CN115512321A - Vehicle weight limit information identification method, computing device and storage medium - Google Patents
Vehicle weight limit information identification method, computing device and storage medium Download PDFInfo
- Publication number
- CN115512321A CN115512321A CN202211183756.5A CN202211183756A CN115512321A CN 115512321 A CN115512321 A CN 115512321A CN 202211183756 A CN202211183756 A CN 202211183756A CN 115512321 A CN115512321 A CN 115512321A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- wheel
- image
- information
- camera
- 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
Images
Classifications
-
- 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
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
-
- 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
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Software Systems (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a vehicle weight limit information identification method, which comprises the following steps: acquiring a vehicle side image and a vehicle chassis image acquired by an image acquisition device for a running vehicle; detecting the wheel hub in the side image of the vehicle based on the trained deep learning wheel hub detection model to obtain the number of vehicle axles and wheel information; detecting whether a differential exists in the vehicle chassis image based on the trained deep learning chassis detection model to obtain differential marking information; determining the axle type of the vehicle according to the number of the axles of the vehicle and the wheel information; determining the position and number of the driving shafts according to the axle type or differential marking information of the vehicle; the weight limit information of the vehicle is determined according to an axle type standard table based on the axle type of the vehicle or the position and number of the drive axles. The scheme increases the identification of the driving shaft so as to provide reliable vehicle type and shaft type information, can improve the accuracy of identifying the weight limit information of the vehicle, and strengthens the effective supervision of the overweight and the overrun of the vehicle.
Description
Technical Field
The invention relates to the technical field of vehicle monitoring, in particular to a vehicle weight limit information identification method, a computing device and a storage medium.
Background
In the field of intelligent transportation, in order to carry out overrun overload control on freight vehicles, dynamic weighing and axle type axle number detection are required to be carried out on the vehicles in the advancing process, and whether the vehicles overrun and overload is judged according to the axle number and axle type of the vehicles and corresponding weight limit standards.
In the prior art, a camera is generally adopted to collect a vehicle image or a laser scanning device is used to collect vehicle contour data to identify a vehicle type and an axle type, for example, a Chinese invention patent document with the patent number of CN112883943A discloses an axle type identification method and a system, wherein the quantity of axles of a current vehicle in the vehicle side image is determined by obtaining the vehicle side image and according to a pre-established depth self-learning model; determining the relative distance of the axle wheels according to the ratio of the side image of the vehicle to the actual vehicle; and determining the current vehicle axle type and vehicle type from a pre-established vehicle axle type characteristic list according to the relative distance between the axles and the number of the axles. In the scheme, the number and the type of the vehicle axles are determined by comparing the characteristics of the axle wheels with the model and combining the proportion of the actual vehicle, the number of the driving axles cannot be identified, the single-axle type and the double-axle type of the driving axles cannot be distinguished, and the identification accuracy is low. The chinese utility model patent of patent number "CN207675122U" discloses "a laser motorcycle type, axle type recognizer", through the system of constituteing by components such as detection device, control box, control panel, base, laser scanner, carry out comprehensive laser scanning to the vehicle of process, compensate the correction to the detected data, and send the detected data after correcting the compensation for the treater, the information that the treater obtained motorcycle type and axle type is analyzed to the detected data to vehicle data in the database according to the treater. The scheme has the disadvantages of complex system structure, high installation and debugging maintenance cost and incapability of determining the number of the driving shafts. The Chinese invention patent with the patent number of CN111325146A discloses a method for identifying the type and the axle type of a truck, which comprises the steps of collecting images or videos of the side surface and the front surface of the truck; after the position of the positioned vehicle is detected, inputting the position into a trained deep learning model to position and identify the wagon axle; calculating the number of axles, the wheelbase and the number of tires of the vehicle according to the number and the position of the axles of the vehicle in the side image of the vehicle detected by the model; positioning the tire position of the vehicle by using a convolutional neural network training model; according to the method, the front images of the vehicles are subjected to vehicle type identification and classification according to the discrimination model, and driving shafts cannot be identified.
Therefore, there is a need for a method for identifying vehicle weight limit information, which can not only improve the accuracy of identifying the number of axles, the type of axles, and the type of vehicle, but also detect and identify the number and positions of the driving axles to provide reliable vehicle weight limit information, so as to solve the above problems in the prior art.
Disclosure of Invention
In view of the above problems, a vehicle weight limit information identification method is provided, which detects the hub image and the chassis image respectively through a deep learning algorithm, can accurately identify the number of axles, the axle type and the driving axle of the vehicle running in the lane, and solves the problems of low axle type identification accuracy and incapability of identifying the driving axle in the prior art.
According to one aspect of the invention, a vehicle weight limit information identification method is provided, which comprises the following steps: acquiring a vehicle side image and a vehicle chassis image acquired by an image acquisition device for a running vehicle; detecting the wheel hubs in the side images of the vehicle based on the trained deep learning wheel hub detection model to obtain the number of vehicle axles and wheel information; detecting whether a differential exists in the vehicle chassis image based on the trained deep learning chassis detection model to obtain differential marking information; determining the axle type of the vehicle according to the number of axles of the vehicle and the wheel information; determining the position and number of the driving shafts according to the axle type or differential marking information of the vehicle; the weight limit information of the vehicle is determined according to an axle type standard table based on the axle type of the vehicle or the position and number of the drive axles.
By adopting the scheme, the vehicle side image containing the hub information and the vehicle chassis image containing the chassis information are respectively obtained, and the vehicle side image and the vehicle chassis image are respectively identified through a deep learning algorithm, so that the number of the vehicle axles, the axle type and the positions and the number of the driving axles are obtained, and the accuracy of identifying the vehicle weight limit information can be improved.
Optionally, in the above method, the image capturing device is vertically installed at an entrance of a vehicle driving lane, the image capturing device includes at least a first camera and a second camera, the image capturing device is a first predetermined distance from a near end of the lane and a second predetermined distance from a ground of the lane, and the vehicle side image may be obtained based on the first camera, and an image capturing center line of the first camera is perpendicular to a driving direction of the vehicle; and acquiring a vehicle chassis image based on the second camera, wherein the image acquisition central line of the second camera forms a preset angle with the vehicle running direction.
Through the multi-camera integrated equipment, the image characteristics of the hub and the chassis are respectively collected and extracted in real time, and the debugging and maintenance complexity of the equipment can be reduced.
Optionally, in the above method, the first camera adopts a fisheye short-focus lens, the second camera adopts a short-focus lens, light supplement lamps are arranged in front of the first camera and the second camera, and the image acquisition frequency of the first camera and the second camera is 50-100 frames per second.
In order to ensure the definition of the acquired image, the distance between the camera and the vehicle and the ground and the shooting visual angle of the camera are required to be within a proper range, and the illuminance of a chassis area and a hub area is ensured.
Optionally, in the method, the obtained multiple vehicle side images are spliced to obtain a vehicle side panoramic image; and detecting the panoramic image of the side surface of the vehicle based on the trained deep learning hub detection model to obtain the number of vehicle axles and wheel information, wherein the wheel information comprises the type of the wheel and a wheel coordinate area.
After obtaining the continuous vehicle side image of multiframe, can obtain complete vehicle side panoramic picture through carrying out real-time concatenation to the image, utilize the degree of depth study wheel hub detection model that trains well to detect the wheel hub in the vehicle side panoramic picture simultaneously, obtain vehicle axle number and wheel information to follow-up judge the vehicle axle type.
Optionally, in the method, scanning and detecting the vehicle side image, and when it is detected that wheels exist in the vehicle side image, triggering to acquire a vehicle chassis image; and scanning and detecting the acquired vehicle chassis image frame by frame through the trained deep learning chassis detection model to acquire corresponding differential marking information, and adding the differential marking information into the vehicle chassis image information. Thus, each vehicle chassis image corresponds to the trigger ID and differential label information for one wheel in the vehicle side image.
The method comprises the steps of scanning and detecting wheels in a side image of a vehicle, triggering a second camera to obtain a vehicle chassis image when the wheels are detected, detecting whether a differential mechanism exists in the obtained vehicle chassis image through a trained deep learning chassis detection model, and adding differential mechanism marking information into corresponding vehicle chassis image information when the differential mechanism exists, so that whether the differential mechanism exists in the vehicle chassis image triggered by the wheels correspondingly is indicated.
Optionally, in the above method, calibrating the first camera based on a training friend calibration method to obtain an actual distance corresponding to the pixel; classifying the vehicles according to the number of the vehicle axles, and classifying the vehicles into 2 types of trucks, 3 types of trucks, 4 types of trucks, 5 types of trucks and 6 types of trucks; determining an axle type according to the wheel type according to an axle type standard table, wherein the wheel type comprises a single wheel and a double wheel; or calculating the actual width of the wheels and the actual distance between the wheels through the coordinate areas and the actual distances of the respective wheels, and determining the axle type of the vehicle according to the actual width of the wheels and the actual distance between the wheels.
In the process of determining the axle type, the vehicles are firstly classified according to the number of axles, the axle type of certain types of trucks can be directly determined according to single-wheel and double-wheel information, such as 1 type trucks, 2 types trucks and 6 types of trucks, and the axle type of certain types of trucks needs to be determined according to the actual width of wheels and the actual distance between the wheels, such as 3 types of trucks, 4 types of trucks and 5 types of trucks.
Alternatively, in the above method, for non-category 6 trucks, the number and position of the drive axles are found according to the axle type of the vehicle; and for 6 types of trucks, determining the number and the positions of the driving shafts according to the differential marking information in the vehicle chassis image corresponding to the trigger IDs of the third shaft wheel and the fourth shaft wheel in the vehicle side image.
Optionally, in the above method, the method further comprises: acquiring load information of a vehicle based on a vehicle weighing device installed on the ground of a vehicle running lane; and comparing the acquired vehicle load information with the corresponding vehicle weight limit information to judge whether the vehicle is overloaded.
According to another aspect of the invention, there is provided a computing device comprising: at least one processor; and a memory storing program instructions, wherein the program instructions are configured to be executed by the at least one processor, the program instructions comprising instructions for performing the vehicle weight limit information identification method described above.
According to still another aspect of the present invention, there is provided a readable storage medium storing program instructions that, when read and executed by a computing device, cause the computing device to execute the above-described vehicle weight limit information identifying method.
According to the scheme of the invention, the side images of the running vehicle containing the wheel hub and the chassis images of the running vehicle containing the chassis information are respectively collected, the wheel hub and the chassis in the images are respectively identified and detected based on a deep learning identification algorithm, the axle number, the wheel information and the driving shaft information of the vehicle are obtained, the axle type is determined according to the axle number and the wheel information, the corresponding weight limit information of the vehicle is determined according to the position and the number information of the axle type or the driving shaft, and in an actual test, no matter whether the vehicle is jammed or not, the weight limit identification rate can reach more than 99.5 percent. This scheme can improve vehicle limit for weight information identification's accuracy to solve the problem of unable discernment drive shaft among the prior art, gather wheel hub and chassis image characteristic respectively in real time through many cameras equipment of integration, can reduce the debugging and the maintenance complexity of equipment.
The above description is only an overview of the technical solutions of the present invention, and the present invention can be implemented in accordance with the content of the description so as to make the technical means of the present invention more clearly understood, and the above and other objects, features, and advantages of the present invention will be more clearly understood.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
FIG. 1 illustrates a vehicle information collection device installation diagram according to one embodiment of the present invention;
FIG. 2 illustrates a flow diagram of a vehicle weight limit information identification method 200 according to one embodiment of the invention;
FIG. 3 illustrates a schematic view of a vehicle side panoramic image, according to one embodiment of the present invention;
FIG. 4 illustrates a decision diagram for determining corresponding weight limit information based on vehicle axle type and drive axle, according to an embodiment of the present invention;
FIG. 5 illustrates a flow diagram for determining vehicle weight limit information according to one embodiment of the present invention;
FIG. 6 shows a schematic block diagram of a computing device 100, according to an embodiment of the invention.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
In order to reduce the occurrence of traffic accidents, traffic control departments have different limits on the running speed and the load of vehicles, and the allowed weighing limit values of different vehicle types are different due to the difference of the number of axles, the type of the axles, the distance between the axles, the number of driving axles and the like. And the information such as the axle number, axle type, gross weight of freight train need be transmitted to the toll collector through transfinite overload detecting system at toll booth, charge according to the vehicle transfinite condition, reach the purpose of administering transfinite overload through this kind of mode. In order to accurately identify information such as the number of axles, the type of axles, a driving shaft and the like of a vehicle, the scheme provides a vehicle weight limit information identification method, the number of axles, the type of axles and the driving shaft are identified through a deep learning model, and wheel distances and single-wheel and double-wheel characteristics are calculated, so that whether the vehicle is over-limit and overloaded or not is judged according to the identified vehicle information and vehicle load. In order to gather vehicle image information and vehicle weight information in the process of traveling in real time, this scheme adopts many cameras of integration equipment to gather respectively and contains vehicle wheel hub's side image and contain vehicle chassis's positive image to and be used for carrying out the weighbridge (truck scale) of measuring to the freight train load.
Fig. 1 shows a schematic view of a vehicle information collection apparatus installation according to an embodiment of the present invention. As shown in fig. 1, a grating sensor and an integrated multi-camera device are arranged at an entrance of a vehicle driving direction, and a wagon balance is arranged on the ground in the middle of a lane and used for weighing the vehicle and a control cabinet positioned on one side of the lane. The grating sensor is used for detecting the displacement of a vehicle, the integrated multi-camera equipment is installed perpendicular to a lane, and the equipment at least comprises a first camera used for acquiring a side image of the vehicle and a second camera used for acquiring a chassis image of the vehicle. According to the position of the wheel hub and the chassis of the vehicle away from the ground, the installation position of the integrated multi-camera equipment can be 0.5m-2.0m away from the side face of the vehicle and 0.5m-1.5m away from the ground of the lane. The center line of the visual field of the first camera is vertical to the driving direction of the vehicle, and a fisheye short-focus lens can be used so as to enable the lens to achieve the maximum shooting visual angle close to 180 degrees. The shooting visual angle of the second camera is an area which forms 30-65 degrees with the driving direction of the vehicle, and the second camera can use a short-focus lens to ensure a preset shooting visual angle. In order to make the image of shooing more clear, avoid influencing the image acquisition effect owing to lack illuminance, first camera and second camera can also embed the light filling lamp among the many camera equipment of integration for regional and the chassis region of vehicle wheel hub carries out the light filling.
Fig. 2 shows a schematic flow diagram of a vehicle weight limit information identification method 200 according to an embodiment of the invention. As shown in fig. 2, the method starts with step S210 of acquiring a vehicle side image and a vehicle chassis image captured by an image capturing device for a running vehicle.
The image acquisition device can use an integrated multi-camera device which is installed perpendicular to a lane as shown in fig. 1, during image acquisition, a vehicle can normally run on the lane, the image acquisition device is a first preset distance away from the vehicle running lane and a second preset distance away from the ground of the lane, the image acquisition center line of the first camera is perpendicular to the vehicle running direction, and a vehicle side image can be acquired based on the first camera; the image acquisition center line of the second camera and the vehicle driving direction form a preset angle, and the vehicle chassis image can be acquired based on the second camera. In one embodiment of the invention, the integrated multi-camera device is installed at a position 1m away from the near end edge of the lane and 1m away from the ground of the lane, the first camera uses a short-focus lens with a focal length of 1.6mm, the second camera uses a short-focus lens with a focal length of 6mm, and the shooting angle is in a direction of 45 degrees with respect to the driving direction of the vehicle. The first camera and the second camera may acquire the vehicle side images and the vehicle chassis images at a preset acquisition frequency, for example, 50 frames per second to 100 frames per second. When the grating sensor at the entrance of the lane detects that the vehicle runs in the lane, an image acquisition instruction can be sent to the first camera to acquire a plurality of continuous vehicle side images. And calibrating the first camera on the basis of a Zhang-Yong calibration method for the acquired vehicle side image to obtain a conversion relation between the pixel coordinate and the world coordinate. Specifically, a checkerboard calibration board may be first manufactured, a camera is used to shoot the checkerboard calibration board at different angles to obtain a group of images, and the calibration board angular points in the images are detected to obtain the pixel coordinates of each angular point; then fixing the world coordinate system on a chessboard grid calibration plate to obtain the physical coordinates of each corner point under the world coordinate system; and finally, acquiring an internal reference matrix, an external reference matrix and distortion parameters of the camera based on a Python-openCV calibration program, further acquiring a conversion relation between pixel coordinates and world coordinates, and further acquiring an actual distance corresponding to the pixels so as to obtain an actual distance between objects in the image.
After a plurality of continuous vehicle side images are obtained, the plurality of vehicle side images with overlapped parts can be spliced to obtain a seamless vehicle side panoramic image, and the vehicle side panoramic image is obtained. FIG. 3 shows a schematic view of a vehicle side panoramic image, according to one embodiment of the present invention. As shown in fig. 3, the vehicle side panoramic image is a vehicle side panoramic image mainly containing wheel information obtained by splicing a plurality of real-time acquired continuous vehicle side images. In the image acquisition process, scanning detection needs to be carried out on the vehicle side image, and when the vehicle side image is detected to have wheels and pass a preset distance, the second camera is triggered to acquire the vehicle chassis image. The vehicle chassis image comprises information of a drive train, a running train, a steering train and a braking train of a vehicle chassis, wherein the drive train comprises a clutch, a transmission, a universal transmission device, a main speed reducer, a differential mechanism, a half shaft (driving shaft) and the like, and the information of the differential mechanism and the driving shaft can be obtained by obtaining the vehicle chassis image, so that the load standard of the vehicle is determined according to the number and the position of the driving shafts and the number and the type of the shafts of the vehicle. In one embodiment of the invention, the capturing area of the second camera is 1.5m away from the device, and the second camera can be triggered to capture the vehicle chassis image when one wheel is detected in the vehicle side image and 1.5m passes. That is, each wheel has a unique trigger ID associated therewith, and each vehicle floor image corresponds to the trigger ID associated with one of the wheels in the vehicle side images.
And step S220 is executed, the wheel hubs in the panoramic image of the side surface of the vehicle are detected based on the trained deep learning wheel hub detection model, and the number of vehicle axles and wheel information are obtained.
Specifically, the vehicle side images can be spliced in real time, and simultaneously, the trained deep learning hub detection model is used for detecting the hub in the vehicle side images, so that the number of vehicle axles and a wheel coordinate region (a wheel rectangular region determined by the coordinates of the upper left pixel point and the coordinates of the lower right pixel point of the wheel) are obtained. The number of vehicle axles refers to the number of axles connected with tires, and large trucks, military vehicles, special vehicles and large buses are multi-axle vehicles generally. For trucks, the greater the number of axles the greater the load. Because the vehicle parameters of the single wheel and the double wheels are different, for example, the working quality, the dead line load, the turning radius, the walking speed and the wheel base are different, the trained deep learning wheel hub classification model is further required to be used for extracting and calculating the wheel texture in the wheel rectangular area, and the information of the single wheel and the double wheels is judged.
And step S230 is executed, whether the differential exists in the vehicle chassis image is detected based on the trained deep learning chassis detection model, and differential marking information is obtained.
The parallel wheels of each group of the automobile are connected with each other through a shaft, and the shaft can be a through shaft or two disconnected half shafts. A half shaft, also called an automobile drive shaft, is a shaft that connects a differential mechanism and drive wheels, and transmits power of the differential mechanism to left and right drive wheels, so that the left and right wheels can rotate at different angular velocities. Differential marking information can be obtained by identifying and detecting the differential in the vehicle chassis image, so that the detected vehicle chassis image comprises the differential information and the wheel trigger ID information.
Step S240 is then executed to determine the axle type of the vehicle based on the number of vehicle axles and the wheel information. The wheel information comprises wheel types and wheel coordinate information, the wheel types comprise single wheels and double wheels, and the wheel types can be identified through a classification model based on a deep learning algorithm. The wheel coordinate information is a detected wheel rectangular area determined by the pixel point coordinate p1 (x 1, x 2) at the upper left corner and the pixel point coordinate p2 (x 2, y 2) at the lower right corner. When determining the axle type of the vehicle, the vehicle is firstly classified according to the number of axles, and when the number of wheels is a few from front to back when viewed from one side of the truck, the vehicle is a few-axle vehicle. Light trucks with a total length of two axles of less than 6 meters and a total weight of less than 4.5 tons are listed in class 1 trucks. Two-axle trucks having a total length of 6m or more or a maximum allowable total mass of 4.5 tons or more are classified as class 2 trucks. Three to six trucks are class 3, class 4, class 5 and class 6 trucks, respectively. The axle types of the passenger car, the class-1 truck, the class-2 truck and the class-6 truck can be obtained by directly classifying according to single-wheel and double-wheel information and an axle type standard table. The 3-class trucks, the 4-class trucks and the 5-class trucks need to be classified according to the actual wheel spacing and the actual wheel width in combination with an axle type standard table to obtain corresponding axle types. The actual width (x 2-x 1) × k of the wheel can be calculated according to the pixel coordinates (x 1, y 1) at the upper left corner of the wheel coordinate area, the pixel coordinates (x 2, y 2) at the lower right corner, the pixel coordinates (x 3, y 3) at the upper left corner of the adjacent wheel coordinate area and the actual distance k corresponding to one pixel, and the actual distance of the wheel is (x 3-x 2) × k.
Step S250 is then performed to determine the position and number of drive axles based on the axle type or differential signature information of the vehicle.
For 6 types of trucks, the number and the position of the driving shafts can be directly obtained through the axle type, and for 6 types of trucks, the vehicle chassis image with the same trigger ID is searched through the trigger ID of the wheels, the chassis differential marking information is extracted from the image information, and the position of the driving shaft is judged according to the corresponding chassis differential marking information.
Finally, step S260 is performed to determine weight limit information of the vehicle according to the axle type standard table based on the axle type or the position and number of the driving axles of the vehicle.
The number of axles, the axle type and the driving axle of the vehicle can be accurately judged by combining the calculation and analysis, and then the corresponding vehicle weight limit information can be determined. Fig. 4 shows a schematic decision diagram for determining corresponding weight limit information according to a vehicle axle type or a drive axle according to an embodiment of the invention. As shown in fig. 4, the vehicles are firstly classified according to the number of axles of the vehicles, 2 axles correspond to 2 types of trucks, 3 axles correspond to 3 types of trucks, 4 axles correspond to 4 types of trucks, 5 axles correspond to 5 types of trucks, and 6 axles correspond to 6 types of trucks, the axle types of the vehicles with the same number of axles are firstly distinguished according to single-wheel and double-wheel information, the axle type of the 2 types of trucks can be directly determined according to single-wheel and double-wheel information, the axle type corresponding to a single wheel is 1+2, the corresponding weight limit information is 18 tons, the axle type corresponding to a double wheel is 1+1, and the corresponding weight limit information is 18 tons. The other 3 types of trucks, 4 types of trucks and 5 types of trucks cannot be classified according to single-double wheel information, and can be classified according to the actual distance between the wheels and the actual width of the wheels in combination with an axle type standard table, and the corresponding axle type is determined by comparing the wheel distance with the wheel width, for example, for the 3 types of trucks, it can be firstly determined whether the distance between the second axle wheel and the third axle wheel is smaller than the width of the wheels, if the distance between the second axle wheel and the third axle wheel is smaller than the wheel width, the axle type of the vehicle is determined to be 1+22, the corresponding weight limit information is 25 tons, if the distance between the second axle wheel and the third axle wheel is not smaller than the wheel width, it is continuously determined whether the distance between the 1 st axle wheel and the 2 nd axle wheel is smaller than the wheel width, if the distance between the first axle wheel and the second axle wheel is smaller than the wheel width, the axle type of the vehicle is determined to be 11+2, the corresponding weight limit information is 25 tons, and if the distance between the first axle and the second axle is not smaller than the wheel width, the corresponding weight limit information is determined to be 27 tons.
The axle type determination procedures for other class 3 trucks, class 4 trucks, class 5 trucks and class 6 trucks are similar and are not described in detail herein.
For non-6 types of trucks, the position and number of the driving axle can be given by the axle type standard, while for 6 types of trucks, in the process of determining the vehicle weight limit information, the position and number of the driving axle need to be determined by using the chassis differential information, namely, a vehicle chassis image with the same trigger ID is searched by the trigger IDs of the third axle wheel and the fourth axle wheel, and the chassis differential mark information is extracted from the image information. Finally, after the vehicle axle type and the driving axle are determined, the corresponding weight limit information of the vehicle is obtained. As shown in fig. 4, for 6 types of trucks, it is first determined whether the distances between the fourth axle wheel, the fifth axle wheel and the sixth axle wheel are all smaller than the wheel width, if all the distances are smaller than the wheel width, it is continuously determined whether the distance between the first axle wheel and the second axle wheel is smaller than the wheel width, if the distance between the 1 st axle wheel and the 2 nd axle wheel is not smaller than the wheel width, it is determined that the axle shape of the vehicle is 1+22+222, it is continuously determined whether the third axle wheel has a differential, if there is a differential, it is determined that the vehicle weight limit is 49 tons, and if there is no differential, it is determined that the vehicle weight limit is 46 tons. If the distances among the fourth axle wheel, the fifth axle wheel and the sixth axle wheel are not all smaller than the wheel width, whether the distance between the fifth axle wheel and the sixth axle wheel is smaller than the wheel width is judged, if so, the axle shape of the vehicle is determined to be 11+22, if not, the axle shape of the vehicle is determined to be 11+22+2, at this time, whether a differential mechanism exists on the fourth axle wheel needs to be further judged, if so, the weight limit of the vehicle is determined to be 49 tons, and if not, the weight limit of the vehicle is determined to be 46 tons.
FIG. 5 illustrates a flow diagram for determining vehicle weight limit information according to one embodiment of the present invention. As shown in fig. 5, firstly, image acquisition is performed, a vehicle side image and a vehicle chassis image are respectively acquired through an integrated multi-camera device, then, image stitching is performed on a plurality of acquired continuous vehicle side images to obtain a complete vehicle side image, a vehicle axle number is obtained by detecting the complete vehicle side image through a deep learning hub detection model, and axle types of certain types of vehicles are determined according to the axle number; and simultaneously, triggering scanning and identification of the vehicle chassis image when the wheels in the vehicle side image are detected, detecting whether a differential exists in the vehicle chassis image through a deep learning chassis detection model to obtain corresponding differential marking information, and further determining the position and the number of the driving shafts according to the differential marking information or the axle type of the vehicle. And finally, after the vehicle axle type and the driving axle are determined, the corresponding weight limit information of the vehicle is obtained.
FIG. 6 shows a block diagram of a computing device 100, according to one embodiment of the invention. As shown in FIG. 6, in a basic configuration 102, a computing device 100 typically includes a system memory 106 and one or more processors 104. A memory bus 108 may be used for communication between the processor 104 and the system memory 106.
Depending on the desired configuration, the processor 104 may be any type of processor, including but not limited to: a microprocessor (μ P), a microcontroller (μ C), a digital information processor (DSP), or any combination thereof. The processor 104 may include one or more levels of cache, such as a level one cache 110 and a level two cache 112, a processor core 114, and registers 116. The example processor core 114 may include an Arithmetic Logic Unit (ALU), a Floating Point Unit (FPU), a digital signal processing core (DSP core), or any combination thereof. The example memory controller 118 may be used with the processor 104, or in some implementations the memory controller 118 may be an internal part of the processor 104.
Depending on the desired configuration, system memory 106 may be any type of memory, including but not limited to: volatile memory (such as RAM), non-volatile memory (such as ROM, flash memory, etc.), or any combination thereof. The physical memory in the computing device is usually referred to as a volatile memory RAM, and data in the disk needs to be loaded into the physical memory to be read by the processor 104. System memory 106 may include an operating system 120, one or more applications 122, and program data 124. In some implementations, the application 122 can be arranged to execute instructions on an operating system with program data 124 by one or more processors 104. Operating system 120 may be, for example, linux, windows, etc., which includes program instructions for handling basic system services and performing hardware dependent tasks. The application 122 includes program instructions for implementing various user-desired functions, and the application 122 may be, for example, but not limited to, a browser, instant messenger, a software development tool (e.g., an integrated development environment IDE, a compiler, etc.), and the like. When the application 122 is installed into the computing device 100, a driver module may be added to the operating system 120.
When the computing device 100 is started, the processor 104 reads program instructions of the operating system 120 from the memory 106 and executes them. The application 122 runs on top of the operating system 120, utilizing the operating system 120 and interfaces provided by the underlying hardware to implement various user-desired functions. When the user starts the application 122, the application 122 is loaded into the memory 106, and the processor 104 reads the program instructions of the application 122 from the memory 106 and executes the program instructions.
The computing device 100 also includes a storage device 132, the storage device 132 including removable storage 136 and non-removable storage 138, the removable storage 136 and the non-removable storage 138 each connected to the storage interface bus 134.
Computing device 100 may also include an interface bus 140 that facilitates communication from various interface devices (e.g., output devices 142, peripheral interfaces 144, and communication devices 146) to the basic configuration 102 via the bus/interface controller 130. The example output device 142 includes a graphics processing unit 148 and an audio processing unit 150. They may be configured to facilitate communication with various external devices, such as a display or speakers, via one or more a/V ports 152. Example peripheral interfaces 144 may include a serial interface controller 154 and a parallel interface controller 156, which may be configured to facilitate communication with external devices such as input devices (e.g., keyboard, mouse, pen, voice input device, touch input device) or other peripherals (e.g., printer, scanner, etc.) via one or more I/O ports 158. An example communication device 146 may include a network controller 160, which may be arranged to facilitate communications with one or more other computing devices 162 over a network communication link via one or more communication ports 164.
A network communication link may be one example of a communication medium. Communication media may typically be embodied by computer readable instructions, data structures, program modules, and may include any information delivery media, such as carrier waves or other transport mechanisms, in a modulated data signal. A "modulated data signal" may be a signal that has one or more of its data set or its changes made in such a manner as to encode information in the signal. By way of non-limiting example, communication media may include wired media such as a wired network or private-wired network, and various wireless media such as acoustic, radio Frequency (RF), microwave, infrared (IR), or other wireless media. The term computer readable media as used herein may include both storage media and communication media. In the computing device 100 according to the invention, the application 122 comprises instructions for carrying out the vehicle weight limit information identification method 200 of the invention.
According to the scheme, the side images of the running vehicle containing the wheel hubs and the chassis images of the running vehicle containing the chassis information are respectively collected, the wheel hubs and the chassis in the images are respectively identified and detected based on a deep learning identification algorithm, the number of axles, the wheel information and the driving shaft information of the vehicle are obtained, the axle type is determined according to the number of axles and the wheel information, the corresponding weight limit information of the vehicle is determined according to the axle type or the position and number information of the driving shaft, and in an actual test, no matter whether the vehicle is jammed or not, the weight limit identification rate can reach more than 99.5%. This scheme can improve vehicle limit for weight information identification's accuracy to solve the problem of unable discernment drive shaft among the prior art, gather wheel hub and chassis image characteristic respectively in real time through many cameras equipment of integration, can reduce the debugging and the maintenance complexity of equipment.
In the description provided herein, numerous specific details are set forth. However, it is understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.
Similarly, it should be appreciated that in the foregoing description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. However, the disclosed method should not be interpreted as reflecting an intention that: rather, the invention as claimed requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this invention.
Those skilled in the art will appreciate that the modules or units or components of the devices in the examples disclosed herein may be arranged in a device as described in this embodiment or alternatively may be located in one or more devices different from the devices in this example. The modules in the foregoing examples may be combined into one module or may be further divided into multiple sub-modules.
Those skilled in the art will appreciate that the modules in the device in an embodiment may be adaptively changed and disposed in one or more devices different from the embodiment. The modules or units or components of the embodiments may be combined into one module or unit or component, and furthermore they may be divided into a plurality of sub-modules or sub-units or sub-components. All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or elements of any method or apparatus so disclosed, may be combined in any combination, except combinations where at least some of such features and/or processes or elements are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.
Furthermore, those skilled in the art will appreciate that while some embodiments described herein include some features included in other embodiments, rather than other features, combinations of features of different embodiments are meant to be within the scope of the invention and form different embodiments. For example, in the following claims, any of the claimed embodiments may be used in any combination.
Furthermore, some of the embodiments are described herein as a method or combination of method elements that can be implemented by a processor of a computer system or by other means of performing a function. A processor with the necessary instructions for implementing a method or method elements thus forms an apparatus for implementing the method or method elements. Further, the elements of the apparatus embodiments described herein are examples of the following apparatus: the apparatus is used to implement the functions performed by the elements for the purpose of carrying out the invention.
As used herein, unless otherwise specified the use of the ordinal adjectives "first", "second", "third", etc., to describe a common object, merely indicate that different instances of like objects are being referred to, and are not intended to imply that the objects so described must be in a given sequence, either temporally, spatially, in ranking, or in any other manner.
While the invention has been described with respect to a limited number of embodiments, those skilled in the art, having benefit of this description, will appreciate that other embodiments can be devised which do not depart from the scope of the invention as described herein. Moreover, it should be noted that the language used in the specification has been principally selected for readability and instructional purposes, and may not have been selected to delineate or circumscribe the inventive subject matter. Accordingly, many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the appended claims. The present invention has been disclosed in an illustrative rather than a restrictive sense with respect to the scope of the invention, as defined in the appended claims.
Claims (10)
1. A vehicle weight limit information identification method, adapted to be executed in a computing device, comprising:
acquiring a vehicle side image and a vehicle chassis image acquired by an image acquisition device on a running vehicle;
detecting the wheel hub in the side image of the vehicle based on the trained deep learning wheel hub detection model to obtain the number of vehicle axles and wheel information;
detecting whether a differential mechanism exists in the vehicle chassis image or not based on the trained deep learning chassis detection model to obtain differential mechanism marking information;
determining the axle type of the vehicle according to the number of the vehicle axles and the wheel information;
determining the position and the number of driving shafts according to the shaft type or differential marking information of the vehicle;
and determining the weight limit information of the vehicle according to an axle type standard table based on the axle type of the vehicle or the position and the number of the driving axles.
2. The method of claim 1, wherein the image capturing device is vertically installed at an entrance of a driving lane of the vehicle, the image capturing device is located a first predetermined distance from a near end of the driving lane and a second predetermined distance from a ground of the driving lane, the image capturing device comprises at least a first camera and a second camera, and the step of acquiring the side images and the chassis images of the vehicle captured by the image capturing device on the driving vehicle comprises:
acquiring a vehicle side image based on a first camera, wherein the image acquisition center line of the first camera is vertical to the vehicle running direction;
and acquiring a vehicle chassis image based on a second camera, wherein an image acquisition central line of the second camera forms a preset angle with the vehicle running direction.
3. The method according to claim 2, wherein the first camera adopts a fisheye short-focus lens, the second camera adopts a short-focus lens, a fill light is arranged in front of the first camera and the second camera, and the image acquisition frequency of the first camera and the second camera is 50-100 frames per second.
4. The method of claim 1, wherein the step of detecting the wheel hub in the vehicle side image based on the trained deep learning wheel hub detection model to obtain the number of vehicle axles and wheel information comprises:
splicing the acquired multiple vehicle side images to obtain a vehicle side panoramic image;
and detecting a vehicle side panoramic image based on the trained deep learning hub detection model to obtain the number of vehicle axles and wheel information, wherein the wheel information comprises the type of a wheel and a wheel coordinate area.
5. The method of claim 1, wherein the step of detecting whether a differential exists in the vehicle chassis image based on the trained deep learning chassis detection model to obtain differential labeling information comprises:
scanning and detecting the vehicle side image, and triggering to acquire a vehicle chassis image when wheels exist in the vehicle side image;
and scanning and detecting the acquired vehicle chassis images frame by frame through the trained deep learning chassis detection model to acquire corresponding differential marking information, wherein each vehicle chassis image corresponds to the trigger ID of one wheel in the vehicle side image and the differential marking information.
6. The method of claim 3, wherein the step of determining the axle type of the vehicle based on the number of vehicle axles and the wheel information comprises:
calibrating the first camera based on a Zhangyingyou calibration method to obtain an actual distance corresponding to an image pixel;
classifying the vehicles according to the number of vehicle axles, and classifying the vehicles into 2 types of trucks, 3 types of trucks, 4 types of trucks, 5 types of trucks and 6 types of trucks;
determining an axle type according to the wheel type according to an axle type standard table, wherein the wheel type comprises a single wheel and a double wheel; or
And calculating the actual width of the wheels and the actual distance between the wheels through the coordinate areas and the actual distances of the wheels, and determining the axle type of the vehicle according to the actual width of the wheels and the actual distance between the wheels.
7. The method of claim 6, wherein the step of determining the location and number of drive axles based on axle type or differential signature information of the vehicle comprises:
for non-6 types of trucks, the number and the positions of the driving shafts are obtained according to the axle type of the vehicle;
and for 6 types of trucks, determining the number and the positions of the driving shafts according to the differential marking information in the vehicle chassis image corresponding to the trigger IDs of the third shaft wheel and the fourth shaft wheel in the vehicle side image.
8. The method of claim 1, further comprising:
acquiring load information of a vehicle based on a vehicle weighing device installed on the ground of a vehicle driving lane;
and comparing the acquired vehicle load information with the corresponding vehicle weight limit information, and judging whether the vehicle is overloaded.
9. A computing device, comprising:
at least one processor; and
a memory storing program instructions configured for execution by the at least one processor, the program instructions comprising instructions for performing the method of any of claims 1-8.
10. A readable storage medium storing program instructions that, when read and executed by a computing device, cause the computing device to perform the method of any of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211183756.5A CN115512321A (en) | 2022-09-27 | 2022-09-27 | Vehicle weight limit information identification method, computing device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211183756.5A CN115512321A (en) | 2022-09-27 | 2022-09-27 | Vehicle weight limit information identification method, computing device and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115512321A true CN115512321A (en) | 2022-12-23 |
Family
ID=84506926
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211183756.5A Pending CN115512321A (en) | 2022-09-27 | 2022-09-27 | Vehicle weight limit information identification method, computing device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115512321A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116189113A (en) * | 2022-12-29 | 2023-05-30 | 北京中科神通科技有限公司 | Truck type recognition method and system |
| CN117953460A (en) * | 2024-03-26 | 2024-04-30 | 江西众加利高科技股份有限公司 | Vehicle wheel axle identification method and device based on deep learning |
-
2022
- 2022-09-27 CN CN202211183756.5A patent/CN115512321A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116189113A (en) * | 2022-12-29 | 2023-05-30 | 北京中科神通科技有限公司 | Truck type recognition method and system |
| CN116189113B (en) * | 2022-12-29 | 2024-03-08 | 北京中科神通科技有限公司 | Truck type recognition method and system |
| CN117953460A (en) * | 2024-03-26 | 2024-04-30 | 江西众加利高科技股份有限公司 | Vehicle wheel axle identification method and device based on deep learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108898044B (en) | Loading rate obtaining method, device and system and storage medium | |
| CN115512321A (en) | Vehicle weight limit information identification method, computing device and storage medium | |
| US10198619B1 (en) | System and method for evaluating the perception system of an autonomous vehicle | |
| CN111882882B (en) | Method for detecting cross-lane driving behavior of automobile in dynamic flat-plate scale weighing area | |
| CN102809413B (en) | Detecting method and detecting device for rigged dynamic weighing of vehicle | |
| CN102556066B (en) | Lane departure warning device for passenger vehicle and judgment method thereof | |
| CN106127107A (en) | The model recognizing method that multi-channel video information based on license board information and vehicle's contour merges | |
| CN112966582A (en) | Vehicle type three-dimensional recognition method, device and system, electronic equipment and storage medium | |
| CN104029680A (en) | Lane departure warning system and method based on monocular camera | |
| CN201825037U (en) | Lane departure alarm device for vehicles on highway | |
| CN111292432A (en) | Vehicle charging type distinguishing method and device based on vehicle type recognition and wheel axle detection | |
| CN111915883A (en) | Road traffic condition detection method based on vehicle-mounted camera shooting | |
| CN114822029B (en) | Bridge deck traffic flow load space-time distribution reconstruction method, system and device | |
| CN105021126A (en) | Truck side guard rail mounting size measurement system and method based on machine vision technology | |
| CN110991264A (en) | Front vehicle detection method and device | |
| CN114119955A (en) | Method and device for detecting potential dangerous target | |
| CN110660226A (en) | Method, system and equipment for detecting vehicle safety standard and storage device | |
| CN112183206B (en) | Traffic participant positioning method and system based on road side monocular camera | |
| CN105043253A (en) | Truck side protection guard installation size measurement system based on surface structure light technology and method thereof | |
| CN205788764U (en) | The automatic evidence-collecting device of steering indicating light is not opened in a kind of vehicle modified line | |
| CN107506739A (en) | To vehicle detection and distance-finding method before a kind of night | |
| CN114693722B (en) | Vehicle driving behavior detection method, detection device and detection equipment | |
| CN112990117B (en) | Installation data processing method and device based on intelligent driving system | |
| CN113375618B (en) | Vehicle type identification method based on axle distance calculation | |
| CN109686107A (en) | A kind of intelligent transport vehicle vehicle bottom identifying system |
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 |