CN116818009A - Truck overrun detection system and method - Google Patents
Truck overrun detection system and method Download PDFInfo
- Publication number
- CN116818009A CN116818009A CN202310681027.0A CN202310681027A CN116818009A CN 116818009 A CN116818009 A CN 116818009A CN 202310681027 A CN202310681027 A CN 202310681027A CN 116818009 A CN116818009 A CN 116818009A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- truck
- detection
- license plate
- overrun
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 172
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000005303 weighing Methods 0.000 claims abstract description 49
- 238000009826 distribution Methods 0.000 claims abstract description 20
- 238000004140 cleaning Methods 0.000 claims abstract description 15
- 238000012544 monitoring process Methods 0.000 claims description 62
- 230000008569 process Effects 0.000 claims description 11
- 230000001502 supplementing effect Effects 0.000 claims description 11
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 8
- 239000010453 quartz Substances 0.000 claims description 7
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N silicon dioxide Inorganic materials O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 claims description 7
- 230000006399 behavior Effects 0.000 claims description 6
- 238000005520 cutting process Methods 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 5
- 230000002159 abnormal effect Effects 0.000 claims description 4
- 230000001960 triggered effect Effects 0.000 claims description 4
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 238000010801 machine learning Methods 0.000 claims description 3
- 238000012706 support-vector machine Methods 0.000 claims description 3
- 230000026676 system process Effects 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 7
- 238000004364 calculation method Methods 0.000 description 5
- 230000001133 acceleration Effects 0.000 description 4
- 238000013527 convolutional neural network Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000005484 gravity Effects 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000011272 standard treatment Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000000725 suspension Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/625—License plates
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01G—WEIGHING
- G01G19/00—Weighing apparatus or methods adapted for special purposes not provided for in the preceding groups
- G01G19/02—Weighing apparatus or methods adapted for special purposes not provided for in the preceding groups for weighing wheeled or rolling bodies, e.g. vehicles
- G01G19/03—Weighing apparatus or methods adapted for special purposes not provided for in the preceding groups for weighing wheeled or rolling bodies, e.g. vehicles for weighing during motion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
- G06V20/54—Surveillance or monitoring of activities, e.g. for recognising suspicious objects of traffic, e.g. cars on the road, trains or boats
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/017—Detecting movement of traffic to be counted or controlled identifying vehicles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/08—Detecting or categorising vehicles
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a truck overrun detection system and method, comprising a license plate recognition device, a dynamic weighing device, a vehicle length detection device, an axle detection camera cleaning device, an information display device, an early warning device, a broadcasting system and a control center. After the vehicle passes through the system, the control center can obtain a series of passing information, including the passing time of a truck, the carrying capacity, license plates, the height, the length, the axle number, the axle type, the axle distribution and the like of the truck, if the truck overruns, the control center displays the license plate number of the overrun vehicle and the information exceeding the national limit on an information display screen and gives out a warning to the overrun vehicle, and meanwhile, the control center uploads the information of the vehicle to a database so as to facilitate the later inquiry of national law enforcement personnel, the condition of efficiently and automatically detecting the overrun of the truck throughout the day can be realized, and the vehicle blockage is avoided.
Description
Technical Field
The invention relates to the technical field of truck overrun detection, in particular to a truck overrun detection system and method.
Background
In recent years, with the rapid development of economy, the construction scale of highways in China is larger and larger, and the traffic volume of highways is increased year by year, however, because the overtravel overload of trucks caused by freely refitting vehicles is a common phenomenon for the purpose of pursuing profits by transportation owners, the safety of people and the property safety of countries are seriously endangered, the overtravel transportation of trucks is the key point of the current road treatment, the service life of roads and bridges is shortened due to the overtravel transportation, the road surfaces of highways are damaged, traffic accidents are easily caused, the transportation efficiency of highways is reduced, the cost of maintenance road surfaces of traffic departments is huge, and the construction difficulty is high, so that the overtravel overweight phenomenon of trucks is imperative to be treated.
At present, the high-speed overrun control method for trucks mainly comprises the steps of firstly entering an overrun detection station before the trucks enter a highway toll station, carrying out vehicle detection by matching with national law enforcement personnel, and having the advantages of needing a large amount of manpower, low efficiency, extremely easily causing vehicle blockage due to long detection time and being incapable of ensuring the full-day overrun control due to manual participation.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a truck overrun detection system and method which can efficiently monitor the condition of truck overrun overload.
In order to achieve the purpose of efficiently supervising the excessive overload of the truck, the invention provides the following technical scheme: a truck overrun detection system and method comprises a license plate recognition device, a dynamic weighing device, a vehicle length detection device, an axle detection camera cleaning device, an information display device, an early warning device, a broadcasting system and a control center.
The license plate recognition device is used for recognizing front and rear license plates of vehicles and comprises a vehicle head license plate recognition camera, a license plate display screen and a vehicle tail license plate recognition camera.
The dynamic weighing device is used for detecting the weight of the vehicle, and a quartz type dynamic weighing sensor is arranged inside the dynamic weighing device.
The vehicle length detection device is used for detecting the length and the height of a vehicle and comprises an inverted-L-shaped monitoring rod, a laser sensor, an I-shaped monitoring rod, a laser sensor, an inverted-L-shaped monitoring rod and a laser sensor.
The axle detection device is used for detecting the axle type and distribution of a vehicle and comprises an axle detection camera, a photoelectric sensor and an I-type monitoring rod.
And a vehicle information display screen is arranged at the rearmost part of the detection system and is used for displaying the detection result of the basic information of the detected vehicle.
The axle detection camera cleaning device is used for cleaning a camera used for axle detection and comprises a small wiper, an axle camera protection device and a small water gun.
The control center is used for storing the PP-YOLOE model for detecting the types and the distribution of the axles and various data of detected vehicles, comparing the data with the national standard, the power module provides voltage for each device, and the information display device is used for displaying the information of the current vehicle, the early warning device and the broadcasting system are used for determining the warning after the vehicle is out of limit. The national standard of overrun vehicle identification is stored in a list as a key according to the principle that the main shaft is 1 and the slave shaft is 2, and the total mass limit is stored in a database as a value, so that the comparison with the data uploaded by the weighing device is facilitated.
The vehicle license plate recognition camera and the vehicle tail license plate recognition camera are respectively installed at the front and the rear of the license plate display screen, the dynamic weighing device is arranged between the license plate recognition device and the inverted-L-shaped monitoring rod, a laser sensor installed on the inverted-L-shaped monitoring rod is used for detecting the height of a vehicle, the laser sensor and the laser sensor are used for detecting the length of the vehicle, a photoelectric sensor installed on the I-shaped monitoring rod is used for detecting the number of wheels, and the axle detection camera is arranged between the inverted-L-shaped monitoring rod and the I-shaped monitoring rod.
Preferably, the output ends of the license plate recognition device, the vehicle length detection device, the dynamic weighing device and the axle detection device are connected with the control center, the output end of the control center is connected with the information display device, the early warning device and the broadcasting system, the axle detection camera cleaning device is connected with the axle detection device, and the control center judges through receiving the data uploaded by the license plate recognition device, the vehicle length detection device, the dynamic weighing device and the axle detection device and the national vehicle overrun regulation standard.
When a truck enters a high-speed intersection, the truck needs to carry out overrun detection before the high-speed intersection, the truck passes through a license plate recognition device, a vehicle length detection device, a dynamic weighing device and an axle detection device detection area at a low speed under the condition of no parking, and the system can automatically detect the effective truck data including the truck passing time, the truck carrying capacity, the license plate, the height, the length, the number of axles, the axle types and the axle distribution of the truck vehicles and upload the truck vehicle detection data to a control center;
when a vehicle runs into the overrun detection area and enters the license plate recognition device, the device can detect and recognize front and rear license plates of a truck and upload detected data to a control center;
when a vehicle runs into the overrun detection area and enters the dynamic weighing device, the truck passes through the sensor at a low speed and at a uniform speed, and after the truck passes through the sensor, the instrument obtains weighing information and uploads the weighing information to the control center;
when the vehicle moves into the overrun detection area and enters the vehicle length and height detection device, the device can detect the length and the height of the vehicle and upload detection data to the control center;
when the vehicle runs into the overrun detection area and enters the axle detection device, the device can detect the number of the axles of the truck, the types of the axles and the distribution positions of different axles and upload the detected data to the control center;
the control center judges whether the vehicle is overrun or not through the data uploaded by different detection devices and the vehicle overrun standard specified by the country, and when the vehicle overruns, the control center can send the length, height and weight information of the vehicle to the information display device, start the early warning device and the broadcasting system, remind a driver to accept punishment of law enforcement personnel and timely process overrun behaviors of the vehicle.
Preferably, the license plate recognition device is provided with a light supplementing lamp, and when a vehicle enters the overrun detection area, the license plate recognition camera of the vehicle head and the license plate recognition camera of the vehicle tail acquire license plate images when the vehicle runs. The image is subjected to preprocessing operations including image graying and the like. Converting the color image into gray image can reduce image noise and increase image processing speed, and is convenient for obtaining license plate number in real time. And (3) positioning the license plate, and determining the position of the license plate in the image according to the specific features of the color, the shape, the proportion and the like of the license plate and combining with the trained CNN model. And then, character cutting is carried out on the license plate through a CNN model, and each character is extracted independently. Accurate character cutting can ensure the accuracy of subsequent character recognition. And finally, performing character recognition by using a machine learning algorithm such as a support vector machine, a neural network and the like, and transmitting the recognized license plate information to a license plate display screen.
Preferably, the load cell is mounted on the ground or on a suspension above the ground, ensuring that the sensor is in contact with the vehicle and that the sensor is stable against movement. The sensor is connected with the data acquisition unit, so that the connection stability is ensured and the data acquisition unit works normally. The vehicle passes over the sensor at a constant speed and is allowed to pass through the sensor. The sensor will measure the weight of the vehicle as it passes and transmit data to the data collector. The weight calculation formula: weight = force measured by the sensor/gravitational acceleration. The force measured by the sensor refers to the force generated when the vehicle passes through the sensor, and the gravity acceleration is generally 9.8 m/s.
Preferably, the length of the transverse rod of the inverted-L-shaped monitoring rod and the length of the transverse rod of the inverted-L-shaped monitoring rod are 3m, the length of the vertical rod is 4m, and the distance between the two monitoring rods is s=10m. And 3 laser sensors arranged at the position of 1.5m of the inverted L-shaped monitoring rod cross rod are used for detecting the height of the vehicle. The height of the vehicle can be calculated by subtracting the distance returned by the laser sensor from the height of the vertical rod of the inverted L-shaped monitoring rod, the calculated height of the vehicle is recorded for a plurality of times, and the maximum value is taken as the final height of the vehicle. At the same time, the time t1 when the vehicle passes the sensor is recorded for the calculation of the subsequent vehicle length. The laser sensor arranged at the position of the inverted L-shaped monitoring rod cross rod 1.5m is used for detecting the passing time t2 of the vehicle. The speed v of the vehicle is calculated from s, t1 and t2. When the return distance of the 2 laser sensors arranged at the position, 2m away from the ground, of the vertical rod of the I-type monitoring rod is smaller than 50cm, the vehicle entering detection device is shown. The current time t3 when the vehicle passes the sensor is recorded, and when the distance returned by the laser sensor is greater than 50cm, the vehicle is indicated to pass the detection device. The time t4 at which the current vehicle passes the sensor is recorded. The length of the vehicle can be calculated according to t3, t4 and v.
Preferably, the axle detecting device is provided with a light supplementing lamp, the photoelectric sensor is arranged on the inner side of the I-shaped monitoring rod and used for detecting the number of wheels of the vehicle, the number of wheels is equal to the number of axles, when the fact that the vehicle enters an overrun detecting area is detected, an axle detecting camera arranged between the inverted L-shaped monitoring rod and the I-shaped monitoring rod is lifted, the light supplementing lamp is turned on, then the axle detecting camera photographs and records the bottom of the vehicle, and meanwhile, a photo is transmitted to a control center to be compared with a deployed PP-YOLOE model in real time for detection, and when the vehicle passes through the vehicle completely, the camera falls back to the protecting device again.
Preferably, the deployed PP-YOLOE model firstly pre-processes the image or video uploaded by the camera to reduce the influence of target shadows on detection. The deployed model divides the entire image into grids, then each grid predicts a different number and class of objects, and outputs its upper left corner coordinates, height, width, and confidence score for that class. Thereby obtaining the type of axle in the image or video. According to the sequence of the input images or videos, the identified axle type is stored in a list according to the principle that the main axle is 1 and the main axle is 2, and the axle type is compared with data stored in a database, so that the maximum limit value of the weight of the current truck is obtained. And the method is used for judging whether the truck is overweight or not.
Preferably, the camera shielding detection method is to detect a camera picture, when abnormal situations of a video shielding, a black screen and the like in a certain range occur, the camera shielding early warning is triggered, and the small wiper and the small water gun are matched to clean the camera.
Compared with the prior art, the invention provides a truck overrun detection system and method, through the cooperation arrangement of a license plate recognition device, a dynamic weighing device, a vehicle length detection device, an axle detection camera cleaning device, an information display device, an early warning device, a broadcasting system and a control center, after a vehicle passes through the system, the control center can obtain a series of truck passing information, including truck passing time, carrying capacity, license plates, height, length, axle number, axle type, axle distribution and the like of the truck, if the truck overrun is detected, the control center displays license plate number of the overrun vehicle and information exceeding national limits on an information display screen, and gives a warning to the overrun vehicle, and meanwhile, the control center uploads the information of the vehicle to a database, so that the later inquiry of national law enforcement personnel can be conveniently and efficiently and automatically detected, and the condition of the truck overrun is avoided.
Drawings
FIG. 1 is a schematic diagram of the technical solution of the present invention;
FIG. 2 is a flow chart of the technical solution of the present invention;
FIG. 3 is a schematic diagram of an overrun detecting system according to the present invention;
FIG. 4 is a schematic diagram of a subsystem vehicle length detection device;
FIG. 5 is a schematic diagram of a model step;
FIG. 6 is a diagram of a PP-YOLOE model architecture;
FIG. 7 is a block diagram of an axle inspection camera protection device and an axle inspection camera cleaning device;
FIG. 8 is a diagram of comparison of test data with database data.
In the figure: 1. vehicle license plate recognition camera; 2. license plate display screen; 3. the vehicle tail license plate identifies the camera; 4. a quartz type dynamic weighing sensor; 5. an inverted L-shaped monitoring rod; 6. a laser sensor; 7. i-type monitoring rod; 8. a laser sensor; 9. an inverted L-shaped monitoring rod; 10. a laser sensor; 11. an axle detection camera; 12. a photoelectric sensor; 13. i-type monitoring rod; 14. a vehicle information display screen; 15. a small wiper; 16. an axle camera protection device; 17. and a small water gun.
Description of the embodiments
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1-7, a truck overrun detection system and method includes a license plate recognition device, a dynamic weighing device, a vehicle length detection device, an axle detection camera cleaning device, an information display device, an early warning device, a broadcasting system and a control center, wherein the output ends of the license plate recognition device, the vehicle length detection device, the dynamic weighing device and the axle detection device are connected with the control center, the output end of the control center is connected with the information display device, the early warning device and the broadcasting system, the axle detection camera cleaning device is connected with the axle detection device, the control center judges the license plate recognition device, the vehicle length detection device, the dynamic weighing device and the data uploaded by the axle detection device with national vehicle overrun regulation standard through receiving the license plate recognition device, the vehicle length detection device and the data uploaded by the axle detection device, and the license plate recognition device are used for recognizing front and rear license plates of a vehicle, the license plate recognition device comprises a license plate recognition camera 1, a license plate display screen 2 and a license plate recognition camera 3, and the license plate recognition device is provided with light supplementing lamps for supplementing light when the vehicle drives into the overrun detection area, and the license plate recognition camera 1 and the license plate 3 obtains images of the vehicle when the vehicle drives. The image is subjected to preprocessing operations including image graying and the like. Converting the color image into gray image can reduce image noise and increase image processing speed, and is convenient for obtaining license plate number in real time. And (3) positioning the license plate, and determining the position of the license plate in the image according to the specific features of the color, the shape, the proportion and the like of the license plate and combining with the trained CNN model. And then, character cutting is carried out on the license plate through a CNN model, and each character is extracted independently. Accurate character cutting can ensure the accuracy of subsequent character recognition. Finally, a machine learning algorithm, such as a support vector machine, a neural network and the like, is used for carrying out character recognition and sending the recognized license plate information to a license plate display screen, a dynamic weighing device is used for detecting the weight of a vehicle, a quartz dynamic weighing sensor 4 is arranged in the dynamic weighing device, the weighing sensor is arranged on the ground or a hanging device above the ground, the contact between the sensor and the vehicle is ensured, and the sensor is stable and does not move. The sensor is connected with the data acquisition unit, so that the connection stability is ensured and the data acquisition unit works normally. The vehicle passes over the sensor at a constant speed and is allowed to pass through the sensor. The sensor will measure the weight of the vehicle as it passes and transmit data to the data collector. The weight calculation formula: weight = force measured by the sensor/gravitational acceleration. The force measured by the sensor is the force generated when the vehicle passes through the sensor, the gravity acceleration generally takes 9.8m/s, and the vehicle length and height detection device is used for detecting the length and the height of the vehicle and comprises an inverted L-shaped monitoring rod (5), a laser sensor (6), an I-shaped monitoring rod (7), a laser sensor (8), an inverted L-shaped monitoring rod (9) and a laser sensor (10). The length of the transverse rod of the inverted-L-shaped monitoring rod and the length of the transverse rod of the inverted-L-shaped monitoring rod are 3m, the length of the vertical rod is 4m, and the distance between the two monitoring rods is s=10m. And 3 laser sensors arranged at the position of 1.5m of the inverted L-shaped monitoring rod cross rod are used for detecting the height of the vehicle. The height of the vehicle can be calculated by subtracting the distance returned by the laser sensor from the height of the vertical rod of the inverted L-shaped monitoring rod, the calculated height of the vehicle is recorded for a plurality of times, and the maximum value is taken as the final height of the vehicle. At the same time, the time t1 when the vehicle passes the sensor is recorded for the calculation of the subsequent vehicle length. The laser sensor arranged at the position of the inverted L-shaped monitoring rod cross rod 1.5m is used for detecting the passing time t2 of the vehicle. The speed v of the vehicle is calculated from s, t1 and t2. When the return distance of the 2 laser sensors arranged at the position, 2m away from the ground, of the vertical rod of the I-type monitoring rod is smaller than 50cm, the vehicle entering detection device is shown. The current time t3 when the vehicle passes the sensor is recorded, and when the distance returned by the laser sensor is greater than 50cm, the vehicle is indicated to pass the detection device. The time t4 at which the current vehicle passes the sensor is recorded. The length of the vehicle can be calculated according to t3, t4 and v, the axle detection device is used for detecting the axle type and distribution of the vehicle and comprises an axle detection camera (11), a photoelectric sensor (12) and an I-type monitoring rod (13), a vehicle information display screen 14 is arranged at the rearmost part of the detection system, the vehicle information display screen 14 is used for displaying the detected result of basic information of the detected vehicle, the axle detection camera cleaning device is used for cleaning a camera used for axle detection and comprises a small wiper 15, an axle camera protection device 16 and a small water gun 7, the camera shielding detection method is used for detecting a camera picture, when abnormal situations of a video shielding, a black screen and the like occur in a certain range, the camera shielding early warning is triggered, the small wiper 15 is matched with the small water gun 17 to clean the camera, the control center is used for storing the PP-YOLOE model for detecting the types and the distribution of the axles and detecting various data of the vehicles, comparing the PP-YOLOE model with the national standard, the power module provides voltage for each device, the information display device is used for displaying the information of the current vehicle, the early warning device and the broadcasting system are used for determining the warning after the vehicles overrun, the vehicle head license plate recognition cameras (1) and the vehicle tail license plate recognition cameras (3) are respectively arranged in front of and behind the license plate display screen (2), the dynamic weighing device is arranged between the license plate recognition devices and the inverted L-shaped monitoring rod (5), the laser sensor (6) arranged on the inverted L-shaped monitoring rod (5) is used for detecting the height of the vehicles, the laser sensor (8) and the laser sensor (10) are used for detecting the length of the vehicles, the photoelectric sensor (12) installed on the I-type monitoring rod (13) is used for detecting the number of wheels, the axle detection camera (11) is arranged between the inverted L-type monitoring rod (9) and the I-type monitoring rod (13), the axle detection device is provided with a light supplementing lamp for supplementing light, the photoelectric sensor 12 is arranged on the inner side of the I-type monitoring rod (13) and used for detecting the number of wheels of a vehicle, the number of wheels is equal to the number of axles, when the fact that the vehicle enters an overrun detection area is detected, the axle detection camera (11) arranged between the inverted L-type monitoring rod (9) and the I-type monitoring rod is lifted, the light supplementing lamp is opened, then the axle detection camera (8) photographs and records the bottom of the vehicle, and simultaneously transmits the photograph to a control center to carry out real-time comparison detection with a deployed PP-YOLOE model, after the vehicle completely passes through the vehicle, the camera is lowered back to the protection device again, the PP-YOLOE model carries out image denoising after the influence of target shadows on the detection by preprocessing, and finally carries out axle classification and positioning on the axle in the image, so that the distribution and the number of axles of the driven shafts and the driven shafts of the truck are obtained, and the steps are implemented as follows: when a truck enters a high-speed intersection, the truck is required to carry out overrun detection before the high-speed intersection, the truck passes through a license plate recognition device, a vehicle length detection device, a dynamic weighing device and an axle detection device detection area at a low speed under the condition that the truck is not stopped, the system can automatically detect the effective truck data including the truck passing time, the truck carrying capacity, the license plate, the height, the length, the number of axles, the axle types and the axle distribution of the truck vehicles, and upload the truck vehicle detection data to a control center; when a vehicle runs into the overrun detection area and enters the license plate recognition device, the device can detect and recognize front and rear license plates of a truck and upload detected data to a control center; when a vehicle runs into the overrun detection area and enters the dynamic weighing device, the truck passes through the sensor at a low speed and at a uniform speed, and after the truck passes through the sensor, the instrument obtains weighing information and uploads the weighing information to the control center; when the vehicle moves into the overrun detection area and enters the vehicle length and height detection device, the device can detect the length and the height of the vehicle and upload detection data to the control center; when the vehicle runs into the overrun detection area and enters the axle detection device, the device can detect the number of the axles of the truck, the types of the axles and the distribution positions of different axles and upload the detected data to the control center; the control center judges whether the vehicle is overrun or not through the data uploaded by different detection devices and the vehicle overrun standard specified by the country, and when the vehicle overruns, the control center can send the length, height and weight information of the vehicle to the information display device, start the early warning device and the broadcasting system, remind a driver to accept punishment of law enforcement personnel and timely process overrun behaviors of the vehicle.
In the use process, when a truck enters a high-speed intersection, the truck needs to carry out overrun detection in front of the high-speed intersection, the truck passes through a license plate recognition device, a vehicle length detection device, a dynamic weighing device and an axle detection device detection area at a low speed under the condition of no parking, the system can automatically detect the effective truck data including the truck passing time, the truck carrying capacity, the license plate and the height, the length, the axle number, the axle type and the axle distribution of the truck, and upload the truck vehicle detection data to a control center, wherein when the truck passes through the overrun detection area and enters the license plate recognition device, the device can detect and recognize the front license plate and the rear license plate of the truck, after detection, the truck enters the dynamic weighing detection device, when the truck passes through the overrun detection area and enters the dynamic weighing device, the truck passes through a dynamic weighing plate at a low speed, after passing through the truck, the instrument can acquire the weight data of the truck and upload the weight data to the control center, when the truck runs into an overrun detection area and enters a length and height detection device of the truck, the device can detect the length and the height of the truck and upload the data to the control center, when the truck runs into the overrun detection area and enters an axle detection device, the device can detect the number of axles of the truck, the types of the axles and the distribution positions of different axles and upload the detected data to the control center, when the control center judges whether the truck is overrun or not through the data uploaded by the different detection devices and the overrun standard of the truck specified by the country, when the truck is overrun, the control center can send the length, the height and the weight information of the truck to the information display device, the early warning device and the broadcasting system are started, the driver is reminded of accepting punishment of law enforcement personnel and timely processing the overrun behavior of the truck, the control center forms complete passing information by using information such as passing time, loading capacity, license plate and height, length, axle number, axle type, axle distribution and the like of the trucks detected by different detection devices, and uploads the passing information to a database of the control center, law enforcement personnel can check whether the vehicles have overrun behaviors or not in the past through the database, wherein the model detection process is as follows: firstly, the model carries out pretreatment normalization on a photo shot by a camera, including graying, geometric transformation and image enhancement, and simultaneously carries out a series of standard treatment transformation on the image to convert the image into a fixed standard form, so that the image is adjusted to be uniform in size, and the image is convenient to process. The main purpose of image preprocessing is to eliminate irrelevant information in the image, recover useful real information, enhance the detectability of relevant information, simplify data to the maximum extent, and thereby improve the reliability of feature extraction, image segmentation, matching and recognition. Then, the image is denoised and input into a model PP-YOLOE deployed by a server, a step structure consisting of 3 stacked convolutions and 4 CSPRepResstage are formed through a backbond, cross-scale feature extraction is carried out through the CSPRepResnet network of the backbond part, then the extracted features are input into a piece consisting of 5 CSPRepResstage, and after FPN and PAN structure, the extracted features are further fused, so that the detection effect of the model is improved. The FPN network is from top to bottom, the high-level features are fused with the bottom-level features through upsampling to obtain a predicted feature map, and the PAN structure transmits the strong positioning information of the bottom level through downsampling. The ET-head of the head portion incorporates ESE Block (Effective Squeeze-and-specification Block), DFL and VFL for improving accuracy and speed of detection, and simplifies the classification task alignment module to shortcut, further improving speed, inputting features extracted by the backbox and the negk into the head portion for classification and localization. After model detection, the number and distribution positions of detected driving shafts and driven shafts are obtained, compared with national regulation data stored in a database, and whether the limit is exceeded or not is judged, and in a deep learning model, mAP, precision, accuracy and Recall are generally used as measurement standards of model performance. The accuracy represents the ratio of the prediction pairs in all samples predicted as positive samples, and the accuracy formula is:
TP: true Positive, the classifier predicts that the result is a Positive sample, and actually is also a Positive sample, i.e., the number of Positive samples that are correctly identified. FP: false Positive, the classifier prediction result is a Positive sample, the actual negative sample is the number of False negative samples, the Recall rate indicates the accuracy of the accurate result predicted by detection accounting for the total Positive sample, and the Recall rate formula is:
FN: false positive, the classifier prediction result is a Negative sample, the actual positive sample is the number of positive samples of False detection, the Accuracy represents the proportion of the predicted correct sample to the total samples, and the Accuracy formula is:
TN: true Negative, the classifier prediction result is a Negative sample, and is actually a Negative sample, namely, the number of Negative samples correctly identified, and mAP (mean average precision ) is the most commonly used index in target detection, which is helpful for evaluating the effectiveness of the algorithm. The mAP calculation formula is:
wherein, P () represents the precision measured when the recall rate is equal to or greater than the maximum precision of r, when the vehicle passes through the quartz dynamic weighing sensor 4 at a low speed and uniform speed, the pressure applied by the wheels is converted into an electric signal and transmitted to a computer processing system, and the computer processing system processes, analyzes and calculates the data acquired by the sensor in real time and calculates the current weight of the truck according to the data. And simultaneously, the weighing information is uploaded to a control center. The output end of the dynamic weighing device is connected with a control center, the camera shielding detection method is that when abnormal conditions of video shielding, black screen and other lenses occur in a certain range, camera shielding early warning is triggered, when the existence of stains on the camera 11 is detected, a water gun 17 washes the camera, meanwhile, a wiper 15 cleans the camera, the influence of dust and rain and snow on detection results is prevented, photos shot by the camera are clear, the axle detection device can efficiently identify the types of axles and the distribution rules of different axles, the camera shielding detection method is that a camera picture is taken as a detection photo and is transmitted into the control center, firstly, images are binarized, blackish parts are used as foreground, other parts are used as background, communication areas are detected on the foreground, the method comprises the steps of obtaining the area of a maximum communication area, wherein the ratio of the area to the area of the whole image is a shielding rate, when the shielding rate is larger than 0, a water gun 17 and a wiper 15 work to clear a camera, a detection system further comprises a database system of a control center, the control center obtains data from a license plate recognition device, a vehicle length detection device, a dynamic weighing device and an axle detection device to form complete passing information, the complete passing information is uploaded to the database, the control center judges whether a vehicle is out of limit or not through the data uploaded by different detection devices and the vehicle out of limit standard specified by the country, and when the vehicle is out of limit, the control center sends the length, height and weight information of the vehicle to an information display device to start an early warning device and a broadcasting system to remind a driver to accept punishment of law enforcement personnel and timely process the out of limit behavior of the vehicle.
In summary, the system and method for detecting the overrun of the truck,
it is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
Although embodiments of the present 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 therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (9)
1. A truck overrun detection system comprises a license plate recognition device, a dynamic weighing device, a vehicle length detection device, an axle detection camera cleaning device, an information display device, an early warning device, a broadcasting system and a control center;
the license plate recognition device is used for recognizing front and rear license plates of vehicles and comprises a vehicle head license plate recognition camera (1), a license plate display screen (2) and a vehicle tail license plate recognition camera (3);
the dynamic weighing device is used for detecting the weight of the vehicle, and a quartz type dynamic weighing sensor (4) is arranged in the dynamic weighing device;
the vehicle length detection device is used for detecting the length and the height of a vehicle and comprises an inverted L-shaped monitoring rod (5), a laser sensor (6), an I-shaped monitoring rod (7), a laser sensor (8), an inverted L-shaped monitoring rod (9) and a laser sensor (10);
the axle detection device is used for detecting the axle type and distribution of a vehicle and comprises an axle detection camera (11), a photoelectric sensor (12) and an I-type monitoring rod (13);
the rearmost part of the detection system is provided with a vehicle information display screen (14), and the vehicle information display screen (14) is used for displaying the detection result of the basic information of the detected vehicle;
the axle detection camera cleaning device is used for cleaning a camera for axle detection and comprises a small wiper (15), an axle camera protection device (16) and a small water gun (17);
the control center is used for storing the PP-YOLOE model for detecting the types and the distribution of the axles and various data of detected vehicles, comparing the data with the national standard, the power module provides voltage for each device, and the information display device is used for displaying the information of the current vehicle, the early warning device and the broadcasting system are used for determining the warning after the vehicle is out of limit;
the vehicle head license plate recognition camera (1) and the vehicle tail license plate recognition camera (3) are respectively installed on the front and the rear of the license plate display screen (2), the dynamic weighing device is arranged between the license plate recognition device and the inverted-L-shaped monitoring rod (5), a laser sensor (6) installed on the inverted-L-shaped monitoring rod (5) is used for detecting the height of a vehicle, a laser sensor (8) and a laser sensor (10) are used for detecting the length of the vehicle, a photoelectric sensor (12) installed on the I-shaped monitoring rod (13) is used for detecting the number of wheels, and the axle detection camera (11) is arranged between the inverted-L-shaped monitoring rod (9) and the I-shaped monitoring rod (13).
2. The truck overrun detection system of claim 1, wherein: the vehicle license plate recognition device, the vehicle length detection device, the dynamic weighing device and the output end of the vehicle axle detection device are connected with the control center, the output end of the control center is connected with the information display device, the early warning device and the broadcasting system, the vehicle axle detection camera cleaning device is connected with the vehicle axle detection device, and the control center judges whether the vehicle license plate recognition device, the vehicle length detection device, the dynamic weighing device and the vehicle axle detection device upload data and the national vehicle overrun regulation standard.
3. The method for detecting the overrun of the truck is characterized by comprising the following steps of: the detection by the truck overrun detection system according to any one of claims 1-2 is implemented as follows: when a truck enters a high-speed intersection, the truck is required to carry out overrun detection before the high-speed intersection, the truck passes through a license plate recognition device, a vehicle length detection device, a dynamic weighing device and an axle detection device detection area at a low speed under the condition that the truck is not stopped, the system can automatically detect the effective truck data including the truck passing time, the truck carrying capacity, the license plate, the height, the length, the number of axles, the axle types and the axle distribution of the truck vehicles, and upload the truck vehicle detection data to a control center;
when a vehicle runs into the overrun detection area and enters the license plate recognition device, the device can detect and recognize front and rear license plates of a truck and upload detected data to a control center;
when a vehicle runs into the overrun detection area and enters the dynamic weighing device, the truck passes through the sensor at a low speed and at a uniform speed, and after the truck passes through the sensor, the instrument obtains weighing information and uploads the weighing information to the control center;
when the vehicle moves into the overrun detection area and enters the vehicle length and height detection device, the device can detect the length and the height of the vehicle and upload detection data to the control center;
when the vehicle runs into the overrun detection area and enters the axle detection device, the device can detect the number of the axles of the truck, the types of the axles and the distribution positions of different axles and upload the detected data to the control center;
the control center judges whether the vehicle is overrun or not through the data uploaded by different detection devices and the vehicle overrun standard specified by the country, and when the vehicle overruns, the control center can send the length, height and weight information of the vehicle to the information display device, start the early warning device and the broadcasting system, remind a driver to accept punishment of law enforcement personnel and timely process overrun behaviors of the vehicle.
4. The truck overrun detection system and method of claim 3, wherein: the license plate recognition device is provided with a light supplementing lamp, and when a vehicle drives into an overrun detection area, the license plate recognition camera (1) of the vehicle head and the license plate recognition camera (3) of the vehicle tail can acquire license plate images when the vehicle drives; positioning the license plate, and determining the position of the license plate in the image according to the specific features of the license plate, such as color, shape, proportion and the like and by combining image processing and a computer vision algorithm; then, carrying out character cutting on the license plate, and independently extracting each character; accurate character cutting can ensure the accuracy of the subsequent character recognition; and finally, performing character recognition by using a machine learning algorithm such as a support vector machine, a neural network and the like, transmitting the recognized license plate information to a license plate display screen, and uploading the license plate information to a control center.
5. The truck overrun detection system and method of claim 3, wherein: when the vehicle passes through the quartz dynamic weighing sensor (4) at a low speed and at a uniform speed, the quartz dynamic weighing sensor (4) converts the pressure born by the wheels into an electric signal and transmits the electric signal to the computer processing system, and the computer processing system processes, analyzes and calculates the data acquired by the sensor in real time and calculates the current weight of the truck according to the data; and simultaneously, the weighing information is uploaded to a control center.
6. The truck overrun detection system and method of claim 3, wherein: the length of a transverse rod of the inverted L-shaped monitoring rod (5) and the length of a transverse rod of the inverted L-shaped monitoring rod (9) are 3m, the length of a vertical rod of the inverted L-shaped monitoring rod is 4m, and the distance between the two monitoring rods is s=10m; 3 laser sensors (6) arranged at the position of the transverse rod 1.5m of the inverted L-shaped monitoring rod (5) are used for detecting the height of the vehicle; the height of the vehicle can be calculated by subtracting the returned distance of the laser sensor from the height of the vertical rod of the inverted L-shaped monitoring rod (5), the calculated height of the vehicle is recorded for a plurality of times, and the maximum value is taken as the final height of the vehicle; simultaneously recording the time t1 of the vehicle passing through the sensor for calculating the length of the subsequent vehicle; the laser sensor (10) is arranged at the position of the transverse rod 1.5m of the inverted L-shaped monitoring rod (9) and is used for detecting the passing time t2 of the vehicle; calculating the speed v of the vehicle according to s, t1 and t2; when the return distance of the 2 laser sensors (8) arranged at the position, 2m away from the ground, of the vertical rod of the I-type monitoring rod (7) is less than 50cm, a vehicle entering detection device is indicated; recording the time t3 when the current vehicle passes through the sensor, and indicating that the vehicle passes through the detection device after the distance returned by the laser sensor is more than 50 cm; recording the time t4 when the current vehicle passes through the sensor; the length of the vehicle can be calculated according to t3, t4 and v; and upload the data to the control center.
7. The truck overrun detection system and method of claim 3, wherein: the axle detection device is provided with a light supplementing lamp, a photoelectric sensor (12) is arranged on the inner side of an I-type monitoring rod (13) and used for detecting the number of wheels of a vehicle, the number of the wheels is equal to the number of axles, when the fact that the vehicle enters an overrun detection area is detected, an axle detection camera (11) arranged between an inverted L-type monitoring rod (9) and the I-type monitoring rod (13) is lifted, the light supplementing lamp is turned on, then the axle detection camera (11) photographs and records the bottom of the vehicle, and meanwhile, the photograph is transmitted to a control center to be compared with a deployed PP-YOLOE model in real time, and after the vehicle passes through the vehicle completely, the camera falls back to the protection device again.
8. The truck overrun detection system and method of claim 3, wherein: the deployed PP-YOLOE model firstly reduces the influence of target shadows on detection after preprocessing an image or video uploaded by a camera; dividing the whole image into grids by the deployed model, predicting objects of different numbers and categories by each grid, and outputting the coordinates, the height and the width of the upper left corner of the object and the confidence score of the category; thereby obtaining the type of axle in the image or video; according to the order of the incoming images or video.
9. The truck overrun detection system and method of claim 3, wherein: the camera shielding detection method is used for detecting a camera picture, when abnormal situations of a video shielding, a black screen and the like occur in a certain range, the camera shielding early warning is triggered, and the small wiper (15) is matched with the small water gun (17) to clean the camera.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310681027.0A CN116818009A (en) | 2023-06-09 | 2023-06-09 | Truck overrun detection system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310681027.0A CN116818009A (en) | 2023-06-09 | 2023-06-09 | Truck overrun detection system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116818009A true CN116818009A (en) | 2023-09-29 |
Family
ID=88126761
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310681027.0A Pending CN116818009A (en) | 2023-06-09 | 2023-06-09 | Truck overrun detection system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116818009A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117351439A (en) * | 2023-12-06 | 2024-01-05 | 山东博安智能科技股份有限公司 | Dynamic monitoring management system for intelligent expressway overrun vehicle |
| CN117389256A (en) * | 2023-12-11 | 2024-01-12 | 青岛盈智科技有限公司 | Early warning method for truck vehicle state in transportation process |
-
2023
- 2023-06-09 CN CN202310681027.0A patent/CN116818009A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117351439A (en) * | 2023-12-06 | 2024-01-05 | 山东博安智能科技股份有限公司 | Dynamic monitoring management system for intelligent expressway overrun vehicle |
| CN117351439B (en) * | 2023-12-06 | 2024-02-20 | 山东博安智能科技股份有限公司 | Dynamic monitoring management system for intelligent expressway overrun vehicle |
| CN117389256A (en) * | 2023-12-11 | 2024-01-12 | 青岛盈智科技有限公司 | Early warning method for truck vehicle state in transportation process |
| CN117389256B (en) * | 2023-12-11 | 2024-03-08 | 青岛盈智科技有限公司 | Early warning method for truck vehicle state in transportation process |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106652468B (en) | The detection and from vehicle violation early warning alarm set and method in violation of rules and regulations of road vehicle front truck | |
| CN116818009A (en) | Truck overrun detection system and method | |
| US20220196459A1 (en) | Real-time vehicle overload detection method based on convolutional neural network | |
| CN105809679B (en) | Mountain railway side slope rockfall detection method based on visual analysis | |
| CN103714363B (en) | A kind of motor vehicle exhaust smoke video identification system | |
| CN102737247B (en) | Identification system of smoke intensity image of tail gas of diesel vehicle | |
| KR100890625B1 (en) | High-speed Weight In Motion | |
| CN102759347B (en) | Online in-process quality control device and method for high-speed rail contact networks and composed high-speed rail contact network detection system thereof | |
| CN111783638B (en) | System and method for detecting number of wheel axles of vehicle and identifying vehicle type | |
| CN103465857A (en) | Mobile-phone-based active safety early-warning method for automobile | |
| CN106769732A (en) | A kind of rectilinear diesel vehicle smoke intensity detection method | |
| CN115205796B (en) | Rail line foreign matter intrusion monitoring and risk early warning method and system | |
| CN114399744A (en) | Vehicle type recognition method and device, electronic equipment and storage medium | |
| CN115527364B (en) | Traffic accident tracing method and system based on radar data fusion | |
| CN111783700B (en) | Automatic recognition and early warning method and system for pavement foreign matters | |
| CN111667655A (en) | Infrared image-based high-speed railway safety area intrusion alarm device and method | |
| CN103528531A (en) | Intelligent Internet of Things image detection system for small vehicle parameters | |
| CN115600124A (en) | Subway tunnel inspection system and inspection method | |
| CN113034378A (en) | Method for distinguishing electric automobile from fuel automobile | |
| CN113903180B (en) | Method and system for detecting vehicle overspeed on expressway | |
| CN111627221A (en) | Automatic comprehensive detection system based on no-parking dynamic intelligent weighing | |
| Yao et al. | Developing operating mode distribution inputs for MOVES with a computer vision–based vehicle data collector | |
| CN114627286A (en) | Method for detecting wagon staff invasion based on PSPNet and improved YOLOv4 | |
| WO2002052523A1 (en) | Method and apparatus for monitoring vehicle | |
| CN116631187B (en) | Intelligent acquisition and analysis system for case on-site investigation information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |