CN118212170A - Edge detection method and device for guide rail steel plate protection cover - Google Patents
Edge detection method and device for guide rail steel plate protection cover Download PDFInfo
- Publication number
- CN118212170A CN118212170A CN202211568856.XA CN202211568856A CN118212170A CN 118212170 A CN118212170 A CN 118212170A CN 202211568856 A CN202211568856 A CN 202211568856A CN 118212170 A CN118212170 A CN 118212170A
- Authority
- CN
- China
- Prior art keywords
- edge
- protective cover
- image
- steel plate
- rail steel
- 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
- 229910000831 Steel Inorganic materials 0.000 title claims abstract description 85
- 239000010959 steel Substances 0.000 title claims abstract description 85
- 238000003708 edge detection Methods 0.000 title claims abstract description 55
- 238000000034 method Methods 0.000 title claims abstract description 26
- 230000001681 protective effect Effects 0.000 claims abstract description 91
- 238000012545 processing Methods 0.000 claims abstract description 30
- 238000004458 analytical method Methods 0.000 claims abstract description 17
- 238000010191 image analysis Methods 0.000 claims abstract description 9
- 238000001514 detection method Methods 0.000 claims abstract description 6
- 238000012549 training Methods 0.000 claims description 15
- 238000010586 diagram Methods 0.000 claims description 11
- 230000002159 abnormal effect Effects 0.000 claims description 9
- 238000013507 mapping Methods 0.000 claims description 7
- 238000002372 labelling Methods 0.000 claims description 6
- 230000005764 inhibitory process Effects 0.000 claims description 5
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 238000011161 development Methods 0.000 abstract description 4
- 238000007689 inspection Methods 0.000 description 9
- 238000009434 installation Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 230000001629 suppression Effects 0.000 description 3
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 229910052742 iron Inorganic materials 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The invention relates to an edge detection method and device for a guide rail steel plate protective cover. The method comprises the steps that a control module controls a monocular camera arranged in a working area to acquire an original image, an edge detection analysis module performs edge detection through an image processing unit, an edge image processing unit and an edge image analysis unit, a processing result is transmitted to the control module to form a feedback loop, and the control module judges whether a warning signal is generated according to detection result information to stop the operation of a guide rail steel plate protective cover, so that the safety of the numerical control machine tool is ensured. The invention establishes the edge detection of the guide rail steel plate protective cover in the numerical control machine tool, and judges whether the protective cover is in a normal running state or not by analyzing the edge detection result on the embedded development board, and has the advantages of strong practicability and low cost.
Description
Technical Field
The invention belongs to the field of industrial automation, in particular to a method and a device for detecting the edge of a guide rail steel plate protective cover based on monocular vision, which are applied to the automatic detection of the guide rail steel plate protective cover on an industrial conveyor belt and are convenient for carrying out automatic safety management on workshop industry.
Background
Currently in the field of industrial automation, computer vision detection is an important direction on numerical control machines. Compared with the manual inspection monitoring of whether the numerical control machine has abnormal operation conditions, the numerical control machine is judged and detected through computer vision, so that the abnormal conditions can be found more accurately and timely, continuous operation can be carried out for 24 hours, and rest is not needed. The guide rail steel plate protective cover plays an important role in complex and precise part processing due to the fact that high-efficiency and high-speed completion is required on a numerical control machine tool. On one hand, the numerical control machine tool can prevent chips or other sharp objects from entering the machine tool to damage the parts of the numerical control machine tool, and on the other hand, the numerical control machine tool can transport the parts to enter the numerical control machine tool for subsequent processing. Therefore, the visual monitoring is carried out on the guide rail steel plate protective cover, and whether the guide rail steel plate protective cover is in a normal running state or not is judged, so that the guide rail steel plate protective cover is a very important problem.
The method for monitoring the guide rail steel plate protective cover mainly depends on a manual inspection mode. The inspection staff can inspect the steel plate on site at intervals, the operation state of the guide rail protective cover is judged by eyes of the inspection staff, the rough inspection mode is difficult to monitor and evaluate, and meanwhile, the inspection information feedback is delayed, so that the production efficiency and quality are seriously affected. If the problem is not found in time, the guide rail can be stopped, and the production of the machine tool is suspended.
Disclosure of Invention
According to the technical problems, the edge detection method and the device for the guide rail steel plate protection cover are provided, and are used for solving the problems that the existing manual inspection method is low in efficiency and difficult to respond to faults in time, and meanwhile, the workshop management safety is solved, and the degree of automation is improved.
The invention adopts the following technical means: an edge detection method for a guide rail steel plate protective cover comprises the following steps:
s1, collecting a data set: a camera is used for collecting color images of the guide rail steel plate protective cover;
S2, preprocessing a data set: carrying out graying treatment on the color image, and marking the gray image to represent the edge of the protective cover; constructing a data set according to the marked gray level image for training of the neural network;
S3, building a network model: acquiring gradient information from gray images in a dataset by adopting an improved LBP algorithm, and inputting pixel difference values representing the gradient information into ResNet networks for training to acquire edge images containing edge information of the protective cover;
S4, edge detection: and acquiring a protective cover image in real time, processing the protective cover image by adopting an improved LBP algorithm, inputting the obtained pixel difference value representing gradient information into a ResNet network after training to obtain an edge image containing protective cover edge information, and then carrying out non-maximum value inhibition processing on the edge information to obtain a single pixel edge image.
The gradient information is acquired from the gray level image in the data set by adopting an improved LBP algorithm, and the method comprises the following steps:
For each pixel, taking the pixel as a central pixel, respectively calculating the pixel difference value between the surrounding 8 neighborhood pixels and the center, the pixel difference value between every two adjacent pixels in the surrounding 8 neighborhood pixels and the pixel difference value in the radial direction, and adopting at least one of the three difference values for representing gradient information of the image.
The improved LBP algorithm forms a backbone network with the ResNet network.
Training ResNet network, loss function adopts Annotator-robust loss function:
Wherein lambda is a super parameter for balancing positive and negative sample proportion, P (x) is a sigmoid function, and W represents all parameters to be learned; x i represents the feature vector of the current pixel;
The average value of 5 labeling in the pretreatment of the S2 data set is calculated, and is represented by y i, a new edge probability mapping diagram is generated, the range of the new edge probability mapping diagram is [0,1,0 ] which represents that no labeling is carried out in the 5 times, and 1 represents that the new edge probability mapping diagram is labeled as edge pixels in the 5 times of labeling;
Meanwhile, the edge probability value exceeds a threshold value eta as a positive sample Y +, and the edge probability value is smaller than eta as a negative sample Y -.
The edge detection method for the guide rail steel plate protection cover further comprises S5 edge judgment: and judging the distance and angle of the edges between the protective sheets of the protective cover according to a protective cover edge coordinate equation obtained by the single pixel edge image so as to determine whether the protective cover of the guide rail steel plate is in a normal running state.
The S5 edge judgment comprises the following steps:
For a single-pixel edge image, an edge L is one end of a fixed guide rail steel plate protective cover, a coordinate system is established by taking the edge L of the fixed end of the guide rail steel plate protective cover as a y axis, and a coordinate equation of the edge L 1、L2、L3、…、Ln between all the pieces of the guide rail steel plate protective cover is obtained, wherein n is the number of protective pieces of the guide rail steel plate protective cover;
a) Acquiring angles alpha between the edges L 1、L2、L3、…、Ln and the fixed edges L respectively; if alpha is larger than delta, judging that the guide rail steel plate protective cover is abnormal, and alarming; wherein delta is the allowable error angle; otherwise, normal operation is performed;
B) The distance d 1、d2、d3、…、dn between two adjacent edges of the edges L, L 1、L2、L3、…、Ln is calculated, and it is determined whether the following formula is satisfied:
If the protection cover meets the protection cover requirement, judging that the protection cover of the guide rail steel plate is abnormal at the moment, and alarming; otherwise, normal operation is performed; where ε is the allowable error distance.
An edge detection device facing a guide rail steel plate protection cover, comprising:
the multidimensional image acquisition module is used for acquiring a color image of the guide rail steel plate protective cover in the working area through the camera according to the received image acquisition control signal and outputting the color image to the edge detection analysis module;
an edge detection analysis module comprising:
The image processing unit is used for acquiring edge information from the color image of the guide rail steel plate protective cover through the backbone network;
The edge image processing unit is used for carrying out non-maximum value inhibition processing on the obtained edge information to obtain a single-pixel edge image;
And the edge image analysis unit is used for judging according to the edge information in the single-pixel edge image to obtain the detection of whether the protective cover is normal or not.
The beneficial effects of the invention are as follows:
1. The invention establishes the edge detection of the guide rail steel plate protective cover in the numerical control machine tool, judges whether the protective cover is in a normal running state or not by analyzing the edge detection result on the embedded development board, and has the advantages of strong practicability and low cost.
2. According to the edge detection method and device for the guide rail steel plate protective cover, disclosed by the invention, a set of edge detection method and device based on computer vision is constructed through the multidimensional image acquisition module, the edge detection analysis module and the control module, and the running state of the steel plate protective cover can be monitored in real time. Meanwhile, when the steel plate protective cover is detected to be abnormal, the control module sends out a warning signal, so that timely protection measures are taken, and the safety of the numerical control machine tool system is ensured.
Drawings
FIG. 1 is a flow chart of the operation of the edge detection device of the rail steel plate protective cover of the present invention;
FIG. 2 is a schematic view of an image captured by a monocular camera directly above the present invention;
FIG. 3 is an edge pictorial view of the rail steel panel protective cover of the present invention;
FIG. 4a is a schematic diagram showing a failure of the edge detection result of the steel rail protective cover of the present invention;
FIG. 4b is a second schematic diagram of the edge detection result of the steel rail protective cover of the present invention;
fig. 5 is a schematic diagram of three forms of LBP algorithm.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings.
An edge detection device of a rail steel plate protective cover, comprising:
The device comprises a multidimensional image acquisition module, an edge detection analysis module and a control module; the multidimensional image acquisition module receives an image acquisition control signal of the control module, acquires an original image of the guide rail steel plate protective cover equipment in a working area, outputs the original image to the edge detection analysis module for image data processing and analysis, and obtains an edge detection result to be fed back to the control module for warning control of the guide rail steel plate protective cover equipment.
The multidimensional image acquisition module comprises an acquisition bracket arranged in a working area and a monocular camera arranged on the acquisition bracket and is used for acquiring original image information of the guide rail steel plate protective cover.
The edge detection analysis module and the control module are realized on an embedded development board.
The edge detection prediction module comprises an image processing unit, an edge image processing unit and an edge image analysis unit;
the image processing unit acquires edge image information of an original image of the guide rail steel plate protective cover on the embedded development board through a backbone network;
the edge image processing unit refines the obtained edge image information to obtain an edge image of the single-pixel guide rail steel plate protective cover;
And the edge image analysis unit judges the acquired single-pixel edge information of the guide rail steel plate protective cover, and feeds back an analysis result to the control module, wherein the analysis device is in a normal running state.
The backbone network includes three modified LBP algorithms connected in sequence and ResNet backbone networks.
The edge image processing unit refines the edge image by adopting a non-maximum suppression algorithm to obtain a single-pixel edge map of the guide rail steel plate protective cover.
The edge image analysis unit calculates the distance and the angle between each edge of the guide rail steel plate protective cover to determine whether the guide rail steel plate protective cover is in a normal running state.
The control module comprises an acquisition control circuit and a warning control circuit, wherein the acquisition control circuit sends out an image acquisition signal to control a monocular camera in a working area to acquire original image information, and the warning control circuit outputs a warning command to the guide rail steel plate protective cover equipment in combination with an edge detection result.
An edge detection method for a guide rail steel plate protective cover comprises the following steps:
the control module sends out an acquisition control signal through the acquisition control circuit, and controls the monocular camera of the working area to acquire the original image of the guide rail steel plate protective cover;
The edge detection analysis module acquires edge information of the steel plate protective cover in the image processing unit according to the received image data and outputs the edge information to the edge image processing unit; the edge image processing unit obtains a single-pixel edge image through a non-maximum suppression algorithm and outputs the single-pixel edge image to the edge image analysis unit; the edge image analysis unit judges according to the acquired edge information, calculates the distance and the angle between the edges of each piece of shield of the guide rail steel plate shield, compares the distance with a standard preset threshold value, and outputs a feedback result to the control module;
and the control module outputs a warning signal to the guide rail steel plate protective cover equipment through the warning control circuit according to the acquired edge detection result signal so as to warn when faults occur.
The image processing unit adopts three improved LBP differential processing algorithms for capturing image gradient information. And obtaining meaningful semantic features of the processed image pixel difference values through ResNet backbone networks, and finally generating robust and accurate edges.
The detected steel plate protective cover is in the prior art, has the physical shape shown in fig. 2, mainly comprises a left gray part in the figure, is one side of the steel plate fixed at a specified position, and consists of a plurality of strip-shaped protective sheets. And the middle of the protective sheet slightly bulges and stretches and changes along with the guide rail. The main effect of the steel plate protective cover can effectively prevent scrap iron and protect mechanical guide rails.
An edge detection device facing a guide rail steel plate protection cover, as shown in fig. 1, comprises: the system comprises a multidimensional image acquisition module, an edge detection analysis module and a control module. The multidimensional image acquisition module is responsible for acquiring an original image from a monocular camera right above the guide rail steel plate protective cover equipment by receiving an image acquisition control signal of the control module, outputting acquired image data to the edge detection analysis module for image data processing and analysis, and feeding back to the control module for warning control of the guide rail steel plate protective cover equipment according to an edge detection result.
The installation angle of the monocular camera is shown in fig. 2, and the installation angle of the camera is adjusted to enable the range of the guide rail steel plate protective cover which is completely opened to be completely covered in the acquired image. After the installation is completed, the camera is required to be calibrated and corrected, parameters of the camera are calculated, and the coordinate equation of each edge is calculated conveniently in the edge detection process.
The edge image effect diagram finally obtained by the edge image processing unit in the edge detection module is shown in fig. 3. And the edge L is one end of the fixed guide rail steel plate protective cover, a coordinate system is established by taking the edge L as a y axis, and coordinate equations of the edge L 1L2L3…Ln are calculated respectively.
The edge image analysis unit in the edge detection module defines both cases as fault conditions as shown in fig. 4 a-4 b. In fig. 4a, an angle α between the edge L 1L2L3…Ln and the fixed edge L is calculated, if α > δ, it is determined that the protection cover of the guide rail steel plate is abnormal at this time, and a detection result signal is transmitted to the control module to send out a warning signal, so as to halt the operation of the numerically-controlled machine tool. Wherein delta is a preset allowable error angle.
Fig. 4b, calculates the distance d 1d2d3…dn between edges LL 1L2L3…Ln, respectively, if there is the following formula:
And judging that the protective cover of the guide rail steel plate is abnormal at the moment, wherein epsilon is a preset allowable error distance.
The edge detection method for the guide rail steel plate protective cover comprises the following specific steps:
S1, collecting a data set: a camera is used for protecting a large number of images of different angles of the guide rail steel plate, and when shooting, a proper light source is selected, so that the definition and accuracy of the images are ensured as much as possible. The resolution of the acquired image is 2560 x 1920;
S2, preprocessing a data set: the image is annotated 5 times using Geolabel tools, and the color image is first grayed to reduce the amount of data that needs to be processed. The weighted average method is adopted to gray the color image to obtain a reasonable gray image, and the formula is as follows:
L＝R*299/1000+G*587/1000+B*114/1000
S3, data set division: dividing the data set into a training set, a testing set and a verification set according to the ratio of 7:2:1;
S4, building a network model: the baseline model employs a modified LBP algorithm and ResNet backbone, followed by a non-maximal suppression algorithm to obtain edge images of single pixels. Unlike existing edge detection algorithms, instead of directly importing image information into a neural network for training, the image is first used with a modified LBP algorithm to capture gradient information of the image. The ResNet network is then focused on extracting meaningful semantic features. And respectively calculating pixel difference values of the periphery and the center, and pixel difference values of the periphery circulation and pixel difference values of the radial direction, so as to obtain image gradient information. Three forms of improving the LBP algorithm are shown in fig. 5.
The processed image data is transmitted to ResNet network for training, and the loss function adopts Annotator-robust loss function, and the specific expression is as follows:
And (3) calculating the average value of 5 labels in the step (S2), and using y i to represent the average value, and generating a new edge probability mapping diagram, wherein the range is [0,1, 0] which indicates that none of the 5 labels is marked, and 1 which indicates that all of the 5 labels are marked as edge pixels. Meanwhile, the positive sample Y + with the edge probability value exceeding eta and the negative sample Y -.Xi with the edge probability value smaller than eta represent the feature vector of the current pixel. In this patent, η is 0.5.
Where λ is a super parameter used to balance the positive and negative sample ratios. P (x) is a standard sigmoid function, W represents all parameters to be learned.
S5, training a network model: and training by using the network model in the training set S4 obtained in the step S2 to obtain an edge image containing edge information. Training a model suitable for detecting the edge of the guide rail steel plate protective cover on the training set;
S6, inputting an image: acquiring an actual image of a guide rail steel plate protective cover in an actual machine tool from a monocular camera, wherein the acquisition rule is according to S1;
S7, edge detection: and (3) inputting the image acquired in the step (S6) into the step (S4) network model, detecting edge information in the image, and performing non-maximum value inhibition processing on the edge information to obtain a single-pixel edge image. Finally, a coordinate equation of L 1L2L3…Ln shown in FIG. 3 is obtained;
s8, edge judgment: according to the edge coordinate equation generated in the step S7, calculating the ratio and the angle of the edge distance:
αi＞δ
if either of the two formulas is true, a warning signal is sent out, the operation of the numerical control machine tool is suspended, and the numerical control machine tool waits for manual processing. Wherein delta and epsilon are predetermined error values.
According to the edge detection method and device for the guide rail steel plate protective cover, disclosed by the invention, a set of edge detection method and device based on computer vision is constructed through the multidimensional image acquisition module, the edge detection analysis module and the control module, and the running state of the steel plate protective cover can be monitored in real time. The invention aims at the method for detecting the guide rail steel plate protective cover, solves the problem that the efficiency is low when a manual inspection system is adopted, and also solves the problem that the manual inspection is not safe enough and cannot respond to faults in time. The invention has the advantages of strong practicability and low cost.
While the foregoing is directed to the preferred embodiments of the present invention, it will be appreciated by those skilled in the art that various modifications and adaptations can be made without departing from the principles of the present invention, and such modifications and adaptations are intended to be comprehended within the scope of the present invention.
Claims (7)
1. The edge detection method for the guide rail steel plate protection cover is characterized by comprising the following steps of:
s1, collecting a data set: a camera is used for collecting color images of the guide rail steel plate protective cover;
S2, preprocessing a data set: carrying out graying treatment on the color image, and marking the gray image to represent the edge of the protective cover; constructing a data set according to the marked gray level image for training of the neural network;
S3, building a network model: acquiring gradient information from gray images in a dataset by adopting an improved LBP algorithm, and inputting pixel difference values representing the gradient information into ResNet networks for training to acquire edge images containing edge information of the protective cover;
S4, edge detection: and acquiring a protective cover image in real time, processing the protective cover image by adopting an improved LBP algorithm, inputting the obtained pixel difference value representing gradient information into a ResNet network after training to obtain an edge image containing protective cover edge information, and then carrying out non-maximum value inhibition processing on the edge information to obtain a single pixel edge image.
2. The edge detection method for a rail steel plate protection cover according to claim 1, wherein the gradient information is obtained from the gray level image in the dataset by using an improved LBP algorithm, comprising the steps of:
For each pixel, taking the pixel as a central pixel, respectively calculating the pixel difference value between the surrounding 8 neighborhood pixels and the center, the pixel difference value between every two adjacent pixels in the surrounding 8 neighborhood pixels and the pixel difference value in the radial direction, and adopting at least one of the three difference values for representing gradient information of the image.
3. The edge detection method for a rail steel sheet protective cover according to claim 1, wherein the improved LBP algorithm and ResNet network form a backbone network.
4. The edge detection method for a rail steel sheet protection cover according to claim 1, wherein the ResNet network is trained, and a Annotator-robustt loss function is adopted as the loss function:
Wherein lambda is a super parameter for balancing positive and negative sample proportion, P (x) is a sigmoid function, and W represents all parameters to be learned; x i represents the feature vector of the current pixel;
The average value of 5 labeling in the pretreatment of the S2 data set is calculated, and is represented by y i, a new edge probability mapping diagram is generated, the range of the new edge probability mapping diagram is [0,1,0 ] which represents that no labeling is carried out in the 5 times, and 1 represents that the new edge probability mapping diagram is labeled as edge pixels in the 5 times of labeling;
Meanwhile, the edge probability value exceeds a threshold value eta as a positive sample Y +, and the edge probability value is smaller than eta as a negative sample Y -.
5. The edge detection method for a rail steel sheet protection cover according to claim 1, further comprising S5 edge judgment: and judging the distance and angle of the edges between the protective sheets of the protective cover according to a protective cover edge coordinate equation obtained by the single pixel edge image so as to determine whether the protective cover of the guide rail steel plate is in a normal running state.
6. The edge detection method for a rail steel sheet protection cover according to claim 5, wherein the S5 edge judgment comprises the steps of:
For a single-pixel edge image, an edge L is one end of a fixed guide rail steel plate protective cover, a coordinate system is established by taking the edge L of the fixed end of the guide rail steel plate protective cover as a y axis, and a coordinate equation of the edge L 1、L2、L3、…、Ln between all the pieces of the guide rail steel plate protective cover is obtained, wherein n is the number of protective pieces of the guide rail steel plate protective cover;
a) Acquiring angles alpha between the edges L 1、L2、L3、…、Ln and the fixed edges L respectively; if alpha is larger than delta, judging that the guide rail steel plate protective cover is abnormal, and alarming; wherein delta is the allowable error angle; otherwise, normal operation is performed;
B) The distance d 1、d2、d3、…、dn between two adjacent edges of the edges L, L 1、L2、L3、…、Ln is calculated, and it is determined whether the following formula is satisfied:
If the protection cover meets the protection cover requirement, judging that the protection cover of the guide rail steel plate is abnormal at the moment, and alarming; otherwise, normal operation is performed; where ε is the allowable error distance.
7. Edge detection device towards guide rail steel sheet protection casing, characterized in that includes:
the multidimensional image acquisition module is used for acquiring a color image of the guide rail steel plate protective cover in the working area through the camera according to the received image acquisition control signal and outputting the color image to the edge detection analysis module;
an edge detection analysis module comprising:
The image processing unit is used for acquiring edge information from the color image of the guide rail steel plate protective cover through the backbone network;
The edge image processing unit is used for carrying out non-maximum value inhibition processing on the obtained edge information to obtain a single-pixel edge image;
And the edge image analysis unit is used for judging according to the edge information in the single-pixel edge image to obtain the detection of whether the protective cover is normal or not.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211568856.XA CN118212170A (en) | 2022-12-08 | 2022-12-08 | Edge detection method and device for guide rail steel plate protection cover |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211568856.XA CN118212170A (en) | 2022-12-08 | 2022-12-08 | Edge detection method and device for guide rail steel plate protection cover |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118212170A true CN118212170A (en) | 2024-06-18 |
Family
ID=91452529
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211568856.XA Pending CN118212170A (en) | 2022-12-08 | 2022-12-08 | Edge detection method and device for guide rail steel plate protection cover |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118212170A (en) |
-
2022
- 2022-12-08 CN CN202211568856.XA patent/CN118212170A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112173636B (en) | Method for detecting faults of belt conveyor carrier roller by inspection robot | |
| CN101957325A (en) | Substation equipment appearance abnormality recognition method based on substation inspection robot | |
| CN109839386B (en) | Intelligent camera shooting identification system | |
| CN116308300B (en) | Power equipment state monitoring evaluation and command method and system | |
| CN106097380A (en) | A kind of system and method for core rod of insulator defect on-line checking based on image | |
| CN115144399B (en) | Assembly quality detection method and device based on machine vision | |
| CN116152202A (en) | Equipment appearance detection system based on image recognition technology and infrared thermal imaging technology | |
| CN117557570B (en) | Rail vehicle abnormality detection method and system | |
| CN117498225B (en) | Unmanned aerial vehicle intelligent power line inspection system | |
| CN113128555B (en) | Method for detecting abnormality of train brake pad part | |
| CN117372377B (en) | Broken line detection method and device for monocrystalline silicon ridge line and electronic equipment | |
| CN118212170A (en) | Edge detection method and device for guide rail steel plate protection cover | |
| CN115741218B (en) | Machine tool fault early warning method and system based on machining image analysis | |
| CN116862712A (en) | Electric power construction potential safety risk detection method and system based on thunder fusion | |
| CN115587978B (en) | Floor leather laminating embossing on-line measuring system based on degree of depth study | |
| CN115504187B (en) | Intelligent speed regulation and protection system and control method for mining belt | |
| CN108181315B (en) | Image processing-based biscuit damage detection device and detection method | |
| CN112508946B (en) | Cable tunnel anomaly detection method based on antagonistic neural network | |
| CN111105395B (en) | AI intelligent cradle head for monitoring power transmission operation | |
| CN112183342B (en) | Comprehensive convertor station defect identification method with template | |
| CN108956617A (en) | A kind of detection method of the electronic device solder joint thermal cycle failure based on micro- deformation Intelligence Classifier | |
| CN116429170B (en) | Quality detection method for plate blank | |
| CN117911415B (en) | Automatic equipment supervision system and method based on machine vision | |
| CN116579609B (en) | Illegal operation analysis method based on inspection process | |
| CN117765434B (en) | On-site video image processing method and system for constructional engineering |
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 |