image to obtain a fire passage real-time detection image; extract the real-time grid lines of the fire passage real-time detection image to determine the real-time topology graph corresponding to the real-time grid lines; based on the reference topology graph and the real-time topology graph, through fire passage occupation determination, obtain the occupation area data of the fire passage; according to the occupation area data, through substantial occupation verification, determine the occupation state of the fire passage.

[0054] Some embodiments of the present application provide a non-volatile computer storage medium for fire passage occupation identification based on grid line topology corresponding to Figure 1 , which stores computer executable instructions, and the computer executable instructions are set to: The image of the unoccupied fire channel is acquired, and the grid line of the unoccupied fire channel image is extracted to determine the reference topology graph corresponding to the grid line. The real-time fire channel image is acquired, and the real-time fire channel image is optimized by light recognition to obtain a real-time detection image of the fire channel. The real-time grid line of the real-time detection image of the fire channel is extracted to determine a real-time topology graph corresponding to the real-time grid line. Based on the reference topology graph and the real-time topology graph, the occupied area data of the fire channel is obtained through fire channel occupation determination. According to the occupied area data, the occupation state of the fire channel is determined through substantial occupation verification.

[0055] Each of the embodiments in the present application is described in a progressive manner, and the same and similar parts between the embodiments can be referred to each other. Each embodiment focuses on the difference from other embodiments. In particular, for the Internet of Things device and medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the part of the method embodiments.

[0056] The system and medium provided by the embodiments of the present application are one-to-one corresponding with the method, and therefore, the system and medium also have similar beneficial technical effects as the method. Since the beneficial technical effects of the method have been described in detail above, the beneficial technical effects of the system and medium will not be described here.

[0057] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program codes.

[0058] The present application is described with reference to flowcharts and / or block diagrams according to the method, device (system), and computer program product of the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to produce a machine, so that the instructions executed by the computer or other programmable data processing devices produce a device for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 The device for implementing the functions specified in one flow or multiple flows and / or blocks. Figure 1 The device for implementing the functions specified in one flow or multiple flows and / or blocks.

[0059] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the Figure 1 function specified in the flow or flows and / or blocks Figure 1 of the block or blocks.

[0060] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the Figure 1 function specified in the flow or flows and / or blocks Figure 1 of the block or blocks.

[0061] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0062] The memory can include non-persistent memory and / or volatile memory, such as random access memory (RAM) and / or cache memory, non-volatile memory, such as read-only memory (ROM), EPROM, and / or flash memory. The memory is an example of computer-readable media.

[0063] Computer-readable media includes permanent and non-permanent, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD), or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible to a computing device. According to the definition herein, computer-readable media does not include transitory media, such as modulated data signals and carrier waves.

[0064] It should also be noted that the terms "comprising", "comprises" or other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can also include other elements not expressly listed or inherent to such process, method, article, or apparatus.

[0065] The above embodiments are only used to illustrate the present application, but not to limit it. Instead of the above, various modifications and changes can be made to the application by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the application shall fall into the scope of the claims of the application.

Claims

1. A method for identifying the occupancy of fire lanes based on grid line topology, characterized in that, The method includes: Acquire images of unoccupied fire lanes and extract grid lines from the unoccupied fire lane images to determine the baseline topology map corresponding to the grid lines; Acquire real-time fire lane images and perform illumination recognition optimization on the real-time fire lane images to obtain a real-time fire lane detection map; Extract the real-time grid lines from the real-time monitoring map of the fire lane to determine the real-time topology map corresponding to the real-time grid lines; Based on the baseline topology map and the real-time topology map, the occupied area data of the fire lane is obtained by determining the occupancy of the fire lane; Based on the occupied area data, the occupancy status of fire lanes is determined through substantive occupancy verification.

2. The method for identifying fire lane occupancy based on grid line topology according to claim 1, characterized in that, Extracting the grid lines from the unobstructed fire lane image to determine the corresponding baseline topology map, specifically including: Semantic segmentation is used to divide the unoccupied fire lanes into foreground areas to obtain grid area data; Based on the grid region data, the grid lines are determined, and based on the grid lines, the node parameters and edge parameters corresponding to the grid lines are determined through grid feature recognition; wherein, the node parameters include: intersection point type, node pixel coordinates; the edge parameters include: edge start point, edge end point, and edge length; The node parameters and edge parameters are organized into a graph structure set to determine the baseline topology graph.

3. The method for identifying fire lane occupancy based on grid line topology according to claim 1, characterized in that, The real-time fire lane image is optimized using illumination recognition to obtain a real-time fire lane detection map, specifically including: Histogram equalization is performed on the real-time fire lane images to determine the real-time fire lane feature-enhanced images; The Sobel operator is used to perform edge detection in four directions on the real-time fire lane feature enhancement image to obtain the gradient magnitude map of the grid line features. Based on the gradient amplitude map, the real-time detection map of the fire lane is obtained through grid feature fusion.

4. The method for identifying fire lane occupancy based on grid line topology according to claim 1, characterized in that, Extracting the real-time grid lines from the real-time fire lane monitoring map to determine the corresponding real-time topology map, specifically including: Semantic segmentation is used to divide the foreground of the real-time fire lane detection map to obtain real-time grid area data; Based on the real-time grid area data, the real-time grid line is determined, and based on the real-time grid line, the node parameters and edge parameters corresponding to the real-time grid line are determined through grid feature identification; The node parameters and edge parameters are organized into a real-time graph structure set to determine the real-time topology graph.

5. The method for identifying fire lane occupancy based on grid line topology according to claim 1, characterized in that, Based on the baseline topology map and the real-time topology map, the occupancy area data of fire lanes is obtained through fire lane occupancy determination, specifically including: Calculate the missing nodes in the real-time topology graph relative to the baseline topology graph; Calculate the missing edges of the real-time topology graph relative to the reference topology graph; The missing nodes are located by pixel range to determine the missing area at the intersection of the fire lane grid lines; The missing edges are covered with strip pixels to determine the missing area of ​​the fire lane grid lines.

6. The method for identifying fire lane occupancy based on grid line topology according to claim 1, characterized in that, Based on the occupied area data, the occupancy status of fire lanes is determined through substantive occupancy verification, specifically including: Based on the occupied area data, fire lane images are continuously monitored for a preset number of frames to obtain fire lane occupancy frame data. If the overlap between the missing nodes and edge positions in each frame of the fire lane occupancy frame data is less than a preset percentage, the fire lane occupancy status is determined to be temporary occupancy. If the overlap between the missing node and the edge position in each frame of the fire lane occupancy frame data is not less than a preset percentage, the area difference of the missing area is determined to determine the occupancy status of the fire lane.

7. The method for identifying fire lane occupancy based on grid line topology according to claim 1, characterized in that, After determining the occupancy status of fire lanes through substantial occupancy verification based on the occupancy area data, the method further includes: An alarm is triggered if the occupancy time of the fire lane exceeds the preset time. Based on the alarm information, the occupancy information corresponding to the occupancy status of the fire lane is recorded synchronously; wherein, the occupancy information includes: occupancy start time, occupancy location, and real-time image of occupancy.

8. The method for identifying fire lane occupancy based on grid line topology according to claim 1, characterized in that, After synchronously recording the occupancy information corresponding to the occupancy status of the fire lane based on the alarm information, the method further includes: Motion monitoring is performed on the occupant object corresponding to the occupancy information within the detection area to obtain the three-dimensional feature data of the occupant object; Based on the aforementioned three-dimensional feature data, an alarm is directly triggered if the object re-enters the fire lane grid area.

9. A fire lane occupancy identification device based on grid line topology, characterized in that, The device includes: At least one processor; And, a memory communicatively connected to the at least one processor; The memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, enable the at least one processor to: Acquire images of unoccupied fire lanes and extract grid lines from the unoccupied fire lane images to determine the baseline topology map corresponding to the grid lines; Acquire real-time fire lane images and perform illumination recognition optimization on the real-time fire lane images to obtain a real-time fire lane detection map; Extract the real-time grid lines from the real-time monitoring map of the fire lane to determine the real-time topology map corresponding to the real-time grid lines; Based on the baseline topology map and the real-time topology map, the occupied area data of the fire lane is obtained by determining the occupancy of the fire lane; Based on the occupied area data, the occupancy status of fire lanes is determined through substantive occupancy verification.

10. A non-volatile computer storage medium for identifying fire lane occupancy based on grid line topology, storing computer-executable instructions, characterized in that, The computer-executable instructions are set as follows: Acquire images of unoccupied fire lanes and extract grid lines from the unoccupied fire lane images to determine the baseline topology map corresponding to the grid lines; Acquire real-time fire lane images and perform illumination recognition optimization on the real-time fire lane images to obtain a real-time fire lane detection map; Extract the real-time grid lines from the real-time monitoring map of the fire lane to determine the real-time topology map corresponding to the real-time grid lines; Based on the baseline topology map and the real-time topology map, the occupied area data of the fire lane is obtained by determining the occupancy of the fire lane; Based on the occupied area data, the occupancy status of fire lanes is determined through substantive occupancy verification.

Images