A bus in-vehicle event recognition method and system based on machine vision

Abstract

This invention relates to the field of data recognition technology, specifically to a machine vision-based method and system for identifying events inside a bus. The method includes the following steps: analyzing the bus interior space, constructing a three-dimensional spatial model of the bus interior, obtaining the model number of the monitoring equipment, analyzing the functional attributes of the monitoring equipment, and designing and deploying the installation location of the monitoring equipment in the three-dimensional spatial model of the bus interior based on the monitoring equipment's signals and functional attributes; obtaining the confirmed installation points of the monitoring equipment in the three-dimensional spatial model of the bus interior. This invention can effectively record and confirm image data of events occurring on the bus and the corresponding passengers, thereby assisting in the effective preservation of evidence through image data when violations occur on the bus, assisting in subsequent handling of violations. Simultaneously, the alarm configuration can warn vulnerable passengers to avoid danger, making the bus riding environment safer and more stable.

Description

Technical Field

[0001] This invention relates to the field of data recognition technology, specifically to a method and system for recognizing events inside a bus based on machine vision. Background Technology

[0002] Buses are a relatively environmentally friendly mode of transportation and are one of the most common means of transport on city roads.

[0003] Because buses carry a large number of passengers and are often crowded, theft, physical altercations, and other violations are not uncommon. These violations can endanger unrelated individuals on board. Furthermore, the chaotic state of the passengers during such incidents makes it difficult to accurately determine the cause of the violation based solely on passenger testimonies. While cameras are currently installed inside buses to address this issue, most cameras have limited viewing angles, monitoring only a small area. This results in functional deficiencies, and the chaotic environment often makes the video footage unclear, hindering the identification of perpetrators and complicating subsequent investigations. Summary of the Invention

[0004] Technical problems to be solved

[0005] In view of the above-mentioned shortcomings of the existing technology, the present invention provides a method and system for recognizing events inside buses based on machine vision, which solves the problems mentioned in the background art.

[0006] Technical solution

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] Firstly, a machine vision-based method for recognizing events inside a bus includes the following steps:

[0009] Step 1: Analyze the interior space of the bus, construct a three-dimensional space model of the bus, obtain the model of the monitoring equipment, analyze the functional attributes of the monitoring equipment, and design and deploy the installation location of the monitoring equipment in the three-dimensional space model of the bus based on the signal and functional attributes of the monitoring equipment.

[0010] Step 2: Obtain the confirmed installation locations for the monitoring equipment in the 3D spatial model inside the bus, and install the monitoring equipment inside the bus according to the installation locations in the 3D spatial model inside the bus.

[0011] Step 3: Collect facial image data and clothing features of passengers boarding the vehicle, and temporarily label and match facial data images and clothing features of passengers boarding the vehicle.

[0012] Step 4: Distribute the collected facial data images and clothing features of boarding passengers to all monitoring devices other than the data collection and monitoring device. The monitoring devices collect facial data and clothing features of passengers in the monitored area and compare the collected results with the received facial data images and clothing features of boarding passengers.

[0013] Step 5: Obtain the comparison results and configure each monitoring device to use the passengers present in the comparison results as monitoring targets;

[0014] Step 6: Establish a threat behavior determination database, collect audio and image data of the characteristic behaviors of each passenger under each monitoring device in real time, and compare them with the data content in the threat behavior determination database in real time.

[0015] Step 7: Obtain the comparison results. When there are similar items between the collected passenger characteristic behavior data (audio, image) and the data stored in the threat behavior judgment database, trigger the audio alarm based on the result of the similarity judgment.

[0016] Step 8: Obtain the location of the threat behavior, drive the monitoring equipment to shift its operating angle, and collect image data of the target and the threat behavior in real time.

[0017] Furthermore, the analysis of the functional attributes of the monitoring device in Step 1 includes: the viewing angle of the monitoring device, the resolution of the monitoring device camera, the data upload rate of the monitoring device, and the memory size of the monitoring device.

[0018] Furthermore, step 3 includes sub-steps, comprising the following steps:

[0019] Step 31: Build an online identity authentication and payment platform, establish an interaction channel between the identity authentication and payment platform and the public security target personnel query system, and receive real-time updates from the public security target personnel query system;

[0020] Step 32: Store user information through identity authentication and payment platforms in real time. Use the start of the bus as a trigger signal to compare and query each user's information with the public security target personnel query system after the bus starts.

[0021] Furthermore, the monitoring equipment that collects facial image data and clothing features of boarding passengers in Step 3 does not participate in the distribution of facial data images and clothing features of boarding passengers in Step 4. In Steps 4 and 5, the comparison of passenger facial data images and clothing features obtained by the monitoring equipment and the passenger configuration are used as monitoring targets of the monitoring equipment, and all monitoring equipment participates in the operation.

[0022] Furthermore, Step 4 and Step 5 are configured with a refresh cycle, and Step 4 and Step 5 are reset and run after each refresh cycle.

[0023] The initial default refresh cycle is set to refresh every time a bus arrives at a new bus stop.

[0024] Furthermore, a sub-step is arranged between step 4 and step 5:

[0025] Step 51: Obtain the passenger overlap targets from the monitoring devices, determine the passenger with the highest comparison similarity and configure it in the monitoring target passenger directory of the corresponding monitoring device, and discard the target passengers found in the monitoring devices that overlap with the other passengers.

[0026] Furthermore, the threat behavior image data collected in Step 8 is sent in real time to the user terminal of the public security target personnel query system through the data interaction channel established in Step 31, so that the user terminal operator can receive and view it on the electronic device deployed in the system.

[0027] Secondly, a machine vision-based in-bus event recognition system includes:

[0028] The control terminal is the main control terminal of the system, used to issue execution commands:

[0029] The deployment module is used to deploy the monitoring module to the installation location inside the bus;

[0030] The monitoring module is used to capture real-time image data inside the bus.

[0031] The data acquisition module is used to collect behavioral characteristic data of passengers inside the bus.

[0032] The database is used to store image data of passenger behavior characteristics inside the bus collected by the acquisition module; to store image data of threatening behavior characteristics; and to store passenger behavior characteristic data that is determined to be threatening by the analysis module.

[0033] The analysis module is used to analyze whether there are similarities between the passenger behavior feature image data collected by the acquisition module and the threat behavior feature image data stored in the database;

[0034] The capture module is used to capture the real-time location of users whose image data shows threatening behavioral characteristics and are similar to those analyzed in the analysis module.

[0035] The driver module is used to drive the monitoring equipment in the capture module of the monitoring module to adjust the angle of the area where the target user's real-time location is located, with the target user as the tracking object.

[0036] Furthermore, the database is equipped with sub-modules, including:

[0037] The management unit is used to manage the internal storage space of the database and to divide the storage space according to the type of stored data.

[0038] The space partitioning logic is set according to user-defined settings, and the management unit is also used for updating and refreshing the data stored in the internal space.

[0039] Furthermore, the control terminal is electrically connected to a deployment module via a medium. The deployment module is electrically connected to a sub-module monitoring module via a medium. The deployment module is electrically connected to a data acquisition module via a medium. The data acquisition module is electrically connected to a database via a medium. A sub-module management unit is deployed in the database. The database is electrically connected to an analysis module, a capture module, and a drive module via a medium. The drive module is electrically connected to the monitoring module via a medium.

[0040] Beneficial effects

[0041] Compared with known public technologies, the technical solution provided by this invention has the following beneficial effects:

[0042] 1. This invention provides a machine vision-based method for recognizing events inside buses for use in maintaining bus passenger safety. By using this method, it is possible to effectively record and confirm the image data of events occurring on the bus and the corresponding passengers. This helps to effectively preserve evidence through image data when violations occur on the bus, in order to assist in the subsequent handling of violations. At the same time, the alarm can be configured to warn vulnerable passengers to avoid danger, making the bus riding environment safer and more stable.

[0043] 2. This invention provides a machine vision-based bus in-vehicle event recognition system for use in bus passenger safety maintenance. This system can effectively monitor and manage passenger conditions inside the bus, and by establishing a database, it effectively provides a basis for judging violations, thereby achieving an early warning effect.

[0044] 3. When used, this invention can effectively manage the monitoring equipment by monitoring its deployment, ensuring that the image data collected by the monitoring equipment is clear, the operation is orderly, and the operation tasks are clear, thereby making the system modules and methods more stable and reliable in specific applications. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in this invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of this invention, and those skilled in the art can obtain other drawings based on these drawings without any creative effort.

[0046] Figure 1 This is a flowchart illustrating the first part of a machine vision-based method for recognizing events inside a bus.

[0047] Figure 2 This is a flowchart illustrating the second part of a machine vision-based method for recognizing events inside a bus.

[0048] Figure 3 This is a schematic diagram of the structure of a machine vision-based bus in-vehicle event recognition system;

[0049] The numbers in the diagram represent: 1. Control terminal; 2. Deployment module; 21. Monitoring module; 3. Acquisition module; 4. Database; 41. Management module; 5. Analysis module; 6. Capture module; 7. Driver module. Detailed Implementation

[0050] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0051] The present invention will be further described below with reference to embodiments.

[0052] Example 1

[0053] This embodiment presents a machine vision-based method for recognizing events inside a bus, such as... Figure 1 and Figure 2 As shown, it includes the following steps:

[0054] Step 1: Analyze the interior space of the bus, construct a three-dimensional space model of the bus, obtain the model of the monitoring equipment, analyze the functional attributes of the monitoring equipment, and design and deploy the installation location of the monitoring equipment in the three-dimensional space model of the bus based on the signal and functional attributes of the monitoring equipment.

[0055] Step 2: Obtain the confirmed installation locations for the monitoring equipment in the 3D spatial model inside the bus, and install the monitoring equipment inside the bus according to the installation locations in the 3D spatial model inside the bus.

[0056] Step 3: Collect facial image data and clothing features of passengers boarding the vehicle, and temporarily label and match facial data images and clothing features of passengers boarding the vehicle.

[0057] Step 4: Distribute the collected facial data images and clothing features of boarding passengers to all monitoring devices other than the data collection and monitoring device. The monitoring devices collect facial data and clothing features of passengers in the monitored area and compare the collected results with the received facial data images and clothing features of boarding passengers.

[0058] Step 5: Obtain the comparison results and configure each monitoring device to use the passengers present in the comparison results as monitoring targets;

[0059] Step 6: Establish a threat behavior determination database, collect audio and image data of the characteristic behaviors of each passenger under each monitoring device in real time, and compare them with the data content in the threat behavior determination database in real time.

[0060] Step 7: Obtain the comparison results. When there are similar items between the collected passenger characteristic behavior data (audio, image) and the data stored in the threat behavior judgment database, trigger the audio alarm based on the result of the similarity judgment.

[0061] Step 8: Obtain the location of the threat behavior, drive the monitoring equipment to shift its operating angle, and collect image data of the target and the threat behavior in real time.

[0062] Example 2

[0063] At the implementation level, based on Example 1, this example refers to... Figure 1 The following provides a further detailed explanation of the machine vision-based bus in-vehicle event recognition method in Example 1:

[0064] like Figure 1 As shown, the analysis of the functional attributes of the monitoring equipment in Step 1 includes: the viewing angle of the monitoring equipment, the resolution of the monitoring equipment camera, the data upload rate of the monitoring equipment, and the memory size of the monitoring equipment.

[0065] like Figure 1 As shown, Step 3 includes sub-steps, including the following steps:

[0066] Step 31: Build an online identity authentication and payment platform, establish an interaction channel between the identity authentication and payment platform and the public security target personnel query system, and receive real-time updates from the public security target personnel query system;

[0067] Step 32: Store user information through identity authentication and payment platforms in real time. Use the start of the bus as a trigger signal to compare and query each user's information with the public security target personnel query system after the bus starts.

[0068] This setup allows passengers to pay more conveniently while also providing them with identification information. By verifying this information against a public security target personnel query system, it effectively prevents situations that could threaten passengers' boarding experience. This creates a barrier to entry for passengers before they even board, providing them with a higher level of security.

[0069] like Figure 1 As shown, the monitoring equipment that collects facial image data and clothing features of passengers boarding the vehicle in Step 3 does not participate in the distribution of facial image data and clothing features of passengers boarding the vehicle to the monitoring equipment queue in Step 4; in Step 4 and Step 5, the comparison of passenger facial image data and clothing features obtained by the monitoring equipment and the passenger configuration are used as the monitoring targets of the monitoring equipment, and all monitoring equipment participate in the operation.

[0070] This setting allows the method to obtain more feature recognition data from bus passengers, thus providing accurate basis for subsequent target passenger capture.

[0071] like Figure 1 As shown, Step 4 and Step 5 have a refresh cycle. After each refresh cycle, Step 4 and Step 5 are reset and run.

[0072] The initial default refresh cycle is set to refresh every time a bus arrives at a new bus stop.

[0073] like Figure 1 As shown, there are sub-steps between Step 4 and Step 5:

[0074] Step 51: Obtain the passenger overlap targets from the monitoring devices, determine the passenger with the highest comparison similarity and configure it in the monitoring target passenger directory of the corresponding monitoring device, and discard the target passengers found in the monitoring devices that overlap with the other passengers.

[0075] This setup allows passenger data to be updated synchronously when there are passenger changes at bus stops, thereby making the monitoring equipment more effective in monitoring and managing passengers.

[0076] like Figure 2As shown, the threat behavior image data collected in Step 8 is sent in real time to the user terminal of the public security target personnel query system through the data interaction channel established in Step 31, so that the user terminal can receive and view it on the electronic devices deployed in the system.

[0077] Example 3

[0078] At the implementation level, based on Example 1, this example refers to... Figure 3 The following provides a further detailed explanation of the machine vision-based bus in-vehicle event recognition method in Example 1:

[0079] A machine vision-based in-bus event recognition system includes:

[0080] Control terminal 1 is the system's main control terminal, used to issue execution commands:

[0081] Deployment module 2 is used to deploy monitoring module 21 at the installation location inside the bus;

[0082] Monitoring module 21 is used to capture real-time image data inside the bus;

[0083] Data acquisition module 3 is used to collect behavioral characteristic data of passengers inside the bus;

[0084] Database 4 is used to store image data of passenger behavior characteristics inside the bus collected by module 3; image data of threatening behavior characteristics; and passenger behavior characteristic data identified as threatening behavior by analysis module 5.

[0085] Analysis module 5 is used to analyze whether there are similarities between the passenger behavior feature image data collected by the comparison and acquisition module 3 and the threat behavior feature image data stored in the database 4.

[0086] The capture module 6 is used to capture the real-time location of users whose image data shows threatening behavioral characteristics analyzed in the analysis module 5.

[0087] The drive module 7 is used to drive the monitoring device in the monitoring module 21, which is located in the area where the target user's real-time location is captured by the capture module 6, to adjust its angle, with the target user as the tracking object.

[0088] In this embodiment, the control terminal 1 controls the deployment module 2 to deploy the monitoring module 21 to its installation position inside the bus, thereby enabling the monitoring module 21 to capture real-time image data inside the bus. The acquisition module 3 then collects the behavioral characteristic data of the passengers inside the bus and sends the collected passenger behavioral characteristic data to the database 4 for storage. The analysis module 5 then performs analysis and comparison to check whether there are similarities between the passenger behavioral characteristic image data collected by the acquisition module 3 and the threat behavioral characteristic image data stored in the database 4. The capture module 6 captures the real-time location of the target user identified by the analysis module 5 as having similar threat behavioral characteristic image data. When a threat target passenger appears, the drive module 7 drives the monitoring equipment in the monitoring module 21 located in the area where the threat target passenger is captured by the capture module 6 to adjust its angle, using the target passenger as the tracking object, and records the target passenger's behavioral image data in real time.

[0089] like Figure 3 As shown, database 4 contains sub-modules, including:

[0090] Management unit 41 is used to manage the internal storage space of database 4 and to divide the storage space according to the type of stored data;

[0091] The space partitioning logic is set according to user-defined settings, and the management unit 41 is also used for updating and refreshing the data stored in the internal space.

[0092] The module settings can provide a reasonable basis for the use of Database 4, enabling Database 4 to operate and be used more reasonably and stably.

[0093] like Figure 3 As shown, the control terminal 1 is electrically connected to the deployment module 2 via a medium. The deployment module 2 is electrically connected to the sub-module monitoring module 21 via a medium. The deployment module 2 is electrically connected to the acquisition module 3 via a medium. The acquisition module 3 is electrically connected to the database 4 via a medium. The database 4 is deployed with the sub-module management unit 41. The database 4 is electrically connected to the analysis module 5, the capture module 6, and the drive module 7 via a medium. The drive module 7 is electrically connected to the monitoring module 21 via a medium.

[0094] In summary, the technical solutions provided in the above embodiments can effectively record and confirm image data of events occurring on buses and the corresponding passengers. This assists in preserving evidence through image data when violations occur on buses, aiding in subsequent handling of violations. Simultaneously, the alarm configuration can warn vulnerable passengers to avoid danger, making the bus riding environment safer and more stable. Furthermore, the system can effectively monitor and manage passenger conditions on buses, providing a database to effectively determine violations and thus achieving an early warning effect. In addition, the monitoring and deployment of the monitoring equipment allows for effective and reasonable management, ensuring clear image data acquisition, orderly operation, and clear operational tasks, making the system modules and methods more stable and reliable in practical applications.

[0095] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for recognizing events inside a bus based on machine vision, characterized in that, Includes the following steps: Step 1: Analyze the interior space of the bus, construct a three-dimensional space model of the bus, obtain the model of the monitoring equipment, analyze the functional attributes of the monitoring equipment, and design and deploy the installation location of the monitoring equipment in the three-dimensional space model of the bus based on the signal and functional attributes of the monitoring equipment. Step 2: Obtain the confirmed installation locations for the monitoring equipment in the 3D spatial model inside the bus, and install the monitoring equipment inside the bus according to the installation locations in the 3D spatial model inside the bus. Step 3: Collect facial image data and clothing features of passengers boarding the vehicle, and temporarily label and match facial data images and clothing features of passengers boarding the vehicle. Step 4: Distribute the collected facial data images and clothing features of boarding passengers to all monitoring devices other than the data collection and monitoring device. The monitoring devices collect facial data and clothing features of passengers in the monitored area and compare the collected results with the received facial data images and clothing features of boarding passengers. Step 5: Obtain the comparison results and configure each monitoring device to use the passengers present in the comparison results as monitoring targets; Step 6: Establish a threat behavior determination database, collect audio and image data of the characteristic behaviors of each passenger under each monitoring device in real time, and compare them with the data content in the threat behavior determination database in real time. Step 7: Obtain the comparison results. When there are similar items between the collected passenger characteristic behavior data (audio, image) and the data stored in the threat behavior judgment database, trigger the audio alarm based on the result of the similarity judgment. Step 8: Obtain the location of the threat behavior, drive the monitoring equipment to shift its operating angle, and collect image data of the target and the threat behavior in real time.

2. The method for recognizing events inside a bus based on machine vision according to claim 1, characterized in that, The analysis of the functional attributes of the monitoring equipment in Step 1 includes: the viewing angle of the monitoring equipment, the resolution of the camera of the monitoring equipment, the data upload rate of the monitoring equipment, and the memory size of the monitoring equipment.

3. The method for recognizing events inside a bus based on machine vision according to claim 1, characterized in that, Step 3 includes sub-steps, including the following steps: Step 31: Build an online identity authentication and payment platform, establish an interaction channel between the identity authentication and payment platform and the public security target personnel query system, and receive real-time updates from the public security target personnel query system; Step 32: Store user information through identity authentication and payment platforms in real time. Use the start of the bus as a trigger signal to compare and query each user's information with the public security target personnel query system after the bus starts.

4. The method for recognizing events inside a bus based on machine vision according to claim 1, characterized in that, The monitoring equipment that collects facial image data and clothing features of boarding passengers in Step 3 does not participate in the distribution of facial image data and clothing features of boarding passengers to the monitoring equipment queue in Step 4; in Step 4 and Step 5, the comparison of passenger facial image data and clothing features obtained by the monitoring equipment and the passenger configuration are used as the monitoring targets of the monitoring equipment, and all monitoring equipment participate in the operation.

5. The method for recognizing events inside a bus based on machine vision according to claim 1, characterized in that, A refresh cycle is deployed in Step 4 and Step 5. After each refresh cycle, Step 4 and Step 5 are reset and run. The initial default refresh cycle is set to refresh every time a bus arrives at a new bus stop.

6. The method for recognizing events inside a bus based on machine vision according to claim 1, characterized in that, There are sub-steps between Step 4 and Step 5: Step 51: Obtain the passenger overlap targets from the monitoring devices, determine the passenger with the highest comparison similarity and configure it in the monitoring target passenger directory of the corresponding monitoring device, and discard the target passengers found in the monitoring devices that overlap with the other passengers.

7. The method for recognizing events inside a bus based on machine vision according to claim 3, characterized in that, The threat behavior-generating target and threat behavior image data collected in Step 8 are sent in real time to the public security target personnel query system user terminal through the data interaction channel established in Step 31, so that the public security target personnel query system user terminal can receive and view them on the electronic devices deployed in the system.

8. A machine vision-based bus in-vehicle event recognition system, wherein the system is an implementation system of the machine vision-based bus in-vehicle event recognition method of claim 1, characterized in that, include: The control terminal (1) is the main control terminal of the system, used to issue execution commands: Deployment module (2) is used to deploy the monitoring module (21) at the installation location inside the bus; The monitoring module (21) is used to capture real-time image data inside the bus; The data acquisition module (3) is used to collect behavioral characteristic data of passengers inside the bus; Database (4) is used to store the behavioral characteristic image data of passengers in the bus in the acquisition module (3); Used to store image data of threatening behavior characteristics; Used to store passenger behavior characteristic data that are determined to be threatening behaviors by the analysis module (5); Analysis module (5) is used to analyze whether there are similarities between the passenger behavior feature image data collected by comparison and acquisition module (3) and the threat behavior feature image data stored in database (4); The capture module (6) is used to capture the real-time location of users whose image data shows threatening behavior characteristics analyzed in the analysis module (5). The driving module (7) is used to drive the monitoring equipment in the monitoring module (21) located in the area where the target user's real-time location is captured by the capture module (6) to adjust the angle, with the target user as the tracking object.

9. A machine vision-based bus in-vehicle event recognition system according to claim 8, characterized in that, The database (4) contains sub-modules, including: The management unit (41) is used to manage the internal storage space of the database (4) and to divide the storage space according to the storage data category; The space partitioning logic is set according to user-defined settings, and the management unit (41) is also used for updating and refreshing the data stored in the internal space.

10. A machine vision-based bus in-vehicle event recognition system according to claim 8, characterized in that, The control terminal (1) is electrically connected to a deployment module (2) via a medium. The deployment module (2) is electrically connected to a monitoring module (21) via a medium. The deployment module (2) is electrically connected to a data acquisition module (3) via a medium. The data acquisition module (3) is electrically connected to a database (4) via a medium. The database (4) is equipped with a management unit (41). The database (4) is electrically connected to an analysis module (5), a capture module (6), and a drive module (7) via a medium. The drive module (7) is electrically connected to the monitoring module (21) via a medium.

Images