CN112330600A - Fault identification method for vehicle end link line fracture based on image processing - Google Patents
Fault identification method for vehicle end link line fracture based on image processing Download PDFInfo
- Publication number
- CN112330600A CN112330600A CN202011103884.5A CN202011103884A CN112330600A CN 112330600 A CN112330600 A CN 112330600A CN 202011103884 A CN202011103884 A CN 202011103884A CN 112330600 A CN112330600 A CN 112330600A
- Authority
- CN
- China
- Prior art keywords
- link line
- end link
- image
- vehicle end
- fault
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/12—Edge-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/181—Segmentation; Edge detection involving edge growing; involving edge linking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20172—Image enhancement details
- G06T2207/20192—Edge enhancement; Edge preservation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30248—Vehicle exterior or interior
- G06T2207/30252—Vehicle exterior; Vicinity of vehicle
Abstract
A fault identification method for vehicle end link line fracture based on image processing belongs to the technical field of railway wagon operation. The invention aims to solve the problems that the train operation safety is not high because the faults of the train end link lines are easily identified by manpower, the fatigue is easy to occur, the strength is high, and the omission is easy to occur. The method includes the steps that linear array high-speed cameras are installed around a track, and a 2D linear array image of the truck is shot and obtained through the linear array high-speed cameras; roughly positioning the end part of the truck and intercepting a subgraph; further accurately positioning the subgraph; and (3) intercepting the horizontal (vertical) edge image, setting a calculation threshold value, adopting a least square method and a self-contrast design fault identification algorithm, outputting alarm information by the system if a fault of the truck is detected, and uploading the alarm information to a platform for manual confirmation. The invention is used for detecting the breakage fault of the rail wagon end link line.
Description
Technical Field
The invention relates to a fault identification method for vehicle end link line fracture based on image processing. Belongs to the technical field of railway freight car operation.
Background
In order to ensure the safe operation of a train, the fault detection of the train is required, for the fault detection of a train end link line, the conventional method is to photograph the train by a detection device and then manually observe to find out the fault point of the train, although the fault detection can be carried out in the running process of the train without stopping, the method is easy to cause fatigue and high strength due to manual observation, and the train end link line is similar to the background color and changeable in form from an image, so that the missed detection is easily caused, and further the running safety of the train is not high.
At present, more and more equipment fault detection, maintenance and other work can be carried out by a machine instead of manual work, the machine has the characteristics of low cost, unified rule, no fatigue in 24 hours and the like, and the feasibility is realized by replacing the traditional manual detection with an image processing and identifying technology.
Disclosure of Invention
The invention aims to solve the problems that fatigue is easy to occur, the strength is high, and missing detection is easy to occur to cause low running safety of a train when the fault of a train end link line is manually identified. A fault identification method for vehicle end link line fracture based on image processing is provided.
The fault identification method of the vehicle end link line fracture based on image processing is characterized by comprising the following steps:
firstly, acquiring a train gray image by imaging equipment;
secondly, intercepting a sample sub-image containing a vehicle end link part in the gray-scale image according to the prior knowledge and the known hardware data and the wheel base information;
thirdly, performing threshold segmentation on the sample subimage containing the vehicle end link part intercepted in the second step, acquiring an accurate area where the vehicle end link line is located by utilizing a gray level statistics method, and accurately intercepting the sample subimage; the gray scale statistics method comprises the following steps: for the sample subimages, counting the proportion of each row of non-black, namely white pixel points in the whole row width after segmentation from bottom to top, and determining the position of a vehicle-end link line according to the proportion;
judging the type of the fault of the vehicle end link line fracture according to the accurately positioned sample subimage, wherein the type of the fault of the vehicle end link line fracture comprises the following steps: the breakage and the sag of the vehicle end link line exceed the boundary of the normal link line, and the breakage of the vehicle end link line is mixed with other link lines and does not exceed the boundary of the normal link line;
advantageous effects
1. The image processing technology is utilized to carry out fault identification, the manpower is liberated, the manual work amount is reduced, the discovery rate and the accuracy rate of the fault identification are improved, and the safety of train operation is improved.
2. An edge extraction mode is designed aiming at image features, so that the outer edge of a vehicle end link line can be completely extracted and contains few false edges.
3. The detail of low gray level in the gray level image is better enhanced by using a logarithmic image enhancement method.
4. And the least square method is used for fitting the edge of the vehicle-end link line, so that the part with a small protruded broken part can be identified.
5. As shown in fig. 9, the method using self-contrast solves the problem that the flexible deformable member is difficult to unify the threshold value.
Drawings
FIG. 1 is a flow chart of the fault identification of the present invention;
FIG. 2 is a vertical edge extraction result;
FIG. 3 shows the results of horizontal edge extraction;
FIG. 4 shows the results after vertical edge enhancement;
FIG. 5 shows the results after edge enhancement in the horizontal direction;
FIG. 6 is the result after edge blending in the horizontal and vertical directions;
FIG. 7 shows the results of edge screening and fitting;
FIG. 8 is a diagram illustrating the result of dividing the hatched area of the vehicle end link line;
FIG. 9 is a diagram illustrating another segmentation result of the vehicle end link line shadow area.
Detailed Description
The first embodiment is as follows: the present embodiment is described with reference to fig. 1, and the method for identifying a failure of a vehicle end link line break by image processing according to the present embodiment includes:
firstly, installing imaging equipment at a detection station, and acquiring high-definition linear array 2D gray images of each part of a passing train at different time under different weather conditions by the imaging equipment; the diversity of the acquired data is ensured, and the stability of the final design algorithm can be improved;
step two, roughly positioning a sample sub-image:
according to priori knowledge, known hardware data (data given by mechanical parts such as a sensor and frames with other functions) and information such as wheel base and the like, intercepting a sample subimage, namely a roughly positioned sample subimage, from a high-definition linear array 2D gray image, wherein the image comprises a vehicle end link part, and the upper part, the lower part, the left part and the right part of the image are provided with reserved parts to prevent incomplete interception caused by errors such as the wheel base of the hardware and the like; the recognition speed of the program can be accelerated after the sample sub-images are intercepted;
step three, accurately positioning the sub-image of the sample:
after the coarsely positioned sample subimage is obtained, because the gray difference between the carriage and the vehicle end is obvious, the accurate area where the vehicle end link line is located can be obtained by utilizing a gray statistic method, namely the proportion of each row of non-black, namely white pixel points occupying the whole row width after the division is counted from bottom to top, the position of the row of non-black, namely white pixel points is determined according to the proportion, and the sample subimage is further accurately positioned;
step four, the vehicle end link line belongs to a flexible component and is connected with the front and rear carriages, and the middle of the vehicle end link line naturally sags; in the running process of the train, the forms of the train end link lines are different; the shapes of the vehicle end link lines among different carriages are different, so an algorithm suitable for various vehicle end link line fracture conditions is needed;
judging the fault type of the vehicle end link line fracture according to the accurately positioned sample subimage, wherein the vehicle end link line fracture is divided into two situations, one situation is that the vehicle end link line naturally sags beyond the boundary of a normal vehicle end link line after being fractured, and the other situation is that the vehicle end link line is mixed with other lines after being fractured and does not exceed the boundary of the normal vehicle end link line;
reading the image for identification and analysis, outputting alarm information if the identification result is that a fault occurs, and uploading the alarm information to a platform for manual confirmation; and if the recognition result is normal, skipping the image to recognize the next image.
The second embodiment is as follows: the difference between the first embodiment and the specific embodiment is that the fourth step is a boundary fault method for judging that the natural droop exceeds the normal vehicle end link line after the vehicle end link line is broken; the specific process is as follows:
because the gray levels of the vehicle-end connecting line in the sample subimage and the background are similar and are influenced by illumination, environment, vehicle type and the like, the image is not divided by utilizing the gray levels;
the vehicle end chain line sags in the middle under a natural state and is in a parabolic state, so that the edges in the horizontal direction and the vertical direction are used as characteristics, and the important degrees of the edges in the two directions are different, so that the vehicle end chain line sags naturally under different edge extraction methods; the edge image in the horizontal direction is used for obtaining a complete vehicle-end link line profile and fitting a normal boundary, and the edge image in the vertical direction is needed when the vehicle-end link line fracture exceeds the normal boundary;
because the edge in the horizontal direction is only useful in the upper half (i.e. the upper half in the image horizontal direction 1/2) and the other parts are interference (the broken part of the vehicle end link line is judged by the edge in the vertical direction, the edge in the horizontal direction is only an extracted intact part, the normal vehicle end link lines shown in the figure are all in the upper half of the subgraph, and the lower half is not present), the positioning result (calculated by the coordinates obtained above) is obtained;
setting the part 1/2 below the vertical direction of the edge image in the horizontal direction as 0 (namely black), and obtaining the extraction result as shown in fig. 2 and 3, it can be seen that the gray scale of the image is low, so that the gray scale needs to be further enhanced;
according to the method, the gray scales of the image at the edge in the horizontal direction and the edge in the vertical direction are enhanced by utilizing the logarithmic function, and the part with low gray scale can be better enhanced compared with the common linear enhancement;
dividing the image subjected to edge enhancement in the horizontal direction by using a fixed threshold value to obtain a binary image;
calculating a threshold value for the image after edge enhancement in the vertical direction, intercepting a range from 1/3 to 2/3 in the horizontal direction and a range from 0 to 1/3 in the vertical direction in fig. 4 (the gray level of the partial image can represent the gray level of a vehicle end link line), calculating a maximum pixel value in the intercepted area, and subtracting 30 (preferably) from the pixel value to serve as a vertical direction segmentation threshold value;
binarizing the image after the edge enhancement in the vertical direction by using the threshold value, and fusing the obtained horizontal and vertical images to obtain a complete vehicle end link line edge image, as shown in fig. 6;
and (2) fitting the lower boundary of the binary image (the edge image of the vehicle end link line) by using a least square method to further judge whether the fault that the vehicle end link line breaks and falls to exceed the boundary of the normal link line occurs or not (the least square method can fit the boundary of the normal link line, all the edge binary images obtained by fusion are set to be 0 according to the upper part of the curve obtained by fitting, and the rest white part in the image is just the part which exceeds the boundary).
Other steps and parameters are the same as those in the first embodiment.
The third concrete implementation mode: the difference between the present embodiment and the first or second embodiment is that the fault identification is performed after the vehicle end link line is broken and mixed with other lines, and the fault identification does not exceed the boundary of the normal vehicle end link line; the specific process is as follows:
the train end link line is a flexible component, the form of the train end link line changes in the running process of a train, and the states of connecting lines between different carriages are different, so that the train end link line is mixed with other lines after being broken without setting a fixed threshold value, and the boundary fault analysis of the normal train end link line is not exceeded;
according to the flexible natural drooping characteristic of the vehicle end link line, the left and right link lines are basically symmetrical, so the invention provides a self-comparison method, namely, a complete vehicle end link line binary image is divided into a left part and a right part, and the characteristics of the left part image and the right part image are compared;
and if the value exceeds a threshold value (obtained according to image test), judging that the vehicle end link line fracture is mixed with other lines and does not exceed the boundary fault of the normal link line.
Other steps and parameters are the same as those in the first or second embodiment.
The fourth concrete implementation mode: this embodiment is different from the first to third embodiments in that the gray-scale value on the horizontal-direction edge image is expressed as:
Phorizontal[r,c]＝|p[r,c+3]-p[r,c]|
wherein r represents the coordinate on the Y-axis of the pixel; c represents the coordinates of the pixel on the X-axis; p [ r, c ]]Is represented by [ r, c]A pixel value of a point; phorizontal[r,c]Representing the horizontal edge image at coordinates r, c]The gray value of the point.
Other steps and parameters are the same as those in one of the first to third embodiments.
The fifth concrete implementation mode: this embodiment is different from the first to fourth embodiments in that the gray scale value on the vertical edge image is represented by:
Pvertical[r,c]＝|p[r+5,c]-2*p[r,c]|
wherein, Pvertical[r,c]Representing the vertical edge image at coordinates r, c]The gray value of the point.
Other steps and parameters are the same as in one of the first to fourth embodiments.
The sixth specific implementation mode: the present embodiment is different from one of the first to fifth embodiments in that the horizontal direction edge and the vertical direction edge image are enhanced by using a logarithmic function; the concrete expression is as follows:
penhance[r,c]＝40log(p[r,c]+1)
wherein p isenhance[r,c]Expressed in the coordinate [ r, c]The dot-enhanced gray scale values and the enhanced results are shown in fig. 4 and 5.
Other steps and parameters are the same as those in one of the first to fifth embodiments.
The seventh embodiment: the difference between the present embodiment and one of the first to the fourth embodiments is that the least square method is used to fit the lower boundary of the image to further determine the fault that the break and the droop of the vehicle-end link line exceed the boundary of the normal link line; the specific process is as follows:
after the binary image is obtained, because the segmentation result of the broken vehicle end link line is still connected with the normal vehicle end connecting line, the edge part of the normal connecting line needs to be determined at first, and the features of the segmentation result exceeding the edge can be further calculated to analyze whether the failure occurs; because the radian size and the like of each vehicle end link line are different, a universal template cannot be selected to screen the external area of the normal vehicle end link line; a complete edge can be obtained by using least square fitting, so that the fault is further judged;
firstly, selecting the maximum connected domain in the binary image as the area where the vehicle end link line is located, and excluding interference from other parts; selecting the coordinate of the point at the lower end of the area as an input value of least square fitting; because the curve is a parabola, the curve can be fitted to be a quadratic curve; analyzing the characteristics of the part exceeding the fitting curve, such as form, size and the like, and judging whether the fault that the link line exceeds the boundary after the link line at the vehicle end is broken; the obtained edges and the fitting result are shown in fig. 7 (the normal car end link line image is a complete and smooth curve, and in a fracture and sagging state, there are projections, and the features of the projection part can be obtained according to the fitted curve, but the projections may also be the light interference of the image, so that the projections need to be eliminated according to the features, and the interference is more than that of a very slender line).
Other steps and parameters are the same as those in one of the first to sixth embodiments.
The specific implementation mode is eight: the difference between this embodiment and one of the first to seventh embodiments is that the image after horizontal edge enhancement is divided by using a fixed threshold, where the fixed threshold is 30.
Claims (8)
1. The fault identification method of the vehicle end link line fracture based on image processing is characterized by comprising the following steps:
firstly, acquiring a train gray image by imaging equipment;
secondly, intercepting a sample sub-image containing a vehicle end link part in the gray-scale image according to the prior knowledge and the known hardware data and the wheel base information;
thirdly, performing threshold segmentation on the sample subimage containing the vehicle end link part intercepted in the second step, acquiring an accurate area where the vehicle end link line is located by utilizing a gray level statistics method, and accurately intercepting the sample subimage; the gray scale statistics method comprises the following steps: for the sample subimages, counting the proportion of each row of non-black, namely white pixel points in the whole row width after segmentation from bottom to top, and determining the position of a vehicle-end link line according to the proportion;
judging the type of the fault of the vehicle end link line fracture according to the accurately positioned sample subimage, wherein the type of the fault of the vehicle end link line fracture comprises the following steps: the broken vehicle end link line sags beyond the boundary of the normal link line, and the broken vehicle end link line is mixed with other link lines and does not exceed the boundary of the normal link line.
2. The method for identifying the fault of the vehicle end link line fracture based on the image processing as claimed in claim 1, wherein the step four is used for identifying the fault that the vehicle end link line fracture sags beyond the boundary of a normal link line; the specific process is as follows:
accurately positioning the acquired sample sub-images based on the third step, and dividing the sample sub-images into two images of edges in the horizontal direction and the vertical direction by using the edges in the horizontal direction and the vertical direction as features;
setting the lower 1/2 part of the edge image in the horizontal direction as 0, and enhancing the gray scale of the edge image in the horizontal direction and the edge image in the vertical direction by using a logarithmic function;
dividing the image subjected to edge enhancement in the horizontal direction by using a fixed threshold value to obtain a binary image;
calculating a threshold value of the image after the edge enhancement in the vertical direction, intercepting an interval from 1/3 to 2/3 in the horizontal direction and an interval from 0 to 1/3 in the vertical direction in the image after the edge enhancement in the vertical direction, wherein the interval is the gray level of a vehicle end link line, calculating the maximum pixel value in the area, and subtracting 30 from the pixel value to be used as a vertical direction segmentation threshold value;
binarizing the image after the edge enhancement in the vertical direction by using the threshold value, and fusing the two obtained images to obtain a complete vehicle end link line edge image;
and the fault that the vehicle end link line breaks and falls beyond the boundary of the normal link line can be judged by fitting the lower boundary of the edge image of the vehicle end link line by using a least square method.
3. The fault identification method for the vehicle end link line fracture based on the image processing is characterized in that the fault identification method for the vehicle end link line fracture which is mixed with other lines and does not exceed the boundary of a normal link line is carried out; the specific process is as follows:
finding out the lowest point of each vehicle end link line in the shadow area of each vehicle end link line, dividing the complete vehicle end link line binary image into a left part and a right part by using a self-comparison method, comparing the values of the divided left part and right part images, wherein the values of the left part and the right part are the number of non-black or white pixel points in the image, and if the values exceed a threshold value, judging that the vehicle end link line is broken and mixed with other lines and does not exceed the boundary fault of a normal link line; the threshold is obtained according to image testing.
4. The method for identifying the fault of the vehicle end link line fracture based on the image processing as claimed in claim 2, wherein the gray values on the horizontal direction edge image are expressed as:
Phorizontal[r,c]＝|p[r,c+3]-p[r,c]|
wherein r represents the coordinate on the Y-axis of the pixel; c represents the coordinates of the pixel on the X-axis; p [ r, c ]]Is represented by [ r, c]A pixel value of a point; phorizontal[r,c]Representing the horizontal edge image at coordinates r, c]The gray value of the point.
5. The method for identifying the fault of the vehicle end link line fracture based on the image processing as claimed in claim 2, wherein the gray value on the vertical direction edge image is represented as:
Pvertical[r,c]＝|p[r+5,c]-2*p[r,c]|
wherein, Pvertical[r,c]Representing the vertical edge image at coordinates r, c]The gray value of the point.
6. The fault identification method for the vehicle end link line fracture based on the image processing is characterized in that the gray scales of the horizontal direction edge image and the vertical direction edge image are enhanced by using a logarithmic function; the concrete expression is as follows:
penhance[r,c]＝40log(p[r,c]+1)
wherein p isenhance[r,c]Expressed in the coordinate [ r, c]Point enhanced gray scale values.
7. The method for identifying the fault of the vehicle end link line fracture based on the image processing as claimed in claim 1, wherein the step of fitting the lower boundary of the image by using a least square method is further used for judging the fault that the vehicle end link line fracture and the droop exceed the boundary of a normal link line; the specific process is as follows:
selecting the largest connected domain in the image as the region where the connecting line is located, and eliminating interference from other parts; and selecting the coordinate of the point at the lower end of the area as an input value for least square fitting, fitting the parabola into a quadratic curve, and analyzing the shape and size characteristics of the part exceeding the fitted curve, namely judging whether the fault that the vehicle-end link line breaks and falls beyond the boundary of the normal link line occurs or not.
8. The method for identifying the fault of the vehicle end link line fracture based on the image processing as claimed in claim 2, wherein the image after the edge enhancement in the horizontal direction is segmented by using a fixed threshold, wherein the fixed threshold value is 30.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011103884.5A CN112330600B (en) | 2020-10-15 | 2020-10-15 | Fault identification method for vehicle end link line fracture based on image processing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011103884.5A CN112330600B (en) | 2020-10-15 | 2020-10-15 | Fault identification method for vehicle end link line fracture based on image processing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112330600A true CN112330600A (en) | 2021-02-05 |
| CN112330600B CN112330600B (en) | 2021-06-04 |
Family
ID=74313617
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011103884.5A Active CN112330600B (en) | 2020-10-15 | 2020-10-15 | Fault identification method for vehicle end link line fracture based on image processing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112330600B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114219806A (en) * | 2022-02-22 | 2022-03-22 | 成都数联云算科技有限公司 | Automobile radar detection method, device, equipment, medium and product |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104657706A (en) * | 2015-01-27 | 2015-05-27 | 北京大学 | Image-based high-speed railway line pole breakage abnormality and connecting structural body abnormality detection method |
| CN105654461A (en) * | 2014-12-02 | 2016-06-08 | 中国科学院沈阳自动化研究所 | Machine vision detection method for fracture of multi-bundle conductor spacer |
| CN106600581A (en) * | 2016-12-02 | 2017-04-26 | 北京航空航天大学 | Train operation fault automatic detection system and method based on binocular stereoscopic vision |
-
2020
- 2020-10-15 CN CN202011103884.5A patent/CN112330600B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105654461A (en) * | 2014-12-02 | 2016-06-08 | 中国科学院沈阳自动化研究所 | Machine vision detection method for fracture of multi-bundle conductor spacer |
| CN104657706A (en) * | 2015-01-27 | 2015-05-27 | 北京大学 | Image-based high-speed railway line pole breakage abnormality and connecting structural body abnormality detection method |
| CN106600581A (en) * | 2016-12-02 | 2017-04-26 | 北京航空航天大学 | Train operation fault automatic detection system and method based on binocular stereoscopic vision |
Non-Patent Citations (2)
| Title |
|---|
| 戴鹏，等: "货车转向架安全链故障计算机视觉识别算法研究", 《计算机测量与控制》 * |
| 李骏: "动车组运行故障动态图像检测系统（TEDS）设计与实现", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114219806A (en) * | 2022-02-22 | 2022-03-22 | 成都数联云算科技有限公司 | Automobile radar detection method, device, equipment, medium and product |
| CN114219806B (en) * | 2022-02-22 | 2022-04-22 | 成都数联云算科技有限公司 | Automobile radar detection method, device, equipment, medium and product |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112330600B (en) | 2021-06-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111079747B (en) | Railway wagon bogie side frame fracture fault image identification method | |
| CN112434695B (en) | Upper pull rod fault detection method based on deep learning | |
| US20170323435A1 (en) | Defect quantification method, defect quantification device, and defect evaluation value display device | |
| CN111080611A (en) | Railway wagon bolster spring fracture fault image identification method | |
| CN113465541B (en) | Contact line abrasion measuring method and system | |
| KR20090101356A (en) | Defect detecting device, and defect detecting method | |
| CN116337868B (en) | Surface defect detection method and detection system | |
| JP2017215220A (en) | Railway vehicle appearance inspection device | |
| CN112330600B (en) | Fault identification method for vehicle end link line fracture based on image processing | |
| CN113111903A (en) | Intelligent production line monitoring system and monitoring method | |
| CN112001908B (en) | Railway freight car sleeper beam hole carried foreign matter detection method | |
| CN112907532B (en) | Improved truck door falling detection method based on fast RCNN | |
| CN114399518A (en) | Method for monitoring tightness of vehicle bottom bolt and electronic equipment | |
| CN111832562B (en) | Spring supporting plate fault identification method based on image processing | |
| CN112085723B (en) | Automatic detection method for spring jumping fault of truck bolster | |
| CN111652228B (en) | Railway wagon sleeper beam hole foreign matter detection method | |
| CN112329858A (en) | Image recognition method for breakage fault of anti-loosening iron wire of railway motor car | |
| CN116433938A (en) | Abnormal detection method, system and storage medium for bolt looseness of subway gearbox | |
| CN112132824B (en) | Automatic detection method for failure of freight car axle box spring | |
| CN113933309A (en) | Method for retesting defects in blind hole by AOI machine | |
| CN112102293B (en) | Rapid detection method for foreign matters in triangular holes of railway wagon | |
| CN112614097B (en) | Method for detecting foreign matter on axle box rotating arm of railway train | |
| CN109978879A (en) | Case angle based on loading goods train video monitor enters slot condition detection method | |
| CN112365481B (en) | Method for detecting bolt loss of cross beam assembly based on image processing | |
| CN112733771B (en) | Railway train jumper wire foreign matter fault identification method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |