High-risk road section monitoring snapshot method based on intelligent traffic internet of things
By combining tilt vibration integrated sensors and ultra-black light cameras to collect data in high-risk road sections, and using intelligent transportation IoT and lightweight YOLOv5 models for vehicle detection, the problem of incomplete early warning in existing technologies has been solved, and faster and more accurate early warning response has been achieved.
Patent Information
- Application Number
- CN202511500921.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-21
- Publication Date
- 2026-02-17
AI Technical Summary
Existing methods for monitoring and capturing high-risk road sections fail to effectively combine on-site vehicle environmental data with data on high-risk road structure data, resulting in incomplete early warnings.
The system employs an integrated tilt and vibration sensor to collect tilt and vibration data of structures in high-risk road sections in real time, and combines it with an ultra-black light camera to collect traffic environment videos in real time. The data is then transmitted to the cloud via the intelligent transportation Internet of Things, and a pre-trained lightweight YOLOv5 model is used for vehicle detection and congestion assessment to achieve joint analysis of data and images.
It improves the comprehensiveness and accuracy of early warning, shortens emergency response time, reduces wiring costs, and enhances deployment efficiency and stability.
Smart Images
Figure CN121545362A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of road safety monitoring technology, specifically relating to a method for monitoring and capturing images of high-risk road sections based on the Internet of Things for intelligent transportation. Background Technology
[0002] High-risk road sections have a much higher accident rate than ordinary road sections due to their complex terrain and harsh environment. Real-time and accurate monitoring and capture can quickly complete accident tracing, violation penalties and safety warnings.
[0003] The current monitoring and capture system for high-risk road sections has the following problems: it only focuses on on-site vehicle and environmental data, which is separated from the data on the structures of high-risk road sections, resulting in incomplete early warnings.
[0004] In view of this, a monitoring and capture method for high-risk road sections based on the Internet of Things for intelligent transportation is designed to solve the above problems. Summary of the Invention
[0005] To address the problems mentioned in the background section, this invention provides a method for monitoring and capturing images of high-risk road sections based on the Internet of Things for intelligent transportation. This method features improved early warning comprehensiveness, shorter emergency response time, and higher early warning accuracy.
[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for monitoring and capturing images of high-risk road sections based on the Internet of Things for intelligent transportation, comprising the following steps: S1: The tilt and vibration integrated sensor collects tilt and vibration data of structures in high-risk road sections in real time and transmits them to the cloud via the intelligent transportation Internet of Things, including tilt value, vibration value, timestamp and location coordinates; S2: The ultra-black light camera collects real-time traffic environment videos of high-risk road sections and transmits them to the cloud via the intelligent transportation Internet of Things, including environmental videos, vehicle videos, timestamps, and location coordinates; S3: Preset the tilt angle and vibration abnormality thresholds for structures in high-risk road sections in the cloud; S4: If the tilt angle and / or vibration value of a high-risk road section structure received by the cloud exceeds the preset tilt angle abnormality threshold and / or vibration abnormality threshold, an alarm will be triggered, and the abnormal tilt angle and / or vibration value will be pushed, along with an environmental video of the received abnormal tilt angle and / or vibration value. S5: The cloud receives vehicle videos of the moments preceding and / or following abnormal tilt angle values and vibration values, as well as the current vehicle environment. It uses a pre-trained lightweight YOLOv5 model to detect vehicles in the video of the moments preceding and following abnormal tilt angle values and vibration values. Based on the detected vehicles, it calculates the traffic flow of the moments preceding and following abnormal tilt angle values and vibration values, determines the degree of congestion based on the traffic flow of the moments preceding and following abnormal tilt angle values and vibration values, and determines the correlation between the congestion degree of the moments preceding and following abnormal tilt angle values and vibration values and the abnormal tilt angle values and vibration values. If there is a correlation, a serious alarm is issued.
[0007] Furthermore, in step S4, the specific steps for pushing environmental video of abnormal tilt angle values and / or vibration values include: Obtain the timestamp and location coordinates of the abnormal tilt angle and / or vibration values pushed; Find environmental videos captured by ultra-black light cameras with the same timestamp and location coordinates based on the timestamps and location coordinates of abnormal tilt angle values and / or vibration values. Push environmental videos captured by ultra-black light cameras with the same timestamp and location coordinates.
[0008] Furthermore, in step S5, the specific steps for the cloud to receive the vehicle video before and after the abnormal tilt angle value and / or vibration value include: Obtain the timestamp and location coordinates of the abnormal tilt angle and / or vibration values pushed; Based on the timestamps and location coordinates of abnormal tilt angle values and / or vibration values, find vehicle videos captured by ultra-black light cameras with the same and previous timestamps and location coordinates. The system pushes out vehicle videos captured by ultra-black light cameras with the same timestamp and location coordinates as the previous ones.
[0009] Furthermore, in step S5, the specific steps for detecting abnormal tilt angle values and / or vibration values in the vehicle's current vehicle environment video using the pre-trained lightweight YOLOv5 model include: Decompose the vehicle environment video into continuous video frame images; Convert consecutive video frame images into grayscale images and then normalize them; The normalized grayscale image is scaled to fit the input size of the pre-trained lightweight YOLOv5 model; A grayscale image adapted to the input size is input into a pre-trained lightweight YOLOv5 model. The pre-trained lightweight YOLOv5 model outputs a set of bounding boxes for the vehicle, each bounding box including coordinates and class confidence. A preset category confidence threshold is used to filter the category confidence of each bounding box, thereby selecting valid vehicles.
[0010] Furthermore, in step S5, the specific steps for determining the degree of congestion based on the previous abnormal tilt angle value and / or vibration value and the current traffic flow include: Preset traffic flow thresholds for each congestion level; Based on the preset traffic flow thresholds for each congestion level, the congestion level that matches the traffic flow before and after the abnormal tilt angle value and / or vibration value is identified, and this level is used as the vehicle congestion level before and after the abnormal tilt angle value and / or vibration value.
[0011] Furthermore, in step S5, the specific steps for determining the correlation between the abnormal tilt angle value and / or vibration value, based on the previous moment of the abnormal tilt angle value and / or vibration value and the current level of congestion, include: Determine the magnitude of congestion before and after the abnormal tilt angle and / or vibration value. If the congestion level before the abnormal tilt angle and / or vibration value is greater than or equal to the current congestion level, then it is unrelated to the abnormal tilt angle and / or vibration value. If the congestion level before the abnormal tilt angle and / or vibration value is less than the current congestion level, then it is related to the abnormal tilt angle and / or vibration value.
[0012] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention uses an integrated tilt and vibration sensor to collect the tilt angle and vibration values of structures in high-risk road sections in real time, and an ultra-black light camera to collect vehicle videos in high-risk road sections in real time. It can not only monitor, capture, detect, and alarm the condition of structures in high-risk road sections, but also monitor, capture, detect, and alarm the condition of vehicles in high-risk road sections, thus improving the comprehensiveness of early warning.
[0013] 2. The tilt and vibration integrated sensor of this invention collects the tilt angle and vibration values of structures in high-risk road sections, while simultaneously collecting timestamps and location coordinates. The ultra-black light camera collects vehicle videos of high-risk road sections, environmental videos, timestamps, and location coordinates. When the tilt angle and vibration values of structures in high-risk road sections are abnormal, the environmental video can be quickly retrieved based on the timestamps and location coordinates, improving the comprehensiveness of the early warning. At the same time, the data and images can help to understand the abnormal situation more quickly and shorten the emergency response time (<5min).
[0014] 3. This invention uses a pre-trained lightweight YOLOv5 model to detect vehicles in the current vehicle environment video before and after abnormal tilt angle values and / or vibration values. Based on the detected vehicles, it calculates the traffic flow before and after the abnormal tilt angle values and / or vibration values, determines the degree of congestion, and determines the correlation between the abnormal tilt angle values and / or vibration values and the degree of congestion. If a correlation is found, a serious alarm is issued. The joint analysis of data and video can avoid misjudgment based on single data and improve the accuracy of early warning (+40%).
[0015] 4. This invention is powered by solar energy and batteries, which can reduce wiring costs and improve deployment efficiency (+60%).
[0016] 5. This invention is adapted to the complex environment of high-risk sections of highways, supports wide temperature range of -40 to +55℃, and has strong stability. Attached Figure Description
[0017] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation
[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0019] This invention provides the following technical solution: a method for monitoring and capturing images of high-risk road sections based on the Internet of Things for intelligent transportation, comprising the following steps: S1: The tilt and vibration integrated sensor collects tilt and vibration data of structures in high-risk road sections in real time and transmits them to the cloud via the intelligent transportation Internet of Things, including tilt value, vibration value, timestamp and location coordinates; S2: The ultra-black light camera collects real-time traffic environment videos of high-risk road sections and transmits them to the cloud via the intelligent transportation Internet of Things, including environmental videos, vehicle videos, timestamps, and location coordinates; S3: Preset the tilt angle and vibration abnormality thresholds for structures in high-risk road sections in the cloud; S4: If the tilt angle and / or vibration value of a high-risk road section structure received by the cloud exceeds the preset tilt angle abnormality threshold and / or vibration abnormality threshold, an alarm will be triggered, and the abnormal tilt angle and / or vibration value will be pushed, along with an environmental video of the received abnormal tilt angle and / or vibration value. S5: The cloud receives vehicle videos of the moments preceding and / or following abnormal tilt angle values and vibration values, as well as the current vehicle environment. It uses a pre-trained lightweight YOLOv5 model to detect vehicles in the video of the moments preceding and following abnormal tilt angle values and vibration values. Based on the detected vehicles, it calculates the traffic flow of the moments preceding and following abnormal tilt angle values and vibration values, determines the degree of congestion based on the traffic flow of the moments preceding and following abnormal tilt angle values and vibration values, and determines the correlation between the congestion degree of the moments preceding and following abnormal tilt angle values and vibration values and the abnormal tilt angle values and vibration values. If there is a correlation, a serious alarm is issued.
[0020] Specifically, in step S4, the specific steps for pushing environmental video of abnormal tilt angle values and / or vibration values include: Obtain the timestamp and location coordinates of the abnormal tilt angle and / or vibration values pushed; Find environmental videos captured by ultra-black light cameras with the same timestamp and location coordinates based on the timestamps and location coordinates of abnormal tilt angle values and / or vibration values. Push environmental videos captured by ultra-black light cameras with the same timestamp and location coordinates.
[0021] Specifically, in step S5, the steps for the cloud to receive the vehicle video immediately before and after receiving the abnormal tilt angle value and / or vibration value include: Obtain the timestamp and location coordinates of the abnormal tilt angle and / or vibration values pushed; Based on the timestamps and location coordinates of abnormal tilt angle values and / or vibration values, find vehicle videos captured by ultra-black light cameras with the same and previous timestamps and location coordinates. The system pushes out vehicle videos captured by ultra-black light cameras with the same timestamp and location coordinates as the previous ones.
[0022] Specifically, in step S5, the specific steps for detecting abnormal tilt angle values and / or vibration values in the vehicle's current vehicle environment video using a pre-trained lightweight YOLOv5 model include: Decompose the vehicle environment video into continuous video frame images; Convert consecutive video frame images into grayscale images and then normalize them; The normalization expression is: In the formula: Indicates the frame image in coordinates The grayscale value at that location; This represents the average grayscale value of the frame image; This represents the standard deviation of grayscale values in a frame image. The normalized grayscale image is scaled to fit the input size of the pre-trained lightweight YOLOv5 model; A grayscale image adapted to the input size is input into a pre-trained lightweight YOLOv5 model. The pre-trained lightweight YOLOv5 model outputs a set of bounding boxes for the vehicle, each bounding box including coordinates and class confidence. A preset category confidence threshold is used to filter the category confidence of each bounding box, thus selecting valid vehicles. The expression is: .
[0023] Specifically, in step S5, the specific steps for determining the degree of congestion based on the previous abnormal tilt angle value and / or vibration value and the current traffic flow include: Preset traffic flow thresholds for each congestion level; Based on the preset traffic flow thresholds for each congestion level, the congestion level that matches the traffic flow before and after the abnormal tilt angle value and / or vibration value is identified, and this level is used as the vehicle congestion level before and after the abnormal tilt angle value and / or vibration value.
[0024] Specifically, in step S5, the specific steps for determining the correlation between the abnormal tilt angle value and / or vibration value and the current level of congestion include: Determine the magnitude of congestion before and after the abnormal tilt angle and / or vibration value. If the congestion level before the abnormal tilt angle and / or vibration value is greater than or equal to the current congestion level, then it is unrelated to the abnormal tilt angle and / or vibration value. If the congestion level before the abnormal tilt angle and / or vibration value is less than the current congestion level, then it is related to the abnormal tilt angle and / or vibration value.
[0025] The aforementioned tilt vibration integrated sensor and ultra-black light camera are powered by a combination of solar energy and batteries to reduce wiring costs.
[0026] The aforementioned tilt vibration integrated sensor and ultra-black light camera are adapted to the complex environment of high-risk sections of highways, support wide temperature range of -40 to +55℃, and have strong stability.
[0027] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.
Claims
1. A method for monitoring and capturing images of high-risk road sections based on the Internet of Things for intelligent transportation, characterized in that: Includes the following steps: S1: The tilt and vibration integrated sensor collects tilt and vibration data of structures in high-risk road sections in real time and transmits them to the cloud via the intelligent transportation Internet of Things, including tilt value, vibration value, timestamp and location coordinates; S2: The ultra-black light camera collects real-time traffic environment videos of high-risk road sections and transmits them to the cloud via the intelligent transportation Internet of Things, including environmental videos, vehicle videos, timestamps, and location coordinates; S3: Preset the tilt angle and vibration abnormality thresholds for structures in high-risk road sections in the cloud; S4: If the tilt angle and / or vibration value of a high-risk road section structure received by the cloud exceeds the preset tilt angle abnormality threshold and / or vibration abnormality threshold, an alarm will be triggered, and the abnormal tilt angle and / or vibration value will be pushed, along with an environmental video of the received abnormal tilt angle and / or vibration value. S5: The cloud receives vehicle videos of the moments preceding and / or following abnormal tilt angle values and vibration values, as well as the current vehicle environment. It uses a pre-trained lightweight YOLOv5 model to detect vehicles in the video of the moments preceding and following abnormal tilt angle values and vibration values. Based on the detected vehicles, it calculates the traffic flow of the moments preceding and following abnormal tilt angle values and vibration values, determines the degree of congestion based on the traffic flow of the moments preceding and following abnormal tilt angle values and vibration values, and determines the correlation between the congestion degree of the moments preceding and following abnormal tilt angle values and vibration values and the abnormal tilt angle values and vibration values. If there is a correlation, a serious alarm is issued.
2. The high-risk road section monitoring and capture method based on intelligent transportation IoT as described in claim 1, characterized in that: In step S4, the specific steps for pushing environmental video of abnormal tilt angle values and / or vibration values include: Obtain the timestamp and location coordinates of the abnormal tilt angle and / or vibration values pushed; Find environmental videos captured by ultra-black light cameras with the same timestamp and location coordinates based on the timestamps and location coordinates of abnormal tilt angle values and / or vibration values. Push environmental videos captured by ultra-black light cameras with the same timestamp and location coordinates.
3. The high-risk road section monitoring and capture method based on intelligent transportation Internet of Things as described in claim 2, characterized in that: In step S5, the specific steps for the cloud to receive the vehicle video before and after receiving the abnormal tilt angle value and / or vibration value include: Obtain the timestamp and location coordinates of the abnormal tilt angle and / or vibration values pushed; Based on the timestamps and location coordinates of abnormal tilt angle values and / or vibration values, find vehicle videos captured by ultra-black light cameras with the same and previous timestamps and location coordinates. The system pushes out vehicle videos captured by ultra-black light cameras with the same timestamp and location coordinates as the previous ones.
4. The high-risk road section monitoring and capture method based on intelligent transportation Internet of Things as described in claim 3, characterized in that: In step S5, the specific steps for detecting abnormal tilt angle values and / or vibration values in the vehicle's current vehicle environment video using a pre-trained lightweight YOLOv5 model include: Decompose the vehicle environment video into continuous video frame images; Convert consecutive video frame images into grayscale images and then normalize them; The normalized grayscale image is scaled to fit the input size of the pre-trained lightweight YOLOv5 model; A grayscale image adapted to the input size is input into a pre-trained lightweight YOLOv5 model. The pre-trained lightweight YOLOv5 model outputs a set of bounding boxes for the vehicle, each bounding box including coordinates and class confidence. A preset category confidence threshold is used to filter the category confidence of each bounding box, thereby selecting valid vehicles.
5. The high-risk road section monitoring and capture method based on intelligent transportation Internet of Things as described in claim 4, characterized in that: In step S5, the specific steps for determining the degree of congestion based on the previous abnormal tilt angle value and / or vibration value and the current traffic flow include: Preset traffic flow thresholds for each congestion level; Based on the preset traffic flow thresholds for each congestion level, the congestion level that matches the traffic flow before and after the abnormal tilt angle value and / or vibration value is identified, and this level is used as the vehicle congestion level before and after the abnormal tilt angle value and / or vibration value.
6. The high-risk road section monitoring and capture method based on intelligent transportation Internet of Things as described in claim 5, characterized in that: In step S5, the specific steps for determining the correlation between the abnormal tilt angle value and / or vibration value and the current level of congestion include: Determine the magnitude of congestion before and after the abnormal tilt angle and / or vibration value. If the congestion level before the abnormal tilt angle and / or vibration value is greater than or equal to the current congestion level, then it is unrelated to the abnormal tilt angle and / or vibration value. If the congestion level before the abnormal tilt angle and / or vibration value is less than the current congestion level, then it is related to the abnormal tilt angle and / or vibration value.