CN111735524A - Tire load obtaining method based on image recognition, vehicle weighing method and system - Google Patents
Tire load obtaining method based on image recognition, vehicle weighing method and system Download PDFInfo
- Publication number
- CN111735524A CN111735524A CN202010874283.8A CN202010874283A CN111735524A CN 111735524 A CN111735524 A CN 111735524A CN 202010874283 A CN202010874283 A CN 202010874283A CN 111735524 A CN111735524 A CN 111735524A
- Authority
- CN
- China
- Prior art keywords
- tire
- vehicle
- image
- load
- tested
- 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.)
- Granted
Links
Images
Classifications
-
- 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
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/02—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/24—Aligning, centring, orientation detection or correction of the image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
-
- 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
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Tires In General (AREA)
Abstract
The invention discloses a tire load obtaining method, a vehicle weighing method and a system based on image recognition, which are characterized in that tire images of all tires of a vehicle to be detected are obtained and sent to a data processing assembly, and the data processing assembly is used for detecting the contact area and the tire pressure of each tire and the ground from the tire images of all the tires; inputting the contact area of each tire and the corresponding tire pressure into a load model, and calculating the load of each tire; compared with the prior art, the tire load obtaining method, the vehicle weighing method and the system have the advantages of wide measuring range, no need of any additional sensing equipment, no need of traffic closure, long-term stable work and easy information integration.
Description
Technical Field
The invention relates to the field of image recognition and weighing, in particular to a tire load obtaining method based on image recognition, a vehicle weighing method and a vehicle weighing system.
Background
With the rapid development of the transportation industry, the rapid, long-distance and heavy-load become an important component and development trend of road freight, which also leads to the increasingly serious overload and overrun phenomena. The overload overrun vehicle can cause irreversible damage to the infrastructures of roads, bridges and the like, thereby not only causing huge economic loss, but also seriously threatening the life safety of people.
With the deep combination of artificial intelligence and transportation industry, the intelligent transportation system is developed. The intelligent transportation system is an informationized, automated, intelligentized and socialized novel transportation system formed by transforming a traditional transportation system by utilizing various high and new technologies, and the intelligent transportation system is the development direction of the future transportation system. In an intelligent traffic system, non-contact vehicle identification based on computer vision is an important part of traffic intelligence.
Computer vision is receiving increasing attention as a non-contact advanced measurement technology in intelligent traffic systems. The traditional vehicle weight detection method mainly adopts a static wagon balance and a contact sensing type road weighing system, however, the weighing wagon balance or the road weighing equipment is easy to damage after the vehicle is continuously rolled, the service life is short, the maintenance cost is high, and traffic is easy to block.
Therefore, how to solve the technical problems of traffic blockage, short service life and high maintenance cost which are easily generated by the conventional weighing system becomes an urgent need to be solved by technical personnel in the field.
Disclosure of Invention
The invention provides a tire load obtaining method based on image recognition, a vehicle weighing method and a vehicle weighing system, which are used for solving the technical problems of traffic blockage, short service life and high maintenance cost of the conventional weighing system.
In order to solve the technical problems, the technical scheme provided by the invention is as follows:
a tire load obtaining method based on image recognition comprises the following steps:
obtaining a tire image of a tire to be detected; detecting the contact area of the tire and the ground and the tire pressure from the tire image; and inputting the contact area and the corresponding tire pressure into a load model, and calculating the load of the tire.
Preferably, the method for detecting the contact area of the tire and the ground from the tire image comprises the following steps:
separating a target tire from a background in the tire image to obtain pixel points of a hub of the target tire and the edge of the tire;
identifying a tire specification mark on a target tire, and reading the diameter of a hub and the width of the tire in the tire specification mark;
obtaining a scale factor of a tire image according to the ratio of the physical length of the diameter of the hub to the pixel point of the hub;
and calculating the number of pixels of the target tire in contact with the ground, calculating the contact length of the target tire and the ground according to the number of pixels and the scale factor, and calculating the contact area according to the contact length and the tire width.
Preferably, the step of separating the target tire from the background in the tire image comprises the steps of:
scaling the tire image by adopting equal-scale transformation, and transforming the tire image into a frequency domain by adopting a wavelet threshold denoising algorithm to perform image denoising treatment to obtain a denoised tire image;
the tire object is separated from the background by a JSEG (J-value segmentation) algorithm of color image segmentation and a region growing algorithm.
Preferably, calculating the contact area according to the contact length and the tire width comprises the following steps:
area of contact
= actual tire contact length × tire width × area reduction rate;
preferably, the tire pressure of the tire detected from the tire image is obtained by the following steps:
identifying the tire specification mark on the target tire, reading the reference tire pressure in the tire specification mark, and multiplying the reference tire pressure by a preset correction coefficient to obtain the actual tire pressure value of the target tire.
Preferably, the loading model is:
wherein the content of the first and second substances,iis the first of the vehicle to be testediA shaft;
for the vehicle to be testediAxial load in KN;
for the vehicle to be testediAxle tire pressure in MPa;
for the vehicle to be testediContact area of the tire of the axle in units of
；
，
And
the regression coefficients are respectively a first regression coefficient, a second regression coefficient and a third regression coefficient, and the value ranges of the first regression coefficient, the second regression coefficient and the third regression coefficient are all in the (0, 1) interval.
Preferably, the tire image of the tire to be detected is acquired by an image acquisition device/video acquisition device.
A vehicle weighing method based on image recognition comprises the following steps:
acquiring the load of each tire of the vehicle to be detected by the tire load acquisition method based on the image recognition;
and calculating the total weight of the vehicle to be tested according to the load of each tire of the vehicle to be tested.
Preferably, after calculating the total weight of the vehicle to be tested according to the load of each tire of the vehicle to be tested, the method further comprises the following steps:
identifying the category of the vehicle to be detected according to the tire image of the vehicle to be detected;
comparing the total weight of the vehicle to be tested with a total weight limit value corresponding to the category of the vehicle to be tested, judging that the vehicle to be tested is overweight when the total weight of the vehicle to be tested is greater than the corresponding total weight limit value, and outputting an alarm signal;
and sending the tire image of the vehicle to be tested, the vehicle weight and whether the vehicle is overweight to a cloud server for storage.
A computer system comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of any of the methods described above when executing the computer program.
The invention has the following beneficial effects:
1. the tire load obtaining method, the vehicle weighing method and the system based on image recognition can quickly and accurately recognize the tire load and the vehicle weight of the vehicle to be detected, and compared with the prior art, the vehicle weighing method and the system have the advantages of wide measurement range, no need of any additional sensing equipment, no need of traffic closure, long-term stable work and easy information integration.
2. In a preferred scheme, the vehicle weighing method and the system are non-contact weighing systems, can automatically realize functions of all-weather 24-hour uninterrupted vehicle weight data acquisition, judgment, early warning and the like, are beneficial to off-site law enforcement of overload and overrun, and can conveniently check illegal behaviors of vehicle overload and overrun. The cloud end of the system provides information such as the vehicle weight, the vehicle axle and the vehicle length, evidence can be provided for off-site law enforcement, interference of human factors is reduced, and artificial emergencies in the law enforcement process are avoided, so that the labor cost is effectively saved.
In addition to the objects, features and advantages described above, other objects, features and advantages of the present invention are also provided. The present invention will be described in further detail below with reference to the accompanying drawings.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate embodiments of the invention and, together with the description, serve to explain the invention and not to limit the invention. In the drawings:
FIG. 1 is a pre-processed tire image of a preferred embodiment of the present invention;
FIG. 2 is a tire edge image after image segmentation in accordance with a preferred embodiment of the present invention;
FIG. 3 is a diagram of a pixel point marking of the rim diameter and tire sidewall marking of a preferred embodiment of the present invention;
FIG. 4 is a contact length of a tire according to a preferred embodiment of the present invention;
FIG. 5 is a graph of tire pressure, load, and tire average ground contact pressure for a preferred embodiment of the present invention;
FIG. 6 is a data flow diagram of a method for weighing a vehicle based on image recognition in accordance with a preferred embodiment of the present invention;
FIG. 7 is a first flowchart of a method for weighing a vehicle based on image recognition in accordance with a preferred embodiment of the present invention;
FIG. 8 is a second flowchart of a method for weighing a vehicle based on image recognition in accordance with a preferred embodiment of the present invention;
fig. 9 is a flowchart of a tire load obtaining method based on image recognition according to the present invention.
Detailed Description
The embodiments of the invention will be described in detail below with reference to the drawings, but the invention can be implemented in many different ways as defined and covered by the claims.
The first embodiment is as follows:
as shown in fig. 9, the present invention discloses a tire load obtaining method based on image recognition, comprising the steps of:
obtaining a tire image of a tire to be detected; detecting the contact area of the tire and the ground and the tire pressure from the tire image; and inputting the contact area and the corresponding tire pressure into a load model, and calculating the load of the tire.
In the embodiment, a vehicle weighing method based on image recognition is also disclosed, which comprises the following steps:
acquiring the load of each tire of the vehicle to be detected by the tire load acquisition method based on the image recognition;
and calculating the total weight of the vehicle to be tested according to the load of each tire of the vehicle to be tested.
In this embodiment, a computer system is further disclosed, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and is characterized in that the processor implements the steps of the above method embodiments when executing the computer program.
The tire load obtaining method, the vehicle weighing method and the system based on image recognition can quickly and accurately recognize the tire load and the vehicle weight of the vehicle to be detected, and compared with the prior art, the vehicle weighing method and the system have the advantages of wide measurement range, no need of any additional sensing equipment, no need of traffic closure, long-term stable work and easy information integration.
Example two:
the second embodiment is an expanded embodiment of the first embodiment, and is different from the first embodiment in that specific steps of a vehicle weighing method based on image recognition are refined, and a specific structure of a vehicle weighing system based on image recognition is refined:
the system comprises an acquisition component and a data processing component, wherein the acquisition component is used for acquiring tire images of all tires of a vehicle to be detected and sending the tire images of all the tires to the data processing component, and the data processing component is used for detecting the contact area and the tire pressure of all the tires with the ground from the tire images of all the tires; inputting the contact area of each tire and the corresponding tire pressure into a load model, calculating the load of each tire, and finally calculating the total weight of the vehicle to be tested according to the load of each tire of the vehicle to be tested.
In this embodiment, the acquisition component may be an image acquisition device or a video acquisition device, specifically may be a camera or a traffic camera, in this embodiment, a Nikon D5600 single lens reflex camera (hereinafter referred to as a camera) is selected, and the data processing component is a notebook computer.
In this embodiment, the workflow of the vehicle weighing system based on image recognition is as follows:
a first part: acquiring a tire image of each tire of a vehicle to be detected, and acquiring the load of each tire by using a tire load acquisition method based on image identification on the tire image of each tire:
step 1: obtaining the taken picture of the tire and carrying out image preprocessing
The method includes the steps of firstly carrying out equal ratio compression on a high-resolution tire image on the premise of ensuring that the tire image is subjected to lossless compression, carrying out related processing on the tire image and then restoring, thus reducing the calculated amount and improving the image processing efficiency
Scaling N times in the x and y directions according to the same proportion to obtain a new image
。
Wherein the content of the first and second substances,
for initial abscissa of transformed seed pixel in tire imageThe value of the one or more of,
is the initial abscissa value of the seed pixel in the original tire image,
is the initial ordinate value of the transformed seed pixel in the tire image,
is the initial ordinate value of the seed pixel in the original tire image,
is a multiple of the scaling.
For a tire image containing certain noise, situations such as unclear tire boundaries, small color differences, complex texture structures and the like easily occur. The direct tire image segmentation reduces the segmentation speed and has poor segmentation effect. And transforming the tire image into a frequency domain by adopting a wavelet threshold denoising method to perform image denoising treatment. The method comprises the steps of carrying out orthogonal wavelet transformation on a noisy tire image, selecting a proper wavelet basis and a proper wavelet decomposition layer, and obtaining a wavelet decomposition coefficient by applying an MATLAB (MATLAB is commercial mathematical software produced by MathWorks company in America) decomposition algorithm. Because the wavelet decomposition coefficients of the tire image and the noise on each wavelet hierarchical layer have different characteristics, the wavelet decomposition coefficient caused by the noise is smaller than a threshold lambda and is directly assigned as 0; wavelet decomposition coefficient caused by tire image is larger than thresholdλAnd can be retained.
Wherein the content of the first and second substances,
is a function of the wavelet decomposition coefficients,
for assigned wavelet scoreSolving the coefficient; wherein the threshold valueλSetting a value range within an interval (0, 1) according to actual requirements;
the wavelet threshold denoising method can effectively reduce sensitive noise in most tire images, enhance image contrast and obtain clear tire characteristic images. The preprocessed tire image obtained after the image is subjected to the equal scaling change and wavelet threshold denoising processing is shown in fig. 1.
Step 2: and acquiring the actual contact length, the actual contact area and the tire pressure of the target tire in the tire image.
And (3) assuming that the hub is a rigid body and is not deformed in the using process, and obtaining pixel points of the hub and the tire edge according to a JSEG algorithm and a region growing algorithm of color image segmentation.
The JSEG algorithm can effectively reduce the number of colors in an image, reducing a complex scene to tens of colors. The principle is as follows: z is the set of all N pixels in a region, assuming set Z is classified as class C,
；
is as follows
A collection of pixel points of a species of,
is composed of
The number of pixel points in the class, Z is the coordinate of the pixel point, Z = (x, y), Z ∈ Z, m represents the coordinate mean of N pixel points in the Z set,
is composed of
In class I
The mean value of the coordinates of the individual pixel points.
Is the coordinate variance of the N pixel points in the Z set,
is of class C
The sum of the variance of the pixel coordinates of the set is the class interpolation,
is an intermediate value used for calculating the local similarity of each pixel point.
And constructing J values under different sizes, and calculating the local similarity of each pixel point. When the J value is higher, the probability that the pixel point is close to the boundary of the region is higher. And obtaining different color quantization class mark maps of each pixel of a tire image through a JSEG algorithm, and performing J-value image segmentation on the class mark maps by using a region growing algorithm. In all tire images, pixels within the same sub-region have similar characteristics, and the pixel characteristics generally vary greatly from sub-region to sub-region. The region growing algorithm is to aggregate the same target region of interest with similar characteristics of pixels (color, mean gray value, texture, etc.) in the image. The steps are as followsThe method for judging the local average value is characterized in that 3 × 3 matrix traversal is adopted for the whole picture, the matrix center point with the minimum local image average value is taken as a growth starting point, the manual randomness and uncertainty interference can be effectively avoided, and the automation of the algorithm is realized
Is the initial position of the seed pixel,
is a seed point, and is a seed point,
is as follows
The coordinate values of the individual pixel points are calculated,
is the offset of the seed point.
The seed growth criterion is the absolute value of the difference between the pixel point value of the seed neighborhood and the seed pixel point value, and is compared with a threshold value T. When the value is smaller than the threshold value T, the considered neighborhood pixel and the seed pixel are merged into the same region. And when the value is equal to or larger than the threshold value T, searching the local maximum value of the gradient in the neighborhood along the region growing direction, and stopping growing the seed when the local maximum value is reached. The tire extraction result based on the JSEG algorithm and the region growing combined algorithm is shown in fig. 2.
And accurately separating the tire target object from the background to obtain pixel points of the hub and the tire edge. The actual physical length of the read hub diameter is shown in fig. 3, according to the truck tire specification mark displayed on the tire image. And obtaining the scale factor of the tire image according to the ratio of the physical length of the diameter of the hub to the image point of the hub. And acquiring coordinates of the lowest point of the contact surface of the tire and the peripheral edge point of the tire, and calculating the slope of the edge point and the lowest point of the contact surface. And obtaining left and right critical values of the effective contact edge points of the tire and the road surface by setting a threshold value, and calculating the number of pixels from the left and right critical values of the effective contact edge points to the lowest contact point of the tire by using the arc length. The actual contact length of the tire is finally obtained by multiplying the number of pixels by the tire scale factor as shown in fig. 4. The tire width is read according to the truck tire specification mark displayed by the tire image, and the peripheral area of the tire grounding surface can be obtained by multiplying the actual contact length of the tire by the tire width. The effective area of the tire is smaller than the peripheral area of the tire contacting with a road surface due to the pattern. According to the field actual measurement and finite element simulation, the reduction rate of the tire area is about 80%. So that the actual effective contact area of the tire
= actual tire contact length × tire width × area reduction rate.
The tire pressure of a target tire is obtained through the following steps:
identifying a tire specification mark on a target tire, reading a reference tire pressure in the tire specification mark, and multiplying the reference tire pressure by a preset correction coefficient to obtain an actual tire pressure value of the target tire, wherein the correction coefficient is preferably 1.2 in general, but the correction coefficient can also be obtained by the following steps:
actual measurement tire pressure data of tires with the same contact length and the same type as the target tire and corresponding reference tire pressure data are obtained from historical data, and the ratio between the actual measurement tire pressure data and the corresponding reference tire pressure data is calculated to serve as a correction coefficient of the target tire.
Step 3: obtaining the relationship among vertical load, grounding pressure and tire pressure of tire
Establishing a tire finite element model, simulating the average grounding pressure of the tire under the action of different vehicle loads and different tire pressures, thereby forming an orthogonal test with the vehicle load and the tire pressure as independent variables and the average grounding pressure of the tire as a dependent variable, and carrying out accuracy inspection through a field actual measurement test. The relationship among the tire air pressure, the axial load and the tire average ground contact pressure of part of the data is shown in FIG. 5.
And performing multiple regression on the vehicle load, the tire air pressure and the average ground contact pressure by using SPSS software, thereby obtaining the relationship among the tire inflation pressure, the load and the average ground contact pressure:
in the formula (I), the compound is shown in the specification,iis the first of the vehicle to be testediA shaft;
is the first of the vehicle to be testediThe average ground contact pressure of the shaft,
for the vehicle to be testediAxial load in KN;
for the vehicle to be testediAxle tire pressure in MPa;
for the vehicle to be testediContact area of the tire of the axle in units of
；
，
And
the regression coefficients are respectively a first regression coefficient, a second regression coefficient and a third regression coefficient, and the value ranges of the first regression coefficient, the second regression coefficient and the third regression coefficient are all in the (0, 1) interval.
After simple mathematical transformation of the formula (8), the result is
As can be seen from equation (9), after the effective contact area and the tire air pressure of the tire are obtained through computer vision, the vertical load of the tire can be obtained.
A second part: calculating the total weight of the vehicle to be tested according to the load of each tire of the vehicle to be tested
Step 4: obtaining vehicle axle weight and gross weight
After the vertical load of each tire is obtained, the vertical loads of the tires on the same axle are added and summed to obtain the axle weight of each axle.
After the axial load of each shaft of the vehicle is obtained, the total weight of the vehicle is obtained through accumulation and summation.
In the formula, N is the total number of axles of the vehicle;
an estimated value of the axial load of the vehicle is obtained;
is as followsiAn estimate of axle vehicle weight.
Step 5: cloud vehicle overload judgment and information early warning sending
After the vehicle weight is obtained through a weighing system for recognizing the tire contact length and the tire pressure through images, the vehicle weight is uploaded to a vehicle garage cloud. The data flow chart of the vehicle weight detection system is shown in fig. 6. Weight comparisons were made at the cloud with gross weight limits (table 1) specified for highway management of overrun transportation vehicles. And if the obtained weight is larger than the corresponding total weight limit value, immediately uploading the overweight information to a cloud end, and then timely sending overweight early warning information to people such as traffic departments, toll stations, truck owners and the like.
TABLE 1 overload Standard for management of over-limit transportation vehicles on roads
After the Step1-5 is subjected to algorithm integration and flowchart combing, a vehicle dynamic weighing algorithm for identifying the contact length and the tire pressure of the tire by the non-contact image is obtained and is shown in fig. 7, and a flow of a vehicle dynamic weighing system for identifying the contact length and the tire pressure of the tire by the whole non-contact image is shown in fig. 8.
In conclusion, the tire load obtaining method based on image recognition, the vehicle weighing method and the vehicle weighing system can effectively overcome the defect of contact type weighing. The system can accurately and truly acquire the weight information of the vehicle, automatically realize the functions of all-weather 24-hour uninterrupted vehicle weight data acquisition, judgment, early warning and the like, and provide evidences for off-site law enforcement, thereby effectively solving the practical problem of vehicle overload and overrun.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (9)
1. A tire load obtaining method based on image recognition is characterized by comprising the following steps:
obtaining a tire image of a tire to be detected; detecting the contact area of the tire and the ground and the tire pressure from the tire image; inputting the contact area and the corresponding tire pressure into a load model, and calculating the load of the tire;
wherein the contact area of the tire and the ground is detected from the tire image, and the method comprises the following steps:
separating a target tire from a background in the tire image to obtain pixel points of a hub of the target tire and the edge of the tire;
identifying a tire specification mark on a target tire, and reading the diameter of a hub and the width of the tire in the tire specification mark;
obtaining a scale factor of a tire image according to the ratio of the physical length of the diameter of the hub to the pixel point of the hub;
and calculating the number of pixels of the target tire in contact with the ground, calculating the contact length of the target tire and the ground according to the number of pixels and the scale factor, and calculating the contact area according to the contact length and the tire width.
2. The image recognition-based tire load obtaining method according to claim 1, wherein the target tire in the tire image is separated from the background, comprising the steps of:
scaling the tire image by adopting equal-scale transformation, and transforming the tire image into a frequency domain by adopting a wavelet threshold denoising algorithm to perform image denoising treatment to obtain a denoised tire image;
and separating the tire target from the background through a JSEG algorithm and a region growing algorithm of color image segmentation.
3. The image recognition-based tire load obtaining method according to claim 2, wherein the contact area is calculated from the contact length and the tire width, comprising the steps of:
area of contact
= actual tire contact length × tire width × area reduction rate.
4. The image recognition-based tire load obtaining method according to claim 3, wherein the tire pressure of the tire detected from the tire image is obtained by:
identifying the tire specification mark on the target tire, reading the reference tire pressure in the tire specification mark, and multiplying the reference tire pressure by a preset correction coefficient to obtain the actual tire pressure value of the target tire.
5. The image recognition-based tire load obtaining method according to any one of claims 1 to 4, wherein the load model is:
wherein the content of the first and second substances,iis the first of the vehicle to be testediA shaft;
for the vehicle to be testediAxial load in KN;
for the vehicle to be testediAxle tire pressure in MPa;
for the vehicle to be testediContact area of the tire of the axle in units of
；
，
And
the regression coefficients are respectively a first regression coefficient, a second regression coefficient and a third regression coefficient, and the value ranges of the first regression coefficient, the second regression coefficient and the third regression coefficient are all in the (0, 1) interval.
6. The method for acquiring the tire load based on the image recognition as claimed in claim 5, wherein the tire image of the tire to be measured is acquired by an image acquisition device/video acquisition device.
7. A vehicle weighing method based on image recognition is characterized by comprising the following steps:
acquiring the load of each tire of a vehicle to be tested by the tire load acquisition method based on image recognition according to any one of claims 1 to 6;
and calculating the total weight of the vehicle to be tested according to the load of each tire of the vehicle to be tested.
8. The image recognition-based vehicle weighing method of claim 7, further comprising the following steps after calculating the total weight of the vehicle under test according to the load of each tire of the vehicle under test:
identifying the category of the vehicle to be detected according to the tire image of the vehicle to be detected;
comparing the total weight of the vehicle to be tested with a total weight limit value corresponding to the category of the vehicle to be tested, judging that the vehicle to be tested is overweight when the total weight of the vehicle to be tested is greater than the corresponding total weight limit value, and outputting an alarm signal;
and sending the tire image of the vehicle to be tested, the vehicle weight and whether the vehicle is overweight to a cloud server for storage.
9. A computer system comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the steps of the method of any one of claims 1 to 8 are performed when the computer program is executed by the processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010874283.8A CN111735524B (en) | 2020-08-27 | 2020-08-27 | Tire load obtaining method based on image recognition, vehicle weighing method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010874283.8A CN111735524B (en) | 2020-08-27 | 2020-08-27 | Tire load obtaining method based on image recognition, vehicle weighing method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111735524A true CN111735524A (en) | 2020-10-02 |
| CN111735524B CN111735524B (en) | 2020-11-24 |
Family
ID=72658800
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010874283.8A Active CN111735524B (en) | 2020-08-27 | 2020-08-27 | Tire load obtaining method based on image recognition, vehicle weighing method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111735524B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112614351A (en) * | 2020-12-29 | 2021-04-06 | 芜湖哈特机器人产业技术研究院有限公司 | Truck overload detection system and method based on machine vision |
| CN112767471A (en) * | 2021-01-04 | 2021-05-07 | 湖南大学 | Tire ground contact area measuring method and device based on image feature extraction |
| CN112861401A (en) * | 2021-02-03 | 2021-05-28 | 湖南大学 | Vehicle weight identification method, device, equipment and storage medium based on simulation analysis |
| CN113212073A (en) * | 2021-05-17 | 2021-08-06 | 中国第一汽车股份有限公司 | Tire pressure adjusting method and system based on vehicle-mounted camera and automobile |
| IT202100008822A1 (en) * | 2021-04-08 | 2022-10-08 | Iveco Spa | WEIGHING SYSTEM AND WEIGHING METHOD FOR A VEHICLE |
| CN115311691A (en) * | 2022-10-12 | 2022-11-08 | 山东圣点世纪科技有限公司 | Joint identification method based on wrist vein and wrist texture |
| CN115497058A (en) * | 2022-09-02 | 2022-12-20 | 东南大学 | Non-contact vehicle weighing method based on multispectral imaging technology |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009047629A (en) * | 2007-08-22 | 2009-03-05 | Yamato Scale Co Ltd | Axle load measuring device |
| CN105678768A (en) * | 2016-01-08 | 2016-06-15 | 杭州电子科技大学 | Machine vision-based tire tread detection method |
| CN106515621A (en) * | 2016-10-20 | 2017-03-22 | 深圳市体数科科技有限公司 | Method and system for monitoring automobile load based on intelligent tire pressure gauge |
| US20180249058A1 (en) * | 2017-02-28 | 2018-08-30 | Panasonic Intellectual Property Management Co., Ltd. | Imaging apparatus, axle load measuring system, and imaging method |
| CN109696133A (en) * | 2017-10-24 | 2019-04-30 | 柯尼卡美能达株式会社 | Squeegee action device for calculating and its method and overload detection system |
| CN109781228A (en) * | 2019-01-30 | 2019-05-21 | 内蒙古大学 | Car load determines system and method |
-
2020
- 2020-08-27 CN CN202010874283.8A patent/CN111735524B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009047629A (en) * | 2007-08-22 | 2009-03-05 | Yamato Scale Co Ltd | Axle load measuring device |
| CN105678768A (en) * | 2016-01-08 | 2016-06-15 | 杭州电子科技大学 | Machine vision-based tire tread detection method |
| CN106515621A (en) * | 2016-10-20 | 2017-03-22 | 深圳市体数科科技有限公司 | Method and system for monitoring automobile load based on intelligent tire pressure gauge |
| US20180249058A1 (en) * | 2017-02-28 | 2018-08-30 | Panasonic Intellectual Property Management Co., Ltd. | Imaging apparatus, axle load measuring system, and imaging method |
| CN109696133A (en) * | 2017-10-24 | 2019-04-30 | 柯尼卡美能达株式会社 | Squeegee action device for calculating and its method and overload detection system |
| CN109781228A (en) * | 2019-01-30 | 2019-05-21 | 内蒙古大学 | Car load determines system and method |
Non-Patent Citations (2)
| Title |
|---|
| 张俊杰: "基于CCD的汽车轮胎图像特征分析系统研制", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
| 祝磊 等: "基于机器视觉的轮胎胎面检测系统设计与实现", 《计算机工程与设计》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112614351A (en) * | 2020-12-29 | 2021-04-06 | 芜湖哈特机器人产业技术研究院有限公司 | Truck overload detection system and method based on machine vision |
| CN112614351B (en) * | 2020-12-29 | 2021-10-26 | 芜湖哈特机器人产业技术研究院有限公司 | Truck overload detection system and method based on machine vision |
| CN112767471B (en) * | 2021-01-04 | 2024-03-29 | 湖南大学 | Tire ground contact area measuring method and device based on image feature extraction |
| CN112767471A (en) * | 2021-01-04 | 2021-05-07 | 湖南大学 | Tire ground contact area measuring method and device based on image feature extraction |
| CN112861401A (en) * | 2021-02-03 | 2021-05-28 | 湖南大学 | Vehicle weight identification method, device, equipment and storage medium based on simulation analysis |
| CN112861401B (en) * | 2021-02-03 | 2022-04-19 | 湖南大学 | Vehicle weight identification method, device, equipment and storage medium based on simulation analysis |
| IT202100008822A1 (en) * | 2021-04-08 | 2022-10-08 | Iveco Spa | WEIGHING SYSTEM AND WEIGHING METHOD FOR A VEHICLE |
| EP4071449A1 (en) * | 2021-04-08 | 2022-10-12 | Iveco S.P.A. | Weighting system and weighting method for a vehicle |
| CN113212073A (en) * | 2021-05-17 | 2021-08-06 | 中国第一汽车股份有限公司 | Tire pressure adjusting method and system based on vehicle-mounted camera and automobile |
| CN115497058A (en) * | 2022-09-02 | 2022-12-20 | 东南大学 | Non-contact vehicle weighing method based on multispectral imaging technology |
| CN115497058B (en) * | 2022-09-02 | 2023-09-01 | 东南大学 | Non-contact vehicle weighing method based on multispectral imaging technology |
| CN115311691B (en) * | 2022-10-12 | 2023-02-28 | 山东圣点世纪科技有限公司 | Joint identification method based on wrist vein and wrist texture |
| CN115311691A (en) * | 2022-10-12 | 2022-11-08 | 山东圣点世纪科技有限公司 | Joint identification method based on wrist vein and wrist texture |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111735524B (en) | 2020-11-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111735524B (en) | Tire load obtaining method based on image recognition, vehicle weighing method and system | |
| Ouma et al. | Pothole detection on asphalt pavements from 2D-colour pothole images using fuzzy c-means clustering and morphological reconstruction | |
| Akagic et al. | Pavement crack detection using Otsu thresholding for image segmentation | |
| Yin et al. | Hot region selection based on selective search and modified fuzzy C-means in remote sensing images | |
| CN111274843B (en) | Truck overload monitoring method and system based on monitoring video | |
| CN109840483B (en) | Landslide crack detection and identification method and device | |
| CN110633690B (en) | Vehicle feature identification method and system based on bridge monitoring | |
| CN108960055B (en) | Lane line detection method based on local line segment mode characteristics | |
| CN114998852A (en) | Intelligent detection method for road pavement diseases based on deep learning | |
| CN108760740A (en) | A kind of pavement skid resistance condition rapid detection method based on machine vision | |
| CN105049790A (en) | Video monitoring system image acquisition method and apparatus | |
| CN110070545B (en) | Method for automatically extracting urban built-up area by urban texture feature density | |
| CN110717886A (en) | Pavement pool detection method based on machine vision in complex environment | |
| CN112307989B (en) | Road surface object identification method, device, computer equipment and storage medium | |
| Tiwari et al. | Roadcare: a deep-learning based approach to quantifying road surface quality | |
| CN113221759A (en) | Road scattering identification method and device based on anomaly detection model | |
| Liaw et al. | Using sharpness image with haar function for urban atmospheric visibility measurement | |
| Doycheva et al. | Computer vision and deep learning for real-time pavement distress detection | |
| CN114332644A (en) | Large-view-field traffic density acquisition method based on video satellite data | |
| Al-Suleiman et al. | Assessment of the effect of alligator cracking on pavement condition using WSN-image processing | |
| Li et al. | An unsupervised concrete crack detection method based on nnU-Net | |
| CN115546762A (en) | Image clustering method, device, storage medium and server | |
| CN109934103A (en) | Method based on obvious object in dark channel prior and region covariance detection image | |
| CN114677428A (en) | Power transmission line icing thickness detection method based on unmanned aerial vehicle image processing | |
| CN112461345A (en) | Truck scale rolling line out-of-bounds detection method based on LSD (least squares distortion) linear detection algorithm |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |