Intelligent bus video linkage snapshot control system
Through the smart bus video linkage capture control system, the linkage function of the on-board camera and cloud control center is used to monitor the inside and outside of the bus in real time, solving the problems of degradation in video quality and blind spots in the existing system, improving operational safety and management efficiency, and providing Safer and orderly public transportation services.
Patent Information
- Application Number
- CN202411038206.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-31
- Publication Date
- 2024-12-03
AI Technical Summary
The existing bus monitoring system has caused video quality to decline under factors such as complex light changes, passenger occlusion and vehicle vibration, making it difficult to cover blind spots in the viewing angle, and is unable to effectively monitor the situation outside the vehicle, affecting the safety and management efficiency of bus operations.
Design a smart bus video linkage capture control system. By installing on-board cameras inside and outside the bus, collecting data in real time and uploading it to the cloud control center, using object detection algorithms and deep learning models to detect illegal events, calculate camera adjustment parameters, and realize linkage capture , ensure real-time monitoring of the inside and outside of the car.
Real-time monitoring of the inside and outside of the bus, automatic capture of illegal incidents, improve the safety and management efficiency of bus operations, provide decision-making support for traffic management, and improve the service quality and image of urban public transportation.
Smart Images

Figure CN119071429A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of smart public transportation technology, and in particular to a smart public transportation video linkage capture control system. Background Art
[0002] As an important part of urban transportation, urban public transportation is the preferred means of transportation for low-carbon, environmentally friendly and green travel advocated by the government and the public. The public transportation system has the advantages of large carrying capacity, high transportation efficiency, low energy consumption, relatively low pollution and low transportation cost. It has been recognized by countries around the world as the best strategy to solve the traffic problems of large and medium-sized cities. It is also one of the important infrastructures for the survival of cities. However, with the rapid development of the economy, the number of motor vehicles in cities has continued to increase. At the same time, due to factors such as poor comfort and long waiting time of public transportation, urban traffic flow has continued to increase, traffic accidents have occurred frequently, and traffic congestion has become increasingly severe. Therefore, how to ensure the safe operation of public transportation vehicles and efficiently dispatch limited public transportation vehicle resources has become a major problem that plagues public transportation system managers.
[0003] At present, many cities have built bus monitoring and management systems, which monitor buses by positioning them and installing monitoring devices on them. However, factors such as complex light changes in existing buses, obstruction by passengers, and vibration during vehicle driving may cause video quality to deteriorate and affect the accuracy of capturing. At the same time, although the layout of cameras is constantly being optimized, there may still be some blind spots that are difficult to cover, resulting in the inability to effectively monitor the situation in some areas. In addition, the existing bus monitoring system mostly monitors the inside of the bus and cannot monitor the outside of the bus, which has an adverse impact on bus driving and cannot provide effective decision-making support for the safety and management efficiency of bus operations.
[0004] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention
[0005] In view of the problems in the related technology, the present invention proposes a smart bus video linkage capture control system to overcome the above-mentioned technical problems existing in the existing related technology.
[0006] The technical solution of the present invention is achieved in this way:
[0007] A smart bus video linkage capture control system, comprising: a bus video monitoring module, a cloud control center and a linkage capture strategy module, wherein;
[0008] The bus video monitoring module is used to install on-board cameras inside and outside the bus, collect vehicle driving condition data and vehicle behavior data in real time, and display and store them in the driver's cab. At the same time, the real-time collected vehicle driving condition data and vehicle behavior data are uploaded to the cloud control center through the vehicle-ground wireless network;
[0009] The cloud control center is used to receive vehicle driving road condition data and in-vehicle behavior data, detect illegal events in the field of view, and calibrate the position and range of the illegal events in the field of view, and calculate the adjustment parameters of the vehicle camera based on the position and range in the field of view, and send control instructions to the linkage capture strategy module;
[0010] The linkage capture strategy module is used to receive control instructions to execute vehicle-mounted camera adjustments, and mark illegal events at the center of the screen for capture.
[0011] Furthermore, the bus video surveillance module includes: a plurality of vehicle-mounted cameras, a hard disk recorder and a touch monitoring screen, wherein: the plurality of vehicle-mounted cameras are used to monitor the passenger compartment, the driver's cab and the conditions in front of the vehicle respectively, and transmit the passenger compartment, the driver's cab and the conditions in front of the vehicle to the hard disk recorder for storage, and display the monitoring images on the touch monitoring screen at the same time.
[0012] Furthermore, the cloud control center includes the following steps:
[0013] Pre-mark the types of illegal events, including: illegal events related to vehicle driving conditions and illegal events related to in-vehicle behavior;
[0014] Detect illegal events on vehicle driving data and in-vehicle behavior data respectively;
[0015] Obtain the location and range of the illegal event within the field of view and calculate the adjustment parameters of the vehicle-mounted camera.
[0016] Furthermore, the illegal event detection is performed on the vehicle driving road condition data, including the following steps:
[0017] Use target detection algorithms to detect targets in images in advance and track them to obtain motion trajectories and state changes;
[0018] According to the pre-calibrated illegal event type, it is determined whether the target of the vehicle driving road condition is an illegal event.
[0019] Furthermore, the illegal event detection is performed on the in-vehicle behavior data, including the following steps:
[0020] The object detection model using deep learning algorithms is pre-adopted to detect people in the video;
[0021] The target detection model is used to analyze in-vehicle behavior data in real time. When a pre-calibrated illegal event type is detected, relevant video clips are screened.
[0022] Furthermore, the location and scope of the illegal event within the visual area are demarcated, including the following steps:
[0023] Pre-trained deep learning-based object detection algorithms to identify objects in the field of view, including outputting bounding boxes or key point information of the objects;
[0024] According to the bounding box or key point information of the target, the position and range of the target in the field of view are calculated.
[0025] This technical solution calibrates the entire image area as the field of view, and uses the center position, upper left corner position or other relevant positions of the bounding box to represent the position of the target in the field of view; at the same time, based on the bounding box information of the target, the range of the person in the field of view is calculated. The width, height or area of the bounding box can be used to represent it.
[0026] Furthermore, the linkage capture strategy module also includes: a tracking adjustment unit, which is used to mark the illegal event at the center of the screen and take a close-up zoom capture.
[0027] Beneficial effects of the present invention:
[0028] Through the on-board cameras installed inside and outside the bus, the vehicle driving road condition data and in-vehicle behavior data are collected in real time, displayed and stored in the driver's cab, and uploaded to the cloud control center through the vehicle-ground wireless network to detect illegal events in the field of view, and mark the position and range of illegal events in the field of view. The adjustment parameters of the on-board camera are calculated by the position and range in the field of view. The linkage capture strategy module receives the control command to execute the on-board camera adjustment, and marks the illegal event in the center of the screen for capture, so as to realize real-time monitoring inside and outside the bus. Inside the bus, it can capture the abnormal behavior of passengers; outside the bus, it can capture traffic violations on the road. Its linkage function performs well. When an illegal event is detected, the capture operation will be automatically triggered, and the relevant video data will be quickly transmitted to the cloud control center, so that the management personnel can respond and handle it in time. It not only improves the safety and management efficiency of bus operations, but also provides strong support for the optimization of bus operations and the decision-making of traffic management departments, provides safer and more orderly guarantees, and helps to improve the service quality and image of urban public transportation. BRIEF DESCRIPTION OF THE DRAWINGS
[0029] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0030] Figure 1 It is a principle schematic diagram of a smart public transportation video linkage capture control system according to an embodiment of the present invention. DETAILED DESCRIPTION
[0031] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the present invention.
[0032] According to an embodiment of the present invention, a smart public transportation video linkage capture control system is provided.
[0033] like Figure 1 As shown, the smart bus video linkage capture control system according to an embodiment of the present invention includes: a bus video monitoring module 1, a cloud control center 2 and a linkage capture strategy module 3, wherein;
[0034] The bus video monitoring module 1 is used to install on-board cameras inside and outside the bus, collect vehicle driving condition data and in-vehicle behavior data in real time, and display and store them in the driver's cab. At the same time, the real-time collected vehicle driving condition data and in-vehicle behavior data are uploaded to the cloud control center 2 through the vehicle-ground wireless network;
[0035] Among them, the bus video monitoring module 1 includes: a plurality of vehicle-mounted cameras, a hard disk recorder and a touch monitoring screen, wherein; the plurality of vehicle-mounted cameras are used to monitor the passenger compartment, the driver's cab and the conditions in front of the vehicle respectively, and transmit the passenger compartment, the driver's cab and the conditions in front of the vehicle to the hard disk recorder for storage, and display the monitoring images on the touch monitoring screen at the same time.
[0036] Specifically, the on-board camera can be a digital camera, with two explosion-proof hemispherical cameras in the passenger compartment; the driver's cab is equipped with an infrared panoramic camera that can adapt to low illumination and has an integrated microphone for collecting on-site sounds, and another driving camera is configured to monitor the situation in front of the bus.
[0037] In this technical solution, the vehicle-ground wireless network is established by setting up wireless access points and other equipment in the section and on the bus to ensure a stable, secure and conflict-avoiding communication network between the cloud control center 2 along the line and the buses.
[0038] The cloud control center 2 is used to receive the vehicle driving road condition data and the in-vehicle behavior data, detect the illegal events in the field of view, and calibrate the position and range of the illegal events in the field of view, and calculate the adjustment parameters of the vehicle camera by the position and range in the field of view, and send control instructions to the linkage capture strategy module 3;
[0039] The linkage capture strategy module 3 is used to receive control instructions to execute the vehicle camera adjustment, and mark the illegal event at the center of the screen for capture.
[0040] In addition, the cloud control center 2 includes the following steps:
[0041] Pre-mark the types of illegal events, including: illegal events related to vehicle driving conditions and illegal events related to in-vehicle behavior;
[0042] Detect illegal events on vehicle driving data and in-vehicle behavior data respectively;
[0043] Obtain the location and range of the illegal event within the field of view and calculate the adjustment parameters of the vehicle-mounted camera.
[0044] Among them, the detection of illegal events on vehicle driving road condition data includes the following steps:
[0045] Use target detection algorithms to detect targets in images in advance and track them to obtain motion trajectories and state changes;
[0046] According to the pre-calibrated illegal event type, it is determined whether the target of the vehicle driving road condition is an illegal event.
[0047] Specifically, this technical solution collects a large amount of image data containing possible illegal events, and performs preprocessing, such as cropping, resizing, denoising, contrast enhancement, etc., to improve image quality and the accuracy of subsequent processing; at the same time, it clarifies the type of illegal events that need to be detected and extracts related features. For example, for illegal parking at bus stops, pay attention to the parking position of vehicles in specific areas. At the same time, target detection algorithms, such as Faster R-CNN and YOLO based on deep learning, are used to detect relevant targets in the image, such as vehicles, and track them to obtain their movement trajectories and state changes.
[0048] In addition, for the above-mentioned target detection algorithm, the detection model is trained using labeled data sets, and the model parameters are continuously optimized to improve the accuracy of detection. In addition, based on the pre-calibrated violation event types, it is determined whether the target of the vehicle's driving road condition is a violation event.
[0049] For example: Detecting illegal parking: Detect the vehicle and determine its location, and continue to observe the vehicle at that location. If the vehicle stays in a prohibited parking area for more than a specified time, it will be determined as illegal parking.
[0050] Among them, the illegal event detection of in-vehicle behavior data includes the following steps:
[0051] The object detection model using deep learning algorithms is pre-adopted to detect people in the video;
[0052] The target detection model is used to analyze in-vehicle behavior data in real time. When a pre-calibrated illegal event type is detected, relevant video clips are screened.
[0053] This technical solution installs two explosion-proof hemispherical cameras in the passenger room to ensure full coverage of all areas in the passenger room and collect clear video data. At the same time, the collected video is preliminarily processed, including noise removal, contrast and brightness enhancement, etc., in order to present the image more clearly.
[0054] For the target detection model of the above-mentioned deep learning algorithm, such as the target detection model based on the convolutional neural network (CNN), the target in the video is detected. It uses a large amount of labeled video data containing theft and fighting behaviors to train the model so that it can learn the patterns and characteristics of these behaviors. The target detection model then analyzes the in-vehicle behavior data in real time, and when a pre-calibrated illegal event type is detected, the relevant video clips are screened.
[0055] Specifically, for example, in the calculation and judgment of theft: if a person's hand is detected to quickly approach and touch another person's bag or pocket in a short period of time, accompanied by a nervous expression and looking around, it may be initially judged as a suspected theft. In addition, in the calculation and judgment of fighting: when a rapid and fierce physical conflict occurs between two or more people, lasting for a certain period of time, accompanied by angry expressions and shouting, it may be identified as a fighting incident.
[0056] In addition, the location and scope of the illegal incident within the visual area are marked, including the following steps:
[0057] Pre-trained deep learning-based object detection algorithms to identify objects in the field of view, including outputting bounding boxes or key point information of the objects;
[0058] According to the bounding box or key point information of the target, the position and range of the target in the field of view are calculated.
[0059] This technical solution calibrates the entire image area as the field of view, and uses the center position, upper left corner position or other relevant positions of the bounding box to represent the position of the target in the field of view; at the same time, based on the bounding box information of the target, the range of the person in the field of view is calculated. The width, height or area of the bounding box can be used to represent it.
[0060] In addition, the linkage capture strategy module 3 also includes: a tracking adjustment unit 31, which is used to mark the illegal event at the center of the screen and perform a close-up zoom capture.
[0061] With the help of the above scheme, the vehicle driving road condition data and in-vehicle behavior data are collected in real time through the on-board cameras installed inside and outside the bus, and displayed and stored in the driver's room. At the same time, the real-time collected vehicle driving road condition data and in-vehicle behavior data are uploaded to the cloud control center 2 through the vehicle-ground wireless network, and the illegal events in the field of view are detected, and the position and range of the illegal events in the field of view are marked. The adjustment parameters of the on-board camera are calculated by the position and range in the field of view, and the linkage capture strategy module 3 receives the control instruction to execute the on-board camera adjustment, and marks the illegal event at the center of the screen for capture, so as to realize real-time monitoring inside and outside the bus. In the car, it can capture the abnormal behavior of passengers, such as theft, fighting, etc.; outside the car, it can capture traffic violations on the road, such as vehicles forcibly cutting in, occupying bus lanes, etc. Its linkage function performs well. When an illegal event is detected, the capture operation will be automatically triggered, and the relevant video data will be quickly transmitted to the cloud control center 2, so that the management personnel can respond and handle it in time. For example, when a bus is driving, if the system detects that a vehicle changes lanes illegally and affects driving safety, it will immediately capture the vehicle and send a warning to the driver. For another example, when a bus stops at a station, if a private car illegally parks at the bus station, it will also be captured and the record will be uploaded, which not only improves the safety and management efficiency of bus operations, but also provides strong support for the optimization of bus operations and the decision-making of traffic management departments, providing a safer and more orderly guarantee, and helping to improve the service quality and image of urban public transportation.
[0062] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. After considering the disclosure of the specification and the examples, those skilled in the art will easily think of other embodiments of the present disclosure. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the art that are not disclosed in the present disclosure. The specification and the examples are only regarded as exemplary, and the true scope and spirit of the present disclosure are indicated by the claims.
[0063] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
Claims
1. A smart public transportation video linkage capture control system, characterized in that: include: A bus video monitoring module (1), a cloud control center (2) and a linkage capture strategy module (3), wherein; The bus video monitoring module (1) is used to install on-board cameras inside and outside the bus respectively, collect vehicle driving road condition data and vehicle behavior data in real time, and display and store them in the driver's cab, and upload the real-time collected vehicle driving road condition data and vehicle behavior data to the cloud control center (2) through the vehicle-ground wireless network; The cloud control center (2) is used to receive vehicle driving road condition data and in-vehicle behavior data, detect illegal events within the field of view, and calibrate the position and range of the illegal events within the field of view, and calculate the adjustment parameters of the vehicle camera based on the position and range within the field of view, and send a control instruction to the linkage capture strategy module (3); The linkage capture strategy module (3) is used to receive control instructions to execute vehicle-mounted camera adjustments, and to mark illegal events at the center of the screen for capture.
2. The smart public transportation video linkage capture control system according to claim 1 is characterized in that: The bus video monitoring module (1) comprises: a plurality of vehicle-mounted cameras, a hard disk recorder and a touch monitoring screen, wherein the plurality of vehicle-mounted cameras are used to monitor the passenger compartment, the driver's cab and the conditions in front of the vehicle respectively, and transmit the passenger compartment, the driver's cab and the conditions in front of the vehicle to the hard disk recorder for storage, while displaying the monitoring images on the touch monitoring screen.
3. The smart public transportation video linkage capture control system according to claim 1 is characterized in that: The cloud control center (2) comprises the following steps: Pre-mark the types of illegal events, including: illegal events related to vehicle driving conditions and illegal events related to in-vehicle behavior; Detect illegal events on vehicle driving data and in-vehicle behavior data respectively; Obtain the location and range of the illegal event within the field of view and calculate the adjustment parameters of the vehicle-mounted camera.
4. The smart public transportation video linkage capture control system according to claim 3 is characterized in that: Detecting illegal events on vehicle driving road condition data includes the following steps: Use target detection algorithms to detect targets in images in advance and track them to obtain motion trajectories and state changes; According to the pre-calibrated illegal event type, it is determined whether the target of the vehicle driving road condition is an illegal event.
5. The smart public transportation video linkage capture control system according to claim 3 is characterized in that: Detecting illegal events on in-vehicle behavior data includes the following steps: The object detection model using deep learning algorithms is pre-adopted to detect people in the video; The target detection model is used to analyze in-vehicle behavior data in real time. When a pre-calibrated illegal event type is detected, relevant video clips are screened.
6. The smart public transportation video linkage capture control system according to claim 4 or 5, characterized in that: Determine the location and scope of the illegal incident within the visual area, including the following steps: Pre-trained deep learning-based object detection algorithms to identify objects in the field of view, including outputting bounding boxes or key point information of the objects; According to the bounding box or key point information of the target, the position and range of the target in the field of view are calculated.
7. The smart public transportation video linkage capture control system according to claim 1 is characterized in that: The linkage capture strategy module (3) also includes: a tracking adjustment unit (31) for marking the illegal event at the center of the screen and performing a close-up zoom capture.
