Parking lot monitoring video system

Abstract

The invention discloses a parking lot monitoring video system, belongs to the technical field of parking lot monitoring videos, and solves the problems that the stored monitoring video data is extremely large, and if cache cleaning needs to be carried out for one month or even three months, the memory of data storage is relatively high in requirement, and the data storage efficiency is high. Therefore, the problem that the monitoring video cannot be stored for a long time is solved. Comprising a monitoring module and a data processing module, the output end of the monitoring module is coupled with the input end of the data processing module, and the monitoring module is used for collecting a scene in a parking lot, generating a first-level monitoring picture and transmitting the first-level monitoring picture to the data processing module; and the data processing module is used for receiving the primary monitoring picture and retaining the possibly scratched picture and the scratched picture after multiple times of recognition. According to the system, monitoring pictures are marked in a grading mode, vehicle state changes, whether vehicles exist in adjacent parking spaces or not and rubbing conditions are recognized, relevant videos are finally stored in a classified mode, all monitoring videos are prevented from being stored, the data storage burden is relieved, and a vehicle owner can conveniently search for the relevant videos.

Description

Technical Field

[0001] The present invention relates to the technical field of parking lot monitoring video technology, and in particular to a parking lot monitoring video system. Background Art

[0002] Underground parking lots are usually located in the basement of buildings or commercial complexes. They are enclosed parking spaces designed to effectively utilize limited land resources. As underground parking lots usually have strong fire prevention and anti-theft functions, they can not only provide a large number of parking spaces, but also avoid sharing space with ground parking lots. The design of modern underground parking lots focuses on traffic flow, and is usually equipped with multiple lanes, advanced parking guidance systems, and high-definition monitoring equipment to ensure the safety and convenience of parking. The parking spaces in underground parking lots are flexible in layout. In addition to the general plane layout, many designs also use mechanical or three-dimensional parking systems to maximize the use of space and adapt to different types of vehicles. In order to improve the user experience, many underground parking lots are also equipped with facilities such as automatic payment systems, electric vehicle charging piles, and intelligent navigation guidance. As underground parking lots are usually located on the ground floor of buildings, the external environment is relatively quiet and far away from noise pollution, making them an ideal choice for car owners. However, the design of the ventilation system and lighting facilities of underground parking lots is also very critical to ensure air circulation and sufficient light to improve the comfort and safety of the parking environment.

[0003] At present, many facilities are arranged in underground parking lots to ensure that car owners have a good parking experience when parking their vehicles, so that the vehicles can be properly managed. However, when parking in underground parking lots, it is impossible to have personnel to supervise at all times to prevent vehicles from scratching against the vehicles next to them when parking in and out of the garage. Therefore, a monitoring module is set up in the underground parking lot to monitor the situation. However, the area of ​​the entire parking lot is large, and the monitoring video data that is finally saved is also very large. If the cache is to be cleared every one or even three months, there will be high requirements for the memory for data storage, so the monitoring video cannot be saved for a long time.

[0004] Therefore, a parking lot monitoring video system is proposed to solve or alleviate the above problems. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a parking lot monitoring video system.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] A parking lot monitoring video system includes a monitoring module and a data processing module. The output end of the monitoring module is coupled to the input end of the data processing module. The monitoring module is used to collect scenes in the parking lot and generate a first-level monitoring picture and transmit it to the data processing module. The data processing module is used to receive the first-level monitoring picture and retain possible scratch pictures and scratch pictures after multiple identifications.

[0008] Preferably, there are several monitoring modules, each of which includes a camera, which is arranged at various positions in the parking lot. The camera is used to capture the scene in the parking lot and generate a primary monitoring screen and transmit it to the data processing module.

[0009] Preferably, the data processing module includes a central processing unit, an input end of the central processing unit is coupled to an output end of the monitoring module, and the central processing unit is used to receive the first-level monitoring screen and retain possible scratch screens and scratch screens after multiple identifications.

[0010] Preferably, the data processing module is used to receive the first-level monitoring screen and retain the possible scratch screen and the scratch screen after multiple identifications, including the following steps:

[0011] Use the target detection algorithm to identify the bounding box of the vehicle in each frame of the first-level monitoring screen, and set the position of each vehicle to (x t ,y t ), where t represents the current time frame;

[0012] Use the vehicle tracking algorithm to match each vehicle in multiple frame images to obtain its motion trajectory. The motion trajectory of each vehicle is: Where T is the total number of time frames;

[0013] Calculate the position change of each vehicle, Δx t =|x t -x t-1 | and Δy t =|y t -y t-1 |, where Δx t and Δy t Respectively represent the position changes of the vehicle in the horizontal and vertical directions;

[0014] Set the displacement to Set the displacement threshold to Δx thresh , determine the displacement amount, displacement threshold, and displacement direction. If the change in the rear position of the vehicle at a certain moment compared to the previous moment is moving along the inner direction of the parking space, and x car >Δx thresh, then the result is that the vehicle is reversing, if the change in the rear position of the vehicle at a certain moment compared to the previous moment is moving along the direction outside the parking space, and x car >Δx thresh , then the judgment result is that the vehicle has left the vehicle;

[0015] If the determination result is that the vehicle is reversing or moving out, the first-level monitoring screen is marked as a second-level monitoring screen.

[0016] Preferably, the data processing module is used to receive the first-level monitoring screen and retain the possible scratch screen and the scratch screen after multiple identifications, and also includes the following steps:

[0017] The secondary monitoring image is converted into grayscale and subjected to Gaussian filtering to obtain the processed image;

[0018] The processed image is subjected to the canny edge detection algorithm to identify the boundary of the parking space, and then normalized to obtain a normalized image;

[0019] Choose the displacement to be x car The parking space is taken as the main parking space, and the adjacent parking spaces on the left and right sides of the main parking space are extracted according to the parking space number or spatial layout;

[0020] Use the target detection algorithm to determine whether the adjacent parking space is parked. If the adjacent parking space is parked, the output result is 1, and if the adjacent parking space is not parked, the output result is 0;

[0021] When the output result is greater than or equal to 1, the second-level monitoring screen is marked as a third-level monitoring screen. When the output result is less than 1, the mark of the second-level monitoring screen is cancelled.

[0022] Preferably, the data processing module is used to receive the first-level monitoring screen and retain the possible scratch screen and the scratch screen after multiple identifications, and also includes the following steps:

[0023] Grayscale the third-level monitoring screen;

[0024] The grayscaled three-level monitoring screen is used to detect the position of the vehicle through the target detection algorithm;

[0025] Define the parking space location through the parking space number information and extract the main parking space P main =(x,y) main and the area P of the adjacent parking space left =(x,y) left ,P right =(x,y) right , where (x,y) main 、(x,y) left 、(x,y) rightRespectively represent the bounding box positions of the main parking space and the adjacent parking space;

[0026] The object detection algorithm returns a bounding box for each vehicle, and if the vehicle is parked in an adjacent parking space, the bounding box corresponds to the outline of the vehicle;

[0027] The movement trajectory of each vehicle Calculate the distance from the vehicle outline in the adjacent parking space,

[0028]

[0029] A safety distance threshold is set. When the distance is less than the safety distance threshold, there is a risk of scratching, and the third-level monitoring screen is a fourth-level monitoring screen;

[0030] The third-level monitoring screens and the fourth-level monitoring screens are classified and saved.

[0031] Preferably, the vehicle tracking algorithm includes a DeepSORT algorithm, and the target detection algorithm includes a Yolo algorithm.

[0032] The present invention has the following beneficial effects:

[0033] In the monitoring video system of the present invention, there are multiple monitoring modules to monitor and shoot the situation in the parking lot, and obtain multiple first-level monitoring pictures. When a vehicle status change (reversing, exiting) appears in the first-level monitoring picture, the first-level monitoring picture is marked as a second-level monitoring picture, and the content of the second-level monitoring picture is identified to identify whether there is an adjacent parking space with a car parked next to the parking space where the vehicle status change appears in the second-level monitoring picture. If it is found after identification that there is no adjacent parking space with a car parked, the mark of the second-level monitoring picture is cancelled. If it is found after identification that there is an adjacent parking space with a car parked, the second-level monitoring picture is marked as a third-level monitoring picture, and the content of the third-level monitoring picture is identified to determine whether a vehicle scratch occurs. In the event that a vehicle scratch is identified, the third-level monitoring screen is marked as a fourth-level monitoring screen; if the vehicle scratch is not identified, it is still marked as a third-level monitoring screen. Finally, the third-level monitoring screen and the fourth-level monitoring screen are classified and saved. In this way, even if the entire parking lot needs to be filmed by a surveillance video system, there is no need to retain conventional surveillance videos. Instead, surveillance videos that may have a risk of scratches can be graded and saved. When the owner of a vehicle parked in the parking lot needs to find the corresponding surveillance video, the corresponding surveillance video can be provided by this system, and there is no need to save all surveillance videos, thereby reducing the burden on the memory required for data storage. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0035] Figure 1 It is a structural block diagram of the present invention.

[0036] 1. Monitoring module; 2. Data processing module. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0038] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0040] In the description of the present invention, it should be understood that the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside", etc. indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, or are the orientations or positional relationships in which the inventive product is conventionally placed when in use, or are the orientations or positional relationships conventionally understood by those skilled in the art. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0041] Furthermore, the terms “first”, “second”, “third”, etc. are merely used for distinguishing descriptions and are not to be understood as indicating or implying relative importance.

[0042] In the description of the present invention, it is also necessary to explain that, unless otherwise clearly specified and limited, the terms "set", "install", "connect", and "connect" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0043] A parking lot monitoring video system, such as Figure 1 As shown, it includes a monitoring module 1 and a data processing module 2. The number of monitoring modules 1 is several. The monitoring module 1 includes a camera. The camera is used to be set at various positions in the parking lot. The camera is used to collect the scene in the parking lot and generate a first-level monitoring picture and transmit it to the data processing module 2. The data processing module 2 includes a central processing unit. The input end of the central processing unit is coupled to the output end of the monitoring module 1. The central processing unit is used to receive the first-level monitoring picture and retain possible scratch pictures and scratch pictures after multiple identifications.

[0044] The output end of the monitoring module 1 is coupled to the input end of the data processing module 2. The monitoring module 1 is used to collect the scene in the parking lot and generate a first-level monitoring screen and transmit it to the data processing module 2. The data processing module 2 is used to receive the first-level monitoring screen and retain possible scratch images and scratch images after multiple identifications.

[0045] Preferably, the data processing module 2 is used to receive the first-level monitoring screen and retain the possible scratch screen and the scratch screen after multiple identifications, including the following steps:

[0046] Use the target detection algorithm to identify the bounding box of the vehicle in each frame of the first-level monitoring screen, and set the position of each vehicle to (x t ,y t ), where t represents the current time frame;

[0047] Use the vehicle tracking algorithm to match each vehicle in multiple frame images to obtain its motion trajectory. The motion trajectory of each vehicle is: Where T is the total number of time frames;

[0048] Calculate the position change of each vehicle, Δx t =|x t -x t-1 | and Δy t =|y t -y t-1 |, where Δx t and Δy t Respectively represent the position changes of the vehicle in the horizontal and vertical directions;

[0049] Set the displacement to Set the displacement threshold to Δx thresh , determine the displacement amount, displacement threshold, and displacement direction. If the change in the rear position of the vehicle at a certain moment compared to the previous moment is moving along the inner direction of the parking space, and x car >Δx thresh , then the result is that the vehicle is reversing, if the change in the rear position of the vehicle at a certain moment compared to the previous moment is moving along the direction outside the parking space, and x car >Δx thresh , then the judgment result is that the vehicle has left the vehicle;

[0050] If the determination result is that the vehicle is reversing or moving out, the first-level monitoring screen is marked as a second-level monitoring screen.

[0051] Preferably, the data processing module 2 is used to receive the first-level monitoring screen and retain the possible scratch screen and the scratch screen after multiple identifications, and also includes the following steps:

[0052] The secondary monitoring image is converted into grayscale and subjected to Gaussian filtering to obtain the processed image;

[0053] The processed image is subjected to the canny edge detection algorithm to identify the boundary of the parking space, and then normalized to obtain a normalized image;

[0054] Choose the displacement to be x car The parking space is taken as the main parking space, and the adjacent parking spaces on the left and right sides of the main parking space are extracted according to the parking space number or spatial layout;

[0055] Use the target detection algorithm to determine whether the adjacent parking space is parked. If the adjacent parking space is parked, the output result is 1, and if the adjacent parking space is not parked, the output result is 0;

[0056] When the output result is greater than or equal to 1, the second-level monitoring screen is marked as a third-level monitoring screen. When the output result is less than 1, the mark of the second-level monitoring screen is cancelled.

[0057] Preferably, the data processing module 2 is used to receive the first-level monitoring screen and retain the possible scratch screen and the scratch screen after multiple identifications, and also includes the following steps:

[0058] Grayscale the third-level monitoring screen;

[0059] The grayscaled three-level monitoring screen is used to detect the position of the vehicle through the target detection algorithm;

[0060] Define the parking space location through the parking space number information and extract the main parking space P main =(x,y) main and the area P of the adjacent parking spaceleft =(x,y) left ,P right =(x,y) right , where (x,y) main 、(x,y) left 、(x,y) right Respectively represent the bounding box positions of the main parking space and the adjacent parking space;

[0061] The object detection algorithm returns a bounding box for each vehicle, and if the vehicle is parked in an adjacent parking space, the bounding box corresponds to the outline of the vehicle;

[0062] The movement trajectory of each vehicle Calculate the distance from the vehicle outline in the adjacent parking space,

[0063]

[0064] A safety distance threshold is set. When the distance is less than the safety distance threshold, there is a risk of scratching, and the third-level monitoring screen is a fourth-level monitoring screen;

[0065] The third-level monitoring screens and the fourth-level monitoring screens are classified and saved.

[0066] Preferably, the vehicle tracking algorithm includes the DeepSORT algorithm, and the target detection algorithm includes the Yolo algorithm.

[0067] The monitoring video system of the present invention uses multiple monitoring modules 1 to comprehensively monitor and shoot the parking lot, and obtains a series of first-level monitoring pictures. In these first-level monitoring pictures, when the system detects a change in the state of the vehicle (such as reversing or leaving the vehicle), the relevant picture will be marked as a second-level monitoring picture. Next, the system will further analyze these second-level monitoring pictures to identify whether there are adjacent parking spaces with parked vehicles next to the parking spaces with status changes in the second-level monitoring pictures. If it is detected that the adjacent parking spaces are empty, the system will cancel the second-level mark of the picture; if it is found that there are vehicles parked in the adjacent parking spaces, the second-level monitoring picture will be upgraded to a third-level monitoring picture.

[0068] The third-level monitoring screen is further analyzed to determine whether a vehicle scratch has occurred. If the system determines that a vehicle scratch has occurred, the third-level screen will be further marked as a fourth-level monitoring screen. If no scratch is detected, the third-level monitoring screen will continue to remain a third-level monitoring screen. Ultimately, all third-level and fourth-level monitoring screens will be classified and stored according to different events.

[0069] Through this hierarchical management method, the system can effectively distinguish different levels of surveillance images, avoiding the tedious work of storing all surveillance images. For ordinary surveillance videos, the system does not need to retain them if there are no obvious abnormalities, thus saving storage space. For those images that may have the risk of scratches or other abnormalities, the system will save them in detail as needed.

[0070] This method not only reduces the unnecessary video storage burden, but also ensures that when car owners need to find a certain surveillance video, they can quickly find the surveillance footage related to their vehicle without having to screen through a large number of ordinary surveillance videos. Through this intelligent and hierarchical storage and retrieval mechanism, the system not only improves the efficiency of parking lot video surveillance, but also greatly optimizes the utilization of storage resources.

[0071] The design concept of the system is to accurately record possible changes in vehicle status and scratches in the parking lot through intelligent identification and hierarchical storage, without having to keep every surveillance video. This not only reduces the pressure on storage space, but also improves the convenience of video data management. When car owners or relevant personnel need to find surveillance records for a specific time period, they can quickly find relevant videos, thereby providing timely and effective evidence support for resolving disputes and accidents in the parking lot.

[0072] In summary, the monitoring video system of the present invention reduces the storage burden while ensuring the accuracy and efficiency of monitoring data through hierarchical labeling and intelligent recognition technology, thereby improving the intelligent level of parking lot video management.

[0073] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A parking lot monitoring video system, characterized in that: The invention comprises a monitoring module (1) and a data processing module (2), wherein the output end of the monitoring module (1) is coupled to the input end of the data processing module (2), the monitoring module (1) is used to collect scenes in a parking lot and generate a first-level monitoring screen and transmit it to the data processing module (2), and the data processing module (2) is used to receive the first-level monitoring screen and retain possible scratch screens and scratch screens after multiple identifications.

2. A parking lot monitoring video system according to claim 1, characterized in that: The number of the monitoring modules (1) is several, and the monitoring modules (1) include cameras, which are arranged at various positions in the parking lot, and are used to collect scenes in the parking lot and generate a primary monitoring screen and transmit it to the data processing module (2).

3. A parking lot monitoring video system according to claim 1, characterized in that: The data processing module (2) comprises a central processing unit, the input end of the central processing unit is coupled to the output end of the monitoring module (1), and the central processing unit is used to receive the first-level monitoring screen and retain the possible scratch screen and the scratch screen after multiple identifications.

4. A parking lot monitoring video system according to claim 1, characterized in that: The data processing module (2) is used to receive the first-level monitoring screen and retain the possible scratch screen and the scratch screen after multiple identifications, and includes the following steps: Use the target detection algorithm to identify the bounding box of the vehicle in each frame of the first-level monitoring screen, and set the position of each vehicle to (x t ,y t ), where t represents the current time frame; Use the vehicle tracking algorithm to match each vehicle in multiple frame images to obtain its motion trajectory. The motion trajectory of each vehicle is {(x t ,y t )} T t＝1 , where T is the total number of time frames; Calculate the position change of each vehicle, Δx t =|x t -x t-1 | and Δy t =|y t -y t-1 |, where Δx t and Δy t Respectively represent the position changes of the vehicle in the horizontal and vertical directions; Set the displacement to Set the displacement threshold to Δx thresh , determine the displacement amount, displacement threshold, and displacement direction. If the change in the rear position of the vehicle at a certain moment compared to the previous moment is moving along the inner direction of the parking space, and x car >Δx thresh , then the result is that the vehicle is reversing, if the change in the rear position of the vehicle at a certain moment compared to the previous moment is moving along the direction outside the parking space, and x car >Δx thresh , then the judgment result is that the vehicle has left the vehicle; If the determination result is that the vehicle is reversing or moving out, the first-level monitoring screen is marked as a second-level monitoring screen.

5. A parking lot monitoring video system according to claim 4, characterized in that: The data processing module (2) is used to receive the first-level monitoring screen and retain the possible scratch screen and the scratch screen after multiple identifications, and also includes the following steps: The secondary monitoring image is converted into grayscale and subjected to Gaussian filtering to obtain the processed image; The processed image is subjected to the canny edge detection algorithm to identify the boundary of the parking space, and then normalized to obtain a normalized image; Choose the displacement to be x car The parking space is taken as the main parking space, and the adjacent parking spaces on the left and right sides of the main parking space are extracted according to the parking space number or spatial layout; Use the target detection algorithm to determine whether the adjacent parking space is parked. If the adjacent parking space is parked, the output result is 1, and if the adjacent parking space is not parked, the output result is 0; When the output result is greater than or equal to 1, the second-level monitoring screen is marked as a third-level monitoring screen. When the output result is less than 1, the mark of the second-level monitoring screen is cancelled.

6. A parking lot monitoring video system according to claim 5, characterized in that: The data processing module (2) is used to receive the first-level monitoring screen and retain the possible scratch screen and the scratch screen after multiple identifications, and also includes the following steps: Grayscale the third-level monitoring screen; The grayscaled three-level monitoring screen is used to detect the position of the vehicle through the target detection algorithm; Define the parking space location through the parking space number information and extract the main parking space P main =(x,y) main and the area P of the adjacent parking space left =(x,y) left , P right =(x,y) right , where (x, y) main 、(x,y) lrft 、(x,y) right Respectively represent the bounding box positions of the main parking space and the adjacent parking space; The object detection algorithm returns a bounding box for each vehicle, and if the vehicle is parked in an adjacent parking space, the bounding box corresponds to the outline of the vehicle; The movement trajectory of each vehicle Calculate the distance from the vehicle outline in the adjacent parking space, A safety distance threshold is set. When the distance is less than the safety distance threshold, there is a risk of scratching, and the third-level monitoring screen is a fourth-level monitoring screen; The third-level monitoring screens and the fourth-level monitoring screens are classified and saved.

7. A parking lot monitoring video system according to claim 6, characterized in that: The vehicle tracking algorithm includes the DeepSORT algorithm, and the target detection algorithm includes the Yolo algorithm.

Images