CN117746165A - Method and device for identifying tire types of wheel type excavator - Google Patents
Method and device for identifying tire types of wheel type excavator Download PDFInfo
- Publication number
- CN117746165A CN117746165A CN202410134393.9A CN202410134393A CN117746165A CN 117746165 A CN117746165 A CN 117746165A CN 202410134393 A CN202410134393 A CN 202410134393A CN 117746165 A CN117746165 A CN 117746165A
- Authority
- CN
- China
- Prior art keywords
- tire
- pattern
- image
- threshold
- edge
- 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
- 238000000034 method Methods 0.000 title claims abstract description 50
- 230000011218 segmentation Effects 0.000 claims abstract description 24
- 230000007812 deficiency Effects 0.000 claims abstract description 19
- 238000001514 detection method Methods 0.000 claims abstract description 14
- 238000003708 edge detection Methods 0.000 claims abstract description 14
- 230000001502 supplementing effect Effects 0.000 claims abstract description 9
- 238000004590 computer program Methods 0.000 claims description 18
- 230000006870 function Effects 0.000 claims description 8
- 238000004806 packaging method and process Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 8
- 238000004422 calculation algorithm Methods 0.000 claims description 7
- 238000003860 storage Methods 0.000 claims description 7
- 230000000007 visual effect Effects 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000009499 grossing Methods 0.000 claims description 5
- 239000013072 incoming material Substances 0.000 claims description 5
- 230000003044 adaptive effect Effects 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 4
- 238000009826 distribution Methods 0.000 claims description 3
- 230000008030 elimination Effects 0.000 claims description 3
- 238000003379 elimination reaction Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000003706 image smoothing Methods 0.000 claims description 3
- 230000009467 reduction Effects 0.000 claims description 3
- 230000001629 suppression Effects 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 230000014759 maintenance of location Effects 0.000 claims description 2
- 239000000047 product Substances 0.000 claims 1
- 239000013589 supplement Substances 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 8
- 238000004519 manufacturing process Methods 0.000 description 7
- 239000000463 material Substances 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000005286 illumination Methods 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 230000003139 buffering effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 238000003702 image correction Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000035939 shock Effects 0.000 description 1
- 230000000087 stabilizing effect Effects 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a method and a device for identifying the tire type of a wheel type excavator, wherein the method comprises the steps of establishing a tire image retrieval template according to an acquired tire sidewall image; searching and scoring a template classification library established in advance through a tire image searching template, outputting left and right tire information and a matching score result, and obtaining a tire type recognition result according to the matching score result; the template classification library establishment comprises the following steps: according to the obtained tire pattern patterns of different models, threshold segmentation of the tire pattern characteristics is carried out; edge detection is carried out according to the pattern features segmented by the threshold value, secondary self-adaptive detection is carried out for the light supplementing deficiency of the pattern edge, and the deficiency edge information is fitted; the invention can quickly identify the type of the tire required before the tire assembly of the wheel type excavator and whether the tire is consistent with the tire required by the machine type or not.
Description
Technical Field
The invention relates to a method and a device for identifying the tire type of a wheel type excavator, and belongs to the field of production and assembly of excavating machinery.
Background
The wheel type excavator is an earth moving machine which is driven by wheels, excavates materials higher or lower than a carrying plane by a bucket, and loads the materials into a transport vehicle or unloads the materials to a storage yard. The wheel excavator is an excavator using tires as traveling members, and in recent years, the wheel excavator has been improved in market share year by virtue of the advantages of flexibility in maneuvering, convenience in transferring, high traveling speed, less damage to the ground surface, high shock absorbing and buffering functions, and the like. The tyre is an important component for bearing gravity, transmitting traction force, braking force and steering force of the excavator and bearing the reaction force of the road surface. Therefore, the corresponding tires are selected for different types of excavators, and the mounting of the corresponding pattern tires on the left and right hubs is important for improving the assembly efficiency of the wheel type excavator and stabilizing the working state of the whole excavator.
In the prior art, the sidewalls of the tire contain numerical and alphabetic information, including the size of the tire, the type of tire, and safety standards.
(1) At present, in the tire assembly process of a wheel type excavator, a manual selection mode is mostly adopted, namely, the type of a tire required by a production model is determined and judged by visual tire side information of the tire.
(2) CN202111633297.1 proposes to correct the tire side image, and to determine the tire type by fitting the tire side image to a ring image and converting the ring image to polar coordinates for tire text information recognition.
The prior art has the following defects:
(1) In the actual production process, the tire type of a corresponding model is determined by manually and visually observing the tire surface information, so that the related information is needed to be searched for in a circle of the tire, the time and the labor are consumed, the error is easy to occur, and the subsequent assembly production efficiency is low.
(2) The tire sidewall image correction method limits the tire information-containing part to the visual acquisition side surface, manual interference is needed, and part of tire sidewall characters do not contain the tire left and right information. And secondly, correcting the image for many times increases the running time of the recognition algorithm, and partial information is lost or the recognition is wrong when the image is converted into the polar coordinates to carry out text recognition.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provides a method and a device for identifying the tire type of a wheel excavator, which can quickly identify whether the type and the left and right of the tire type required by the wheel excavator before the tire assembly are consistent with those required by a machine type.
In order to achieve the above purpose, the invention is realized by adopting the following technical scheme:
in a first aspect, the present invention provides a method for identifying a tire type of a wheel excavator, comprising:
building a tire image retrieval template according to the obtained tire sidewall image;
searching and scoring a template classification library established in advance through a tire image searching template, outputting left and right tire information and a matching score result, and obtaining a tire type recognition result according to the matching score result;
the template classification library establishment comprises the following steps:
according to the obtained tire pattern patterns of different models, threshold segmentation of the tire pattern characteristics is carried out;
edge detection is carried out according to the pattern features segmented by the threshold value, secondary self-adaptive detection is carried out for the light supplementing deficiency of the pattern edge, and the deficiency edge information is fitted;
and packaging tire pattern patterns of different models as templates, and establishing a template classification library through the packaged templates. When the tire sidewall image is obtained, a monocular camera is arranged on one side of the tire sidewall, camera calibration by a Zhang Zhengyou method is carried out, and camera internal references are determined; after the monocular camera calibration is completed, the monocular camera is used for collecting tire sidewall images.
Further, the obtained tire sidewall image is preprocessed, and noise information on the tire surface is reduced by adopting an image smoothing method for improving median filtering, which specifically comprises the following steps:
setting the grouping mode groups the 3×3 window arrangement pixels in units of rows, including: the method comprises the steps of selecting a maximum value group Max, a median value group Mid and a minimum value group Min, wherein each group has three groups of values, and taking attribute values in the three groups of values;
and comparing the three groups of value data, excluding the maximum value and the median value in Max, the minimum value and the median value in Min, and the maximum value and the minimum value in Mid, and searching the median value in the rest data to finish pretreatment.
Further, the threshold segmentation of the tire tread pattern features is performed according to the obtained tire tread patterns of different models, including:
determining an ROI (region of interest) area according to the camera visual field range and the tire incoming material fixed area;
setting a segmentation threshold point in the region, wherein the segmentation threshold point meets the tire pattern segmentation requirement, and the operation expression is as follows:
S={(r,c)∈R|g min ≤f r,c ≤g max }
wherein S represents a set of gray values of the input image; r represents pixel row coordinates; c represents pixel column coordinates; r represents a pixel point set; f (f) r,c Representing the gray value of the input image; g min Representing threshold segmentation gray scaleA minimum value; g max Representing a threshold split gray maximum.
Further, the edge detection of the pattern features segmented according to the threshold includes:
the robust Canny algorithm is adopted to search the pattern edge:
the first step of edge information noise elimination, according to the characteristic that the edge of the pattern is from dark to bright, a Gaussian smoothing filter with size=3 is adopted for convolution noise reduction, and the kernel is as follows:
wherein k represents a gaussian filter mask kernel;
smoothing the mask to a threshold segmentation ROI region;
secondly, calculating the gradient amplitude direction, wherein the processes from 0 to 1 are operated by adopting a Sobel filter, and the two directions of the convolution array function x and y are respectively as follows:
wherein G is x Representing gradient of image change in x direction, G y Representing the gradient of the image in the y-direction;
after the convolution is finished, gradient amplitude and direction calculation is carried out, and the gradient amplitude and direction calculation is carried out according to the convolution result:
wherein G represents the amplitude variation gradient of the input image;
the direction is determined by calculating the angle value theta:
according to the image display of the incoming tire picture, different tire pattern shapes are respectively searched and judged by 0 degree, 45 degrees, 90 degrees and 135 degrees;
thirdly, non-maximum value suppression is carried out, edge pixel extraction is guaranteed, and edge amplitude and direction are reserved through setting area distribution values, so that edge maximum values are obtained;
and fourthly, hysteresis thresholds are set, namely a lowest threshold and a highest threshold are set, the classification group with the suppressed maximum value is stored and removed according to a high threshold interval and a low threshold interval, pixels higher than the high threshold are reserved, pixels lower than the low threshold are removed, and hysteresis retention close to the high threshold in the two-threshold pixel interval is performed.
Further, the performing a secondary adaptive detection for the light supplementing deficiency of the pattern edge, when fitting the missing edge information, performing the secondary adaptive detection by adopting straight line fitting, includes:
firstly, carrying out accumulated probability Hough transformation to obtain linear endpoint information;
and secondly, taking the endpoint group as input, performing straight line fitting by using a RANSAC fitting algorithm, and judging the intersection point as an edge endpoint.
Further, the step of packaging tire patterns of different models as templates, and establishing a template classification library through the packaged templates comprises the following steps:
taking the packed tire patterns as a search mode according to the model-left and right branch structures, taking the model as a row input item for different tires and templates packed at the left and right sides of the tires according to different models, and taking the left and right sides of the tires as a column input item to establish a classification library of n x 2, wherein n is the type of a wheel excavator product.
Further, the method further comprises: left and right tire information, a tire type identification result, and a tire sidewall image are recorded, and the tire type identification result and the left and right tire information are output.
In a second aspect, the present invention provides a tire type identification device for a wheel excavator, comprising:
the template building module is used for building a tire image retrieval template according to the obtained tire sidewall image;
the identification module is used for searching and scoring a template classification library established in advance through a tire image searching template, outputting left and right tire information and a matching score result, and obtaining a tire type identification result according to the matching score result, wherein the establishment of the template classification library comprises the following steps:
according to the obtained tire pattern patterns of different models, threshold segmentation of the tire pattern characteristics is carried out;
edge detection is carried out according to the pattern features segmented by the threshold value, secondary self-adaptive detection is carried out for the light supplementing deficiency of the pattern edge, and the deficiency edge information is fitted;
and packaging tire pattern patterns of different models as templates, and establishing a template classification library through the packaged templates.
When the tire sidewall image is obtained, a monocular camera is arranged on one side of the tire sidewall, camera calibration by a Zhang Zhengyou method is carried out, and camera internal references are determined; after the monocular camera calibration is completed, the monocular camera is used for collecting tire sidewall images.
In a third aspect, the present invention provides a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of any of the methods described in the preceding claims.
In a fourth aspect, the invention provides a computer program product comprising computer programs/instructions which when executed by a processor implement the steps of the method of any of the preceding claims.
In a fourth aspect, the present invention provides a computer apparatus/device/system comprising:
a memory for storing computer programs/instructions;
a processor for executing the computer program/instructions to implement the steps of the method of any one of the preceding claims.
Compared with the prior art, the invention has the beneficial effects that:
(1) The readable pattern image of the incoming tire is used as input data, so that the type and the left and right discriminating work of manually identifying the incoming tire are simplified, the incoming tire information can be reserved, and the traceability of the material checking process can be realized.
(2) The camera is fixedly arranged on the sidewall of the incoming tire, so that the position interference of the manual work on the incoming tire is reduced, and the automatic flow of the visual identification device is enhanced.
(3) The improved image preprocessing, binarization segmentation and pattern edge detection are adopted, so that the pattern image data has stronger robustness and representativeness in practical application.
(4) The real-time score comparison of the types and the left and right information of the tires is realized by constructing a crotch template classification library at the output end, the accuracy of judging the incoming materials of the tires in the actual assembly production process is realized, and the working efficiency of assembling the tires is improved.
Drawings
FIG. 1 is a flow chart of a method for identifying tire type of a wheel excavator provided by an embodiment of the present invention;
fig. 2 is a schematic diagram of a data transmission flow provided in an embodiment of the present invention;
fig. 3 is a schematic diagram of a hardware installation location according to an embodiment of the present invention.
In the figure: 1. a camera fixing device; 2. a monocular camera; 21. a signal output interface; 3. an annular aperture; 4. tire tread; 5. a signal retrieval interface; 6. and displaying the data on an analysis platform.
Detailed Description
The invention is further described below with reference to the accompanying drawings. The following examples are only for more clearly illustrating the technical aspects of the present invention, and are not intended to limit the scope of the present invention.
Example 1
The embodiment introduces a method for identifying the tire type of a wheel excavator, which comprises the following steps:
building a tire image retrieval template according to the obtained tire sidewall image;
searching and scoring a template classification library established in advance through a tire image searching template, outputting left and right tire information and a matching score result, and obtaining a tire type recognition result according to the matching score result;
the template classification library establishment comprises the following steps:
according to the obtained tire pattern patterns of different models, threshold segmentation of the tire pattern characteristics is carried out;
edge detection is carried out according to the pattern features segmented by the threshold value, secondary self-adaptive detection is carried out for the light supplementing deficiency of the pattern edge, and the deficiency edge information is fitted;
and packaging tire pattern patterns of different models as templates, and establishing a template classification library through the packaged templates.
As shown in fig. 1, the application process of the method for identifying the tire type of the wheel excavator provided in the embodiment specifically involves the following steps:
step 1: and (5) installing an identification device and calibrating the camera. The mounting and connection of the identification device is performed, as shown in fig. 3, the identification device including: the camera fixing device 1, the monocular camera 2, the signal output interface 21, the annular diaphragm 3, the tire pattern 4, the signal retrieval interface 5 and the data display analysis platform 6, wherein the monocular camera 2 is installed on one side of the tire sidewall through the camera fixing device 1 for camera calibration of Zhang Zhengyou method, camera internal parameters are determined, the annular diaphragm 3 for collecting images and the signal output interface 21 are arranged on the monocular camera 2, and the data display analysis platform 6 is in communication connection with the signal output interface 21 on the monocular camera 2 through the signal retrieval interface 5, and the equipment operation flow is shown in figure 2.
Step 2: and (5) offline image acquisition and image preprocessing. After the monocular camera calibration is completed, tire sidewall images are collected, noise information on the tire surface is reduced by adopting an image smoothing method of improving median filtering, and the image input effect is enhanced.
Setting the grouping mode groups the 3×3 window arrangement pixels in units of rows, including: the maximum value group Max, the median value group Mid and the minimum value group Min, wherein three groups of values exist in each group, and attribute values in the three groups of values are taken. Examples: max (Max) x =Max[p 0 p 1 p 2 ]The method comprises the steps of carrying out a first treatment on the surface of the And comparing the three groups of value data, excluding the maximum value and the median value in Max, the minimum value and the median value in Min, and the maximum value and the minimum value in Mid, and searching the median value in the rest data, thereby improving the operation efficiency.
Step 3: and establishing a template and generating a retrieval classification library. According to the characteristics of tire patterns of different types, the tire patterns are packaged into a template classification library:
(1) Threshold segmentation is carried out according to the pattern features of the patterns; and determining the region of interest (ROI) according to the camera visual field range and the tire incoming material fixed region. Setting a division threshold point in the region, the division threshold point satisfying a tire pattern division requirement, the operation expression s= { (R, c) ∈r|g min ≤f r,c ≤g max Default set to 100 according to scene reality; wherein S represents a set of gray values of the input image; r represents pixel row coordinates; c represents pixel column coordinates; r represents a pixel point set; f (f) r,c Representing the gray value of the input image; g min Representing a threshold segmentation gray minimum; g max Representing a threshold split gray maximum.
(2) Edge detection is carried out according to the pattern features segmented by the threshold value; in order to eliminate the influence of illumination change on the pattern threshold region, a robust Canny algorithm is adopted to search the pattern edge: the first step of edge information noise elimination, according to the characteristic that the edge of the pattern is from dark to bright, a Gaussian smoothing filter with size=3 is adopted for convolution noise reduction, and the kernel is as follows:
wherein k represents a gaussian filter mask kernel;
the mask is smoothed to a threshold segmentation ROI area. Secondly, calculating the gradient amplitude direction, wherein the processes from 0 to 1 are operated by adopting a Sobel filter, and the two directions of the convolution array function x and y are respectively as follows:
wherein G is x Representing gradient of image change in x direction, G y Representing the gradient of the image in the y-direction;
after the convolution is finished, gradient amplitude and direction calculation is carried out, and the gradient amplitude and direction calculation is carried out according to the convolution result:
wherein G represents the amplitude variation gradient of the input image;
the direction is determined by calculating the angle value theta:
according to the image display of the incoming tire picture, different tire patterns and shapes are different, and 0 degree, 45 degrees, 90 degrees and 135 degrees are used for searching and judging respectively. After the amplitude and the direction are obtained, a third step of non-maximum suppression is needed to ensure the extraction of edge pixels. And reserving the edge amplitude and direction by setting the area distribution value to obtain the edge maximum value. And setting a lowest threshold and a highest threshold according to the fourth step hysteresis threshold, and storing and removing the classification group with the suppressed maximum value according to the high and low threshold intervals. Pixels higher than the high threshold value are reserved, pixels lower than the low threshold value are removed, hysteresis of the pixels between the two threshold values and close to the high threshold value is reserved, and in the scheme, the connection threshold value is set to be 10 pixels according to pattern recognition characteristics.
(3) Performing secondary self-adaptive detection on pattern edge light supplementing deficiency, and fitting the deficiency edge information; the field illumination mode is fixed illumination, and when the deviation occurs between the incoming tire and the reference position, the edge of the pattern cannot be completely illuminated by the light source, so that the edge detection cannot form a complete pattern image. In order to enhance the edge information, secondary self-adaptive detection is added after edge detection, edge missing information is fitted, and the matching score of the edge image is improved. According to tire pattern missing examples, most of the tire pattern missing examples are linear connection missing, and linear fitting is adopted for secondary self-adaptive detection:
firstly, carrying out accumulated probability Hough transformation to obtain linear endpoint information;
and secondly, taking the endpoint group as input, performing straight line fitting by using a RANSAC fitting algorithm, and judging the intersection point as an edge endpoint.
(4) And setting a template classification library, and taking the packaged tire patterns as a retrieval mode according to the model-left and right branch structures. According to different models, models are used as row input items for different tires and templates packaged on the left and the right of the tires, the left and the right of the tires are used as column input items, a classification library of n x 2 is established, a comparison search target of the classification library is made for on-site images input in real time, and the real-time identification NG rate is reduced, wherein n is the type of a wheel excavator product.
Step 4: when the incoming tire is identified, the captured tire image is built into a retrieval template, the branch structure of the template classification library is retrieved, and a score judgment mechanism is set. The mechanism adopts a step-by-step grading and screening mode to obtain grading comparison information of each comparison template, and finally outputs a matching grading result and model and left and right tire information.
Step 5: the type and the left and right tire information are recorded and fed back. The target captured image data is recorded and the tire type and left and right text information are output.
The technical scheme of the embodiment has the following beneficial effects:
(1) The readable pattern image of the incoming tire is used as input data, so that the type and the left and right discriminating work of manually identifying the incoming tire are simplified, the incoming tire information can be reserved, and the traceability of the material checking process can be realized.
(2) The camera is fixedly arranged on the sidewall of the incoming tire, so that the position interference of the manual work on the incoming tire is reduced, and the automatic flow of the visual identification device is enhanced.
(3) The improved image preprocessing, binarization segmentation and pattern edge detection are adopted, so that the pattern image data has stronger robustness and representativeness in practical application.
(4) The real-time score comparison of the types and the left and right information of the tires is realized by constructing a crotch template classification library at the output end, the accuracy of judging the incoming materials of the tires in the actual assembly production process is realized, and the working efficiency of assembling the tires is improved.
Example 2
The embodiment provides a recognition device of wheel type excavator tire type, includes:
the template building module is used for building a tire image retrieval template according to the obtained tire sidewall image;
the identification module is used for searching and scoring a template classification library established in advance through a tire image searching template, outputting left and right tire information and a matching score result, and obtaining a tire type identification result according to the matching score result, wherein the establishment of the template classification library comprises the following steps:
according to the obtained tire pattern patterns of different models, threshold segmentation of the tire pattern characteristics is carried out;
edge detection is carried out according to the pattern features segmented by the threshold value, secondary self-adaptive detection is carried out for the light supplementing deficiency of the pattern edge, and the deficiency edge information is fitted;
and packaging tire pattern patterns of different models as templates, and establishing a template classification library through the packaged templates.
When the tire sidewall image is obtained, a monocular camera is arranged on one side of the tire sidewall, camera calibration by a Zhang Zhengyou method is carried out, and camera internal references are determined; after the monocular camera calibration is completed, the monocular camera is used for collecting tire sidewall images.
Example 3
The present embodiment provides a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the method of any of embodiment 1.
Example 4
The present embodiments provide a computer program product comprising a computer program/instruction which, when executed by a processor, implements the steps of the method of any of embodiment 1.
Interpretation of related terms
ROI-Region of Interest (region of interest), in machine vision, image processing, the region to be processed is outlined from the processed image in the form of a square, circle, ellipse, irregular polygon, etc., and is called the region of interest.
The foregoing is merely a preferred embodiment of the present invention, and it should be noted that modifications and variations could be made by those skilled in the art without departing from the technical principles of the present invention, and such modifications and variations should also be regarded as being within the scope of the invention.
It will be appreciated by those skilled in the art that embodiments of the present disclosure may be provided as a method, system, or computer program product. Accordingly, the present disclosure may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present disclosure may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
The present disclosure is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the disclosure. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Finally, it should be noted that: the foregoing embodiments are merely for illustrating the technical solution of the present disclosure and not for limiting the scope thereof, and although the present disclosure has been described in detail with reference to the foregoing embodiments, it will be understood by those skilled in the art that various changes, modifications or equivalents may be made to the specific embodiments of the invention after reading the present disclosure, and these changes, modifications or equivalents are within the scope of the claims appended hereto.
Claims (10)
1. A method for identifying a tire type of a wheel excavator, comprising:
building a tire image retrieval template according to the obtained tire sidewall image;
searching and scoring a template classification library established in advance through a tire image searching template, outputting left and right tire information and a matching score result, and obtaining a tire type recognition result according to the matching score result;
the template classification library establishment comprises the following steps:
according to the obtained tire pattern patterns of different models, threshold segmentation of the tire pattern characteristics is carried out;
edge detection is carried out according to the pattern features segmented by the threshold value, secondary self-adaptive detection is carried out for the light supplementing deficiency of the pattern edge, and the deficiency edge information is fitted;
and packaging tire pattern patterns of different models as templates, and establishing a template classification library through the packaged templates.
When the tire sidewall image is obtained, a monocular camera is arranged on one side of the tire sidewall, camera calibration by a Zhang Zhengyou method is carried out, and camera internal references are determined; after the monocular camera calibration is completed, the monocular camera is used for collecting tire sidewall images.
2. The method for identifying the tire type of the wheel excavator according to claim 1, wherein the obtained tire side image is preprocessed, and the noise information of the tire surface is reduced by adopting an image smoothing method of improving median filtering, and the method specifically comprises the following steps:
setting the grouping mode groups the 3×3 window arrangement pixels in units of rows, including: the method comprises the steps of selecting a maximum value group Max, a median value group Mid and a minimum value group Min, wherein each group has three groups of values, and taking attribute values in the three groups of values;
and comparing the three groups of value data, excluding the maximum value and the median value in Max, the minimum value and the median value in Min, and the maximum value and the minimum value in Mid, and searching the median value in the rest data to finish pretreatment.
3. The method for identifying the tire type of the wheel excavator according to claim 1, wherein the threshold segmentation of the tire pattern features according to the obtained tire pattern patterns of different models comprises:
determining an ROI (region of interest) area according to the camera visual field range and the tire incoming material fixed area;
setting a segmentation threshold point in the region, wherein the segmentation threshold point meets the tire pattern segmentation requirement, and the operation expression is as follows:
S={(r,c)∈R|g min ≤f r,c ≤g max }
wherein S represents a set of gray values of the input image; r represents pixel row coordinates; c represents pixel column coordinates; r represents a pixel point set; f (f) r,c Representing the gray value of the input image; g min Representing a threshold segmentation gray minimum; g max Representing a threshold split gray maximum.
4. The method for identifying the tire type of the wheel excavator according to claim 1, wherein the edge detection is performed by the pattern features segmented according to the threshold value, comprising:
the robust Canny algorithm is adopted to search the pattern edge:
the first step of edge information noise elimination, according to the characteristic that the edge of the pattern is from dark to bright, a Gaussian smoothing filter with size=3 is adopted for convolution noise reduction, and the kernel is as follows:
wherein k represents a gaussian filter mask kernel;
smoothing the mask to a threshold segmentation ROI region;
secondly, calculating the gradient amplitude direction, wherein the processes from 0 to 1 are operated by adopting a Sobel filter, and the two directions of the convolution array function x and y are respectively as follows:
wherein G is x Representing gradient of image change in x direction, G y Representing the gradient of the image in the y-direction;
after the convolution is finished, gradient amplitude and direction calculation is carried out, and the gradient amplitude and direction calculation is carried out according to the convolution result:
wherein G represents the amplitude variation gradient of the input image;
the direction is determined by calculating the angle value theta:
according to the image display of the incoming tire picture, different tire pattern shapes are respectively searched and judged by 0 degree, 45 degrees, 90 degrees and 135 degrees;
thirdly, non-maximum value suppression is carried out, edge pixel extraction is guaranteed, and edge amplitude and direction are reserved through setting area distribution values, so that edge maximum values are obtained;
and fourthly, hysteresis thresholds are set, namely a lowest threshold and a highest threshold are set, the classification group with the suppressed maximum value is stored and removed according to a high threshold interval and a low threshold interval, pixels higher than the high threshold are reserved, pixels lower than the low threshold are removed, and hysteresis retention close to the high threshold in the two-threshold pixel interval is performed.
5. The method for identifying tire type of wheel excavator according to claim 1, wherein the performing the secondary adaptive detection for the light supplement deficiency of the pattern edge, when fitting the deficiency edge information, performing the secondary adaptive detection by adopting straight line fitting, comprises:
firstly, carrying out accumulated probability Hough transformation to obtain linear endpoint information;
and secondly, taking the endpoint group as input, performing straight line fitting by using a RANSAC fitting algorithm, and judging the intersection point as an edge endpoint.
6. The method for identifying tire type of wheel excavator according to claim 1, wherein the steps of packaging tire patterns of different models as templates, and creating a template classification library by the packaged templates comprise:
taking the packed tire patterns as a search mode according to the model-left and right branch structures, taking the model as a row input item for different tires and templates packed at the left and right sides of the tires according to different models, and taking the left and right sides of the tires as a column input item to establish a classification library of n x 2, wherein n is the type of a wheel excavator product.
7. The method for identifying a tire type of a wheel excavator of claim 1, wherein the method further comprises: left and right tire information, a tire type identification result, and a tire sidewall image are recorded, and the tire type identification result and the left and right tire information are output.
8. A wheel excavator tire type identification device comprising:
the template building module is used for building a tire image retrieval template according to the obtained tire sidewall image;
the identification module is used for searching and scoring a template classification library established in advance through a tire image searching template, outputting left and right tire information and a matching score result, and obtaining a tire type identification result according to the matching score result, wherein the establishment of the template classification library comprises the following steps:
according to the obtained tire pattern patterns of different models, threshold segmentation of the tire pattern characteristics is carried out;
edge detection is carried out according to the pattern features segmented by the threshold value, secondary self-adaptive detection is carried out for the light supplementing deficiency of the pattern edge, and the deficiency edge information is fitted;
and packaging tire pattern patterns of different models as templates, and establishing a template classification library through the packaged templates.
When the tire sidewall image is obtained, a monocular camera is arranged on one side of the tire sidewall, camera calibration by a Zhang Zhengyou method is carried out, and camera internal references are determined; after the monocular camera calibration is completed, the monocular camera is used for collecting tire sidewall images.
9. A computer-readable storage medium having stored thereon a computer program, characterized by: the program, when executed by a processor, implements the steps of the method of any of claims 1-7.
10. A computer apparatus/device/system comprising:
a memory for storing computer programs/instructions;
a processor for executing the computer program/instructions to implement the steps of the method of any one of claims 1-7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410134393.9A CN117746165A (en) | 2024-01-31 | 2024-01-31 | Method and device for identifying tire types of wheel type excavator |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410134393.9A CN117746165A (en) | 2024-01-31 | 2024-01-31 | Method and device for identifying tire types of wheel type excavator |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117746165A true CN117746165A (en) | 2024-03-22 |
Family
ID=90252960
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202410134393.9A Pending CN117746165A (en) | 2024-01-31 | 2024-01-31 | Method and device for identifying tire types of wheel type excavator |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117746165A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118096759A (en) * | 2024-04-26 | 2024-05-28 | 深圳市二郎神视觉科技有限公司 | Method and device for detecting tire tread pattern and electronic equipment |
-
2024
- 2024-01-31 CN CN202410134393.9A patent/CN117746165A/en active Pending
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118096759A (en) * | 2024-04-26 | 2024-05-28 | 深圳市二郎神视觉科技有限公司 | Method and device for detecting tire tread pattern and electronic equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110232389B (en) | Stereoscopic vision navigation method based on invariance of green crop feature extraction | |
CN109781737B (en) | Detection method and detection system for surface defects of hose | |
CN117746165A (en) | Method and device for identifying tire types of wheel type excavator | |
CN108052904B (en) | Method and device for acquiring lane line | |
CN103902985B (en) | High-robustness real-time lane detection algorithm based on ROI | |
CN112528868B (en) | Illegal line pressing judgment method based on improved Canny edge detection algorithm | |
CN109815822B (en) | Patrol diagram part target identification method based on generalized Hough transformation | |
CN107909571A (en) | A kind of weld beam shape method, system, equipment and computer-readable storage medium | |
CN115049700A (en) | Target detection method and device | |
CN114331986A (en) | Dam crack identification and measurement method based on unmanned aerial vehicle vision | |
CN113989604B (en) | Tire DOT information identification method based on end-to-end deep learning | |
CN102073852A (en) | Multiple vehicle segmentation method based on optimum threshold values and random labeling method for multiple vehicles | |
CN109063669B (en) | Bridge area ship navigation situation analysis method and device based on image recognition | |
CN115587966A (en) | Method and system for detecting whether parts are missing or not under condition of uneven illumination | |
CN113487538B (en) | Multi-target segmentation defect detection method and device and computer storage medium thereof | |
CN109558877B (en) | KCF-based offshore target tracking algorithm | |
CN111985436B (en) | Workshop ground marking recognition fitting method based on LSD | |
CN113095283A (en) | Lane line extraction method based on dynamic ROI and improved firefly algorithm | |
Xu et al. | A lane detection method combined fuzzy control with ransac algorithm | |
CN116071713A (en) | Zebra crossing determination method, device, electronic equipment and medium | |
CN115331193A (en) | Parking space identification method, parking space identification system, electronic equipment and storage medium | |
CN112489015B (en) | Chemical fiber drift impurity identification method for mobile robot | |
CN113971789A (en) | Method and computing system for detection | |
CN113505811A (en) | Machine vision imaging method for hub production | |
CN112489052A (en) | Line structure light central line extraction method under complex environment |
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 |