CN112950545B - A method for detecting bending of bar steel - Google Patents
A method for detecting bending of bar steel Download PDFInfo
- Publication number
- CN112950545B CN112950545B CN202110146384.8A CN202110146384A CN112950545B CN 112950545 B CN112950545 B CN 112950545B CN 202110146384 A CN202110146384 A CN 202110146384A CN 112950545 B CN112950545 B CN 112950545B
- Authority
- CN
- China
- Prior art keywords
- pixel
- steel
- features
- image
- bending
- 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.)
- Active
Links
- 229910000831 Steel Inorganic materials 0.000 title claims abstract description 104
- 239000010959 steel Substances 0.000 title claims abstract description 104
- 238000005452 bending Methods 0.000 title claims abstract description 28
- 238000000034 method Methods 0.000 title claims abstract description 13
- 239000000463 material Substances 0.000 claims abstract description 58
- 230000011218 segmentation Effects 0.000 claims abstract description 15
- 238000013135 deep learning Methods 0.000 claims abstract description 14
- 238000001514 detection method Methods 0.000 claims abstract description 13
- 238000003708 edge detection Methods 0.000 claims abstract description 12
- 238000010438 heat treatment Methods 0.000 claims abstract description 11
- 230000006870 function Effects 0.000 claims description 6
- 238000011176 pooling Methods 0.000 claims description 6
- 230000000007 visual effect Effects 0.000 claims description 4
- 230000005540 biological transmission Effects 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 4
- 238000003709 image segmentation Methods 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
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
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-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
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- 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
- G06T2207/20081—Training; Learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Biomedical Technology (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Quality & Reliability (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
本发明公开了一种棒材钢料弯曲检测方法，该方法用耐高温摄像机实时拍摄传送带上将送入加热炉的棒材，然后用基于深度学习的分割网络实时分割出钢料的图像，然后对分割出的钢料图像通过边缘检测算法检测钢料的边界特征，从中提取出三条侧棱特征，通过侧棱特征判断钢料是否弯曲，阻止弯曲的钢料送入加热炉中，实现了棒材钢料弯曲自动化检测，避免弯曲钢料进入加热炉导致卡钢的问题。
The present invention discloses a method for detecting the bending of bar steel materials. The method uses a high-temperature resistant camera to shoot the bars to be sent into a heating furnace on a conveyor belt in real time, and then uses a segmentation network based on deep learning to segment the image of the steel materials in real time. Then, the segmented steel material image is subjected to an edge detection algorithm to detect the boundary features of the steel materials, and three side edge features are extracted therefrom. It is judged whether the steel materials are bent according to the side edge features, and the bent steel materials are prevented from being sent into the heating furnace, thereby realizing the automatic detection of the bending of bar steel materials and avoiding the problem of steel jamming caused by the bent steel materials entering the heating furnace.
Description
Technical Field
The invention belongs to the technical field of computer vision and image detection, and particularly relates to a method and a device for detecting bending of bar steel.
Background
The rolled bar steel material can produce bending deformation in the processing or conveying process, the bending deformation can damage subsequent conveying equipment, when the bent steel material is conveyed into a heating furnace, steel clamping can possibly occur, because the temperature of the heating furnace is extremely high, the steel clamping is difficult to take out, time and manpower and material resources can be seriously wasted, and even equipment failure is damaged. Therefore, before the steel is conveyed to the heating furnace, bending measurement is required to be carried out on the steel, when the steel is detected to be bent beyond a certain degree, the conveyor belt is prevented from being conveyed continuously, an alarm is sent to the front end, and a monitoring person or a mechanical arm is enabled to take off the bent steel.
Disclosure of Invention
The invention aims to solve the technical problem of providing a bar steel bending detection method aiming at the defects in the prior art.
In order to achieve the technical purpose, the invention adopts the following technical scheme:
a bar steel bending detection method comprises the following steps:
Step S1, fixing a high-temperature-resistant network camera above a bar steel material obliquely, and collecting an image of the steel material on a conveyor belt in real time;
S2, inputting the acquired images into a segmentation network based on deep learning, and segmenting images of bars in the images in real time;
s3, performing edge detection on the segmented steel material image by using an edge detection algorithm to extract steel material boundary characteristics and three side edge characteristics;
s4, detecting whether the steel is bent or not based on the three side edge characteristics;
and S5, sending an alarm to the front end when the bending of the steel is detected, and preventing the bent steel from being sent into the heating furnace.
In order to optimize the technical scheme, the specific measures adopted further comprise:
further, in the step S1, the high temperature resistant network camera collects two side surfaces and three side edges of the steel material.
Further, the step S2 specifically includes:
S21, inputting image data into a segmentation network based on deep learning, and judging whether each pixel belongs to steel materials or not;
s22, selecting all pixels judged to belong to the steel material in the image to form a steel material image;
S23, calculating the distance between the two farthest pixels in the steel image, and when the distance is smaller than a certain threshold value, judging that the steel does not completely enter the shooting visual field, and detecting the next image by turning to S21.
Further, the deep learning-based segmentation network in step S21 includes the steps of:
S211, extracting each pixel characteristic of the image by using a multi-layer sensor;
s212, carrying out maximum pooling on a3 multiplied by 3 neighborhood, a5 multiplied by 5 neighborhood and a 7 multiplied by 7 neighborhood which take each pixel as a center to obtain local features of 3 scales of each pixel, and linking the three features to form a multi-scale local feature of each pixel;
s213, extracting high-dimensional features of the multi-scale local features of each pixel by using a multi-layer perceptron, and then maximally pooling the high-dimensional features of all pixels to obtain global features;
S214, extracting weight from the high-dimensional feature of each pixel, linking the multi-scale local feature and the global feature of each pixel, and multiplying the multi-scale local feature and the global feature with the weight to obtain weighted features;
S215, the weighted characteristic of each pixel is reduced to one dimension through the full connection layer, the segmentation probability of each pixel is obtained through calculation by using a sigmoid function, a cross entropy loss function is calculated for training, and whether each pixel belongs to the pixel of steel materials or not is judged.
Further, the step S3 specifically includes:
s31, extracting boundary pixels from the segmented steel material image by using an edge detection algorithm, and constructing a minimum spanning tree by using a kruskal algorithm, wherein the boundary pixels are used as nodes of the tree;
s32, calculating a pixel unit normal vector according to the nearest k neighborhood of each boundary pixel;
S33, selecting one pixel in boundary pixels, and traversing all boundary pixels to obtain all break points when the absolute value of the inner product of the unit normal vector of any pixel in the k neighborhood of the pixel and the unit normal vector of the pixel is smaller than a set threshold value;
And S34, dividing the minimum spanning tree into a plurality of subtrees by all break points, and selecting three subtrees with the most nodes as three side edge features.
Further, step S4 specifically includes:
s41, respectively carrying out straight line detection on three side edge characteristics by using a RANSAC algorithm;
step S42, calculating the distance between all pixel points in the three side edge features and corresponding straight lines, and judging that the boundary line corresponding to the pixel is bent when the pixel with the distance larger than the set threshold exists;
And S43, bending the steel material when any one of the three side edge characteristics is bent.
Further, step S5 is specifically that an alarm is sent to the front end controller when the bending of the steel is detected, the steel conveyor belt is interrupted to transmit, and after the bent steel is taken down, the conveyor belt continues to convey the steel, and the next steel is detected.
The invention has the beneficial effects that:
The invention provides a method and a device for detecting steel bar bending based on an image segmentation algorithm and an image edge detection technology of deep learning. According to the method, an image of conveying steel materials is acquired by a high-temperature-resistant camera, the steel materials are segmented by a deep learning image segmentation algorithm, and the bending degree of the steel materials is detected by extracting boundary characteristics of the steel materials by an edge detection algorithm, so that the problem of steel material bending detection before the steel materials are conveyed into a heating furnace is solved, and subsequent accidents caused by conveying the bent steel materials into the heating furnace are avoided.
Drawings
FIG. 1 is a diagram of a method and apparatus for detecting bending of steel bar stock according to the present invention;
FIG. 2 is a schematic view of a shooting position of a high-temperature-resistant camera according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a segmentation network structure based on deep learning according to an embodiment of the present invention;
FIG. 4 is a schematic view of boundary features and three side edge features of a steel material according to an embodiment of the present invention.
Detailed Description
Embodiments of the present invention are described in further detail below with reference to the accompanying drawings.
As shown in fig. 1, the invention provides a bar steel bending detection method, which comprises the following steps:
Step S1, fixing a high-temperature-resistant network camera above a bar steel material obliquely, and collecting an image of the steel material on a conveyor belt in real time;
Because the detection environment is located beside the heating furnace, the environment temperature is higher, in order not to influence the image acquisition effect, the DS-NXCN A204 CMOS star light level high-temperature-resistant air-cooled barrel type network camera is used for acquiring images, when steel materials are bent, at least one surface (two side edges) is bent, so that whether the steel materials are bent or not can be judged by detecting three steel materials from the four side edges, a camera is fixed above a conveying belt of the steel materials of the bar, the two side surfaces (three side edges) of the steel materials can be completely located in the visual field of the camera, after the images are acquired, due to limited wireless transmission speed and higher environment temperature, special high-temperature-resistant cables are required to be used for transmitting image data, and fig. 2 is a schematic diagram of the shooting position of the high-temperature-resistant camera.
S2, inputting the acquired images into a segmentation network based on deep learning, and segmenting images of bars in the images in real time;
Inputting image data into a segmentation network based on deep learning, judging whether each pixel belongs to steel materials, then selecting all pixels judged to belong to the steel materials to form a steel material image, calculating the distance between the two farthest pixels in the steel material image, judging that the steel materials do not enter a shooting visual field completely when the distance is smaller than a certain threshold value, judging that the steel material image shooting is incomplete, and continuously detecting the next image, wherein fig. 3 is a segmentation network structure schematic diagram based on the deep learning;
wherein the deep learning based segmentation network comprises the steps of:
Step (1) extracting each pixel characteristic of an image by using a multi-layer sensor;
Step (2) carrying out maximum pooling on a3×3 neighborhood, a5×5 domain and a 7×7 domain which take each pixel as a center to obtain local features of 3 scales of each pixel, and linking the three features to form a multi-scale local feature of each pixel;
extracting high-dimensional features of multi-scale local features of each pixel by using a multi-layer perceptron, and then maximally pooling the high-dimensional features of all pixels to obtain global features;
Step (4) extracting weight from the high-dimensional feature of each pixel, linking the multi-scale local feature and the global feature of each pixel, and multiplying the multi-scale local feature and the global feature with the weight to obtain weighted features;
and (5) reducing the weighted characteristic of each pixel to one dimension through a full connection layer, obtaining the segmentation probability of each pixel by using a sigmoid function, calculating a cross entropy loss function for training, and judging whether each pixel belongs to the pixel of the steel material.
S3, performing edge detection on the segmented steel material image by using an edge detection algorithm to extract steel material boundary characteristics and three side edge characteristics;
Extracting boundary pixels from the segmented steel material image by using an edge detection algorithm, constructing a minimum spanning tree by using a kruskal algorithm, taking the boundary pixels as nodes of the tree, calculating the unit normal vector of the pixel according to the nearest k neighborhood of each boundary pixel, selecting one pixel in the boundary pixels, taking the pixel as a breakpoint when the inner product absolute value of the unit normal vector of any pixel in the k neighborhood of the pixel and the unit normal vector of the pixel is smaller than a set threshold value, traversing all the boundary pixels to obtain all the breakpoints, dividing the minimum spanning tree into a plurality of subtrees by all the breakpoints, and selecting three subtrees with the most nodes as three side edge features.
S4, detecting whether the steel is bent or not based on the three side edge characteristics;
the method comprises the steps of respectively carrying out straight line detection on three side edge features by using a RANSAC algorithm, calculating the distance between all pixel points in the three side edge features and corresponding straight lines, judging that a boundary line corresponding to a pixel is bent when the pixel with the distance larger than a set threshold exists, and bending steel materials when any one of the three side edge features is bent.
And S5, sending an alarm to the front-end controller when the bending of the steel is detected, interrupting the transmission of the steel conveyor belt, and continuously conveying the steel by the conveyor belt after the bent steel is taken down, so as to detect the next steel.
The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above examples, and all technical solutions belonging to the concept of the present invention belong to the protection scope of the present invention. It should be noted that modifications and adaptations to the invention without departing from the principles thereof are intended to be within the scope of the invention as set forth in the following claims.
Claims (5)
1. The steel bar bending detection method is characterized by comprising the following steps of:
Step S1, fixing a high-temperature-resistant network camera above a bar steel material obliquely, and collecting an image of the steel material on a conveyor belt in real time;
S2, inputting the acquired images into a segmentation network based on deep learning, and segmenting images of bars in the images in real time;
s3, performing edge detection on the segmented steel material image by using an edge detection algorithm to extract steel material boundary characteristics and three side edge characteristics;
s4, detecting whether the steel is bent or not based on the three side edge characteristics;
S5, sending an alarm to the front end when the bending of the steel is detected, and preventing the bent steel from being sent into the heating furnace;
The step S3 specifically comprises the following steps:
s31, extracting boundary pixels from the segmented steel material image by using an edge detection algorithm, and constructing a minimum spanning tree by using a kruskal algorithm, wherein the boundary pixels are used as nodes of the tree;
Step S32, calculating the pixel unit normal vector according to the nearest k neighborhood of each boundary pixel;
step S33, selecting one pixel in boundary pixels, and traversing all boundary pixels to obtain all break points when the absolute value of the inner product of the unit normal vector of any pixel in the k neighborhood of the pixel and the unit normal vector of the pixel is smaller than a set threshold value;
Step S34, dividing the minimum spanning tree into a plurality of subtrees by all break points, and selecting three subtrees with the most nodes as three side edge features;
The step S4 specifically includes:
s41, respectively carrying out straight line detection on three side edge characteristics by using a RANSAC algorithm;
step S42, calculating the distance between all pixel points in the three side edge features and corresponding straight lines, and judging that the boundary line corresponding to the pixel is bent when the pixel with the distance larger than the set threshold exists;
And S43, bending the steel material when any one of the three side edge characteristics is bent.
2. The method for detecting bending of steel bar according to claim 1, wherein the high-temperature-resistant network camera in the step S1 is used for collecting two side surfaces and three side edges of the steel bar.
3. The method for detecting bending of steel bar according to claim 2, wherein the step S2 is specifically:
s21, inputting image data into a segmentation network based on deep learning, and judging whether each pixel belongs to steel materials or not;
S22, selecting all pixels judged to belong to the steel material in the image to form a steel material image;
And S23, calculating the distance between the two farthest pixels in the steel image, and when the distance is smaller than a certain threshold value, judging that the steel does not completely enter the shooting visual field, and detecting the next image by turning to S21.
4. The method for detecting bending of steel bar according to claim 3, wherein the deep learning-based segmentation network in the step S21 comprises the following steps:
Step S211, extracting each pixel characteristic of the image by using the multi-layer perceptron;
Step S212, carrying out maximum pooling on a 3 multiplied by 3 neighborhood, a 5 multiplied by 5 neighborhood and a 7 multiplied by 7 neighborhood which take each pixel as a center to obtain local features of 3 scales of each pixel, and linking the three features to form a multi-scale local feature of each pixel;
Step S213, extracting high-dimensional features from the multi-scale local features of each pixel by using a multi-layer sensor, and then maximally pooling the high-dimensional features of all pixels to obtain global features;
step S214, extracting weight from the high-dimensional feature of each pixel, linking the multi-scale local feature and the global feature of each pixel, and multiplying the multi-scale local feature and the global feature with the weight to obtain weighted features;
And S215, reducing the weighted characteristic of each pixel to one dimension through a full connection layer, calculating the segmentation probability of each pixel by using a sigmoid function, calculating a cross entropy loss function for training, and judging whether each pixel belongs to the pixel of the steel material.
5. The method for detecting bending of steel bar according to claim 4, wherein the step S5 is specifically to send an alarm to the front end controller when bending of the steel bar is detected, interrupt transmission of the steel bar conveyor belt, and continue conveying the steel bar by the conveyor belt after the bent steel bar is removed for detection of the next steel bar.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110146384.8A CN112950545B (en) | 2021-02-03 | 2021-02-03 | A method for detecting bending of bar steel |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110146384.8A CN112950545B (en) | 2021-02-03 | 2021-02-03 | A method for detecting bending of bar steel |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112950545A CN112950545A (en) | 2021-06-11 |
| CN112950545B true CN112950545B (en) | 2024-12-13 |
Family
ID=76241944
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110146384.8A Active CN112950545B (en) | 2021-02-03 | 2021-02-03 | A method for detecting bending of bar steel |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112950545B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115496754B (en) * | 2022-11-16 | 2023-04-11 | 深圳佰维存储科技股份有限公司 | Curvature detection method and device of SSD, readable storage medium and electronic equipment |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102441581A (en) * | 2010-09-30 | 2012-05-09 | 邓玥 | Device and method for online detection of section size of profile steel based on machine vision |
| CN110648316A (en) * | 2019-09-07 | 2020-01-03 | 创新奇智(成都)科技有限公司 | Steel coil end face edge detection algorithm based on deep learning |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5233372B2 (en) * | 2008-04-03 | 2013-07-10 | 新日鐵住金株式会社 | Steel plate warpage detection system and method |
| JP5828817B2 (en) * | 2012-09-03 | 2015-12-09 | 株式会社神戸製鋼所 | Shape inspection method for steel bars |
| CN110926378A (en) * | 2019-12-16 | 2020-03-27 | 太原科技大学 | An improved bar straightness detection system and method based on visual inspection |
| CN211783345U (en) * | 2020-03-16 | 2020-10-27 | 成都桐林铸造实业有限公司 | Curvature detection device |
-
2021
- 2021-02-03 CN CN202110146384.8A patent/CN112950545B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102441581A (en) * | 2010-09-30 | 2012-05-09 | 邓玥 | Device and method for online detection of section size of profile steel based on machine vision |
| CN110648316A (en) * | 2019-09-07 | 2020-01-03 | 创新奇智(成都)科技有限公司 | Steel coil end face edge detection algorithm based on deep learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112950545A (en) | 2021-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110171691A (en) | Belt conveyor belt tearing condition detection method and detection system | |
| CN110287963B (en) | OCR recognition method for comprehensive performance test | |
| CN111369516B (en) | Detection method of transformer bushing heating defects based on infrared image recognition | |
| CN101957325A (en) | Substation equipment appearance abnormality recognition method based on substation inspection robot | |
| CN107084992A (en) | A capsule detection method and system based on machine vision | |
| CN118608504A (en) | A method and system for detecting component surface quality based on machine vision | |
| CN112950545B (en) | A method for detecting bending of bar steel | |
| CN114529839A (en) | Unmanned aerial vehicle routing inspection-oriented power transmission line hardware anomaly detection method and system | |
| CN110493574B (en) | Security monitoring visualization system based on streaming media and AI technology | |
| CN114092478A (en) | Anomaly detection method | |
| CN116309447B (en) | Dam slope crack detection method based on deep learning | |
| CN114612403A (en) | A kind of intelligent detection method and system for damage defect of feeding belt | |
| CN115049955A (en) | Fire detection analysis method and device based on video analysis technology | |
| CN118183215A (en) | AI technology-based coal conveying belt abnormality detection system | |
| CN115861210A (en) | Transformer substation equipment abnormity detection method and system based on twin network | |
| CN108615057B (en) | CNN-based abnormity identification method for cable tunnel lighting equipment | |
| CN117252840B (en) | Photovoltaic array defect elimination assessment method, device and computer equipment | |
| CN117029703B (en) | Communication cable field production data real-time management monitoring system | |
| CN109977962B (en) | Method and system for automatically identifying fault hidden danger of optical cable | |
| CN111325073A (en) | Monitoring video abnormal behavior detection method based on motion information clustering | |
| CN112907524B (en) | Method for detecting fault of fire-proof plate of rail wagon based on image processing | |
| CN116002480A (en) | Automatic detection method and system for accidental fall of passengers in elevator car | |
| CN102514771A (en) | Industrial explosive roll transmission attitude identification and diagnosis system and method thereof | |
| Wang et al. | Thermal defect detection and location for power equipment based on improved VGG16 | |
| CN110930362A (en) | Screw safety detection method, device 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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20210903 Address after: Room 2605, no.1698, Shuanglong Avenue, Jiangning District, Nanjing City, Jiangsu Province 211106 Applicant after: Nanjing Yuntong Technology Co.,Ltd. Applicant after: Suzhou Yuntong Technology Co.,Ltd. Address before: Room 2605, no.1698, Shuanglong Avenue, Jiangning District, Nanjing City, Jiangsu Province 211106 Applicant before: Nanjing Yuntong Technology Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |