CN107729906B - Intelligent robot-based inspection point ammeter numerical value identification method - Google Patents
Intelligent robot-based inspection point ammeter numerical value identification method Download PDFInfo
- Publication number
- CN107729906B CN107729906B CN201710998464.XA CN201710998464A CN107729906B CN 107729906 B CN107729906 B CN 107729906B CN 201710998464 A CN201710998464 A CN 201710998464A CN 107729906 B CN107729906 B CN 107729906B
- Authority
- CN
- China
- Prior art keywords
- image
- standard
- area
- ammeter
- inspection point
- 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
- 238000000034 method Methods 0.000 title claims abstract description 25
- 238000007689 inspection Methods 0.000 title claims description 29
- 238000012549 training Methods 0.000 claims abstract description 15
- 238000013507 mapping Methods 0.000 claims abstract description 8
- 230000003321 amplification Effects 0.000 claims description 2
- 238000003199 nucleic acid amplification method Methods 0.000 claims description 2
- 230000001105 regulatory effect Effects 0.000 claims 1
- 238000005516 engineering process Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- 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
- 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
-
- 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
- G06V10/48—Extraction of image or video features by mapping characteristic values of the pattern into a parameter space, e.g. Hough transformation
-
- 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
- G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
- G06V10/507—Summing image-intensity values; Histogram projection analysis
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C1/00—Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people
- G07C1/20—Checking timed patrols, e.g. of watchman
-
- 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
- G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
- G06V10/467—Encoded features or binary features, e.g. local binary patterns [LBP]
Abstract
The invention relates to a patrol point ammeter numerical identification method based on an intelligent robot, and belongs to the technical field of image identification. The method comprises the following steps: 1) shooting a standard image; 2) selecting a standard area; 3) establishing an ammeter scale set; 4) generating a training data set; 5) training is carried out; 6) acquiring an image to be identified; 7) selecting small blocks of an image to be identified; 8) determining the most similar image block; 9) determining a mapping relation; 10) determining an ammeter area in an image to be identified; 11) acquiring a binary image; 12) fitting the binary image to determine an ammeter scale pointer; 13) and determining the electric meter value. The method has the advantages of strong positioning error tolerance, high instrument identification accuracy, and the instrument identification degree is about 60 percent, and the accuracy is improved to about 99 percent by adopting the method.
Description
Technical Field
The invention relates to a patrol point ammeter numerical identification method based on an intelligent robot, and belongs to the technical field of image identification.
Background
Along with the development of science and technology, the degree of intellectuality deepens, and mechanized intelligent equipment replaces the manual work gradually and is applied to mechanized loaded down with trivial details work. Just like power equipment's the work of patrolling and examining, adopt the intelligence to patrol and examine the robot and patrol and examine the work, solved the potential safety hazard problem and the artifical mistake and/or wrong problem that patrol and examine and probably bring that are brought by the special environment of power supply unit room.
Because of the immature technology, the mechanical error [ -1Rad,1Rad ] of the cradle head in the rotating process can not shoot the image; the tolerance of the instrument recognition algorithm to the position error of picture shooting is low, so that the instrument recognition rate is reduced. Therefore, it is desirable to provide a collection and identification method with high accuracy and capable of adapting to the existing equipment, so as to provide reliable reference for ledger records and alarms.
Disclosure of Invention
The invention aims to solve the technical problem of providing a patrol point electric meter numerical value identification method based on an intelligent robot with high accuracy aiming at the defects of the prior art.
The technical scheme provided by the invention for solving the technical problems is as follows: a patrol inspection point ammeter value identification method based on an intelligent robot executes the following steps:
1) shooting a standard image of the inspection point;
2) selecting a standard area from the standard image; the standard area is between [400 × 300, 800 × 700] pixels, and the electric meter area of the inspection point is contained in the standard area;
3) establishing an ammeter scale set (K, SV) of a patrol point, wherein K ϵ (1 … n) and SV ϵ [0,2 x pi ], K is the slope of a straight line where a pointer of the ammeter is located, and SV is a scale value;
4) generating a training data set based on the standard image;
5) bringing the training data set into a kd tree for training, and meanwhile, calculating corner point characteristics of a standard area;
6) acquiring an image to be identified of a patrol point through the intelligent robot;
7) selecting a plurality of small blocks of the image to be identified from the image to be identified;
8) bringing the set of the small blocks of the image to be recognized into the trained kd tree, and selecting the small block of the image to be recognized with the minimum vector distance with the standard area from the set of the small blocks of the image to be recognized as the most similar image block by a nearest neighbor search method;
9) matching the corner features of the most similar image blocks with the corner features of the standard area to obtain a mapping relation between the most similar image blocks and the standard area;
10) determining an ammeter area in the image to be identified according to the mapping relation;
11) graying and binarizing the ammeter area in the step 10) to obtain a processed binarized image;
12) finding a line segment in the binary image through Hough line segment fitting, and selecting the line segment with the shortest distance from the pointer rotation center as an ammeter scale pointer;
13) determining the electric meter numerical value of the inspection point through the slope and the scale set of the electric meter scale pointer;
before shooting in the step 1), setting the magnification factor, the focusing value, the horizontal angle of a holder, the vertical angle of the holder and the position coordinate of the inspection point of a camera corresponding to the inspection point; and 6) before the image to be identified is obtained, moving to a corresponding inspection point according to the position coordinate of the inspection point, and setting according to the amplification factor, the focusing value, the horizontal angle of the holder and the vertical angle of the holder set in the step 1).
The improvement of the technical scheme is as follows: the standard image has a size of 1920 pixels by 1080 pixels.
The improvement of the technical scheme is as follows: and 7) taking out small image blocks at a distance of 32 pixels in the image to be identified.
The improvement of the technical scheme is as follows: and 4) carrying out blocking processing on the standard image at an interval of 2 pixels to generate a training image small block, wherein the length and the width of the training image small block are the same as those of the image small block to be identified.
The invention adopts the technical scheme that the method has the beneficial effects that: according to the invention, the problem that images cannot be shot due to mechanical errors [ -1Rad,1Rad ] in the rotation process of the cloud deck is solved by determining the horizontal angle of the cloud deck and the vertical angle of the cloud deck corresponding to each inspection point; according to the invention, the problem that the image cannot be shot due to the error of the robot body in the moving process is solved by determining the position of the inspection point during shooting; the problem that the tolerance of the instrument recognition algorithm to the position error of picture shooting is low, so that the instrument recognition rate is reduced is solved through target object searching and matching.
Compared with the common fixed position instrument identification method, the method has the advantages of strong positioning error tolerance, high instrument identification accuracy, and the instrument identification degree is about 60 percent, and the accuracy is improved to about 99 percent by adopting the method.
Detailed Description
Examples
The inspection point electric meter value identification method based on the intelligent robot comprises the following steps:
1) shooting a standard image ImgStd of a patrol point; the robot autonomously moves to a specified position of a room, sets the magnification Z of the camera, the focusing value F, the horizontal angle H of the holder and the vertical angle V of the holder corresponding to the corresponding inspection point, and records the position coordinates (PosX, PosY) of the current inspection point in the room.
2) Selecting a standard region ImgStdArea from the standard image ImgStd; the standard region imgtsdarea is between [400 × 300, 800 × 700] pixels in size, and the electric meter region of the inspection point is included in the standard region imgtsdarea.
3) Establishing an ammeter scale set (K, SV) of a patrol point, wherein K ϵ (1 … n) and SV ϵ [0,2 x pi ], K is the slope of a straight line where a pointer of the ammeter is located, SV is a scale value, and the scale number set of the meter is (1 … n); k = (PRY-PCY)/(PRX-PCX), and the pointer rotation center position is (PCX, PCY) and the farthest position of each scale from the pointer rotation center is (PRX, PRY).
4) Generating a training data set based on the standard image ImgStd; the standard size of the standard image imgtred is 1920 × 1080, the standard area imgtatra ϵ imgtred is in a rectangle (x, y, w, h), wherein (x, y) is the coordinate position of the upper left corner of the standard area imgtatra, w is the width of the standard area imgtatra, and h is the height of the standard area imgtatra. The standard region imgtstdarea was used as a positive sample, and the other regions were used as negative samples at intervals of 2 pixels. The area where the negative samples are located is (Nx, Ny, Nw, Nh), the following condition should be satisfied:
(Nx > =0) ^ (Nx + Nw) < (x +1/3 w)) (Nx > = (x +2/3 w)) ^ (Nx + Nw) < 1920) ^ (Nx = x +2 n). Wherein n is an integer.
② (v/v ((Ny > =0) ^ (Ny + Nh) < (x + 1/3) ×) ((Ny > = (x + 2/3) ×) (Ny + Nh) < 1920) ^ (Ny = y + 2) × n). Wherein n is an integer.
③ Nw = w
④ Nh = h
For each sample graph with the size of (w, h), the sample graphs are unified into 64 × 64 image small blocks, and then HOG + LBP features of the small blocks are calculated to form vector features with the dimension of 8100, wherein the vector features are training data sets.
5) And bringing the training data set into the kd tree for training to obtain a model file for identifying the standard area, and storing a model file ModelKd record. And simultaneously calculating the corner feature of the standard area. Converting the 3-channel RGB image of the standard region ImgStdArea into a single-channel gray-scale image, calculating all corner features of the standard region ImgStdArea, and recording the corner features to a file Modelfeature.
6) Acquiring an image to be identified of a patrol inspection point through an intelligent robot; and (3) informing the robot to patrol and examine a specified patrol and examine point, moving the robot to a specified coordinate according to preset parameters, setting specified parameters such as a magnification factor, a focusing value and the like, and then shooting an image ImgSmp to be recognized according to the method in the step 1.
7) Selecting a plurality of small blocks of the image to be recognized from the image to be recognized ImgSmp; and taking out small blocks of the image to be recognized at intervals of 32 pixels in the image to be recognized ImgSmp, wherein the length and the height of the small blocks of the image to be recognized are consistent with the width and the height of the sample in the calibration process, and the rectangles corresponding to the small blocks of the image to be recognized are (Sx, Sy, Sw and Sh), Sw = w and Sh = h.
8) And bringing the set of the small blocks of the image to be recognized into the trained kd tree, and selecting the small block of the image to be recognized with the minimum vector distance with the standard area from the set of the small blocks of the image to be recognized as the most similar image block by using a nearest neighbor search method.
9) Matching the corner features of the most similar image blocks with the corner features of the standard area to obtain a mapping relation between the most similar image blocks and the standard area;
all corner features within the image patch to be identified are calculated and described using a corner feature descriptor. Matching the corner features of the most similar image blocks with the corner features of the standard area, finding out 5 pairs of feature points with the shortest Euclidean distance between the features to form matched feature points, wherein half of the matched feature points come from the standard area of the most similar image blocks, and establishing a mapping relation between the two images by using coordinates of the features
10) Determining an ammeter area in the image to be identified according to the mapping relation; denoted as image patch ImgSmpFin.
11) Graying and binarizing the ammeter area in the step 10) to obtain a processed binarized image; graying the small image block ImgSmpFan, recording the small image block as a grayscale image ImgSmpGry, calculating a grayscale histogram of the grayscale image ImgSmpGry, dividing the grayscale histogram into 10 parts on a horizontal axis [0,255], finding out an interval with the maximum number of pixels in the interval, recording the interval as [ Amin, Amax ], and carrying out binarization processing on the grayscale image ImgSmpGry according to a threshold Tv = (Amin + Amax)/2 to obtain a binarized image ImgSmpBin.
12) Finding a line segment in the binary image ImgSmpBin through Hough line segment fitting, and selecting the line segment with the shortest distance from the rotation center of the pointer as an ammeter scale pointer;
13) and determining the electric meter value of the patrol inspection point through the slope of the electric meter scale pointer and the scale set.
After the electric meter numerical value identification of the inspection point is completed, the inspection background system records the readings of the voltmeter and the ammeter into a database; the inspection background system judges whether the identification result is in a normal range according to preset voltage and current limit values, if not, an alarm message is popped up on a system interface, and related personnel are notified by short messages; and querying and identifying historical data and pictures through a historical query tool interface.
The present invention is not limited to the above-described embodiments. All technical solutions formed by equivalent substitutions fall within the protection scope of the claims of the present invention.
Claims (2)
1. A patrol point ammeter value identification method based on an intelligent robot is characterized by comprising the following steps:
1) shooting a standard image of the inspection point;
2) selecting a standard area from the standard image; the standard area is between [400 × 300, 800 × 700] pixels, and the electric meter area of the inspection point is contained in the standard area;
3) establishing an ammeter scale set (K, SV) of a patrol point, wherein K ϵ (1 … n) and SV ϵ [0,2 x pi ], K is the slope of a straight line where a pointer of the ammeter is located, and SV is a scale value;
4) a set of training data is generated based on the standard images,
the rectangle of the standard area is (x, y, w, h), wherein (x, y) is the coordinate position of the upper left corner of the standard area, w is the width of the standard area, and h is the height of the standard area;
taking the standard area as a positive sample, and taking the area of other parts as a negative sample at the interval of 2 pixels;
the area of the negative sample is (Nx, Ny, Nw, Nh), and the following conditions are satisfied:
(Nx > =0) ^ (Nx + Nw) < (x +1/3 w)) (Nx > = (x +2/3 w)) ^ (Nx + Nw) < 1920) ^ (Nx = x +2 n), where n is an integer;
(n =0) ^ (Ny + Nh) < (x +1/3 × h)) (Ny > = (x +2/3 × h)) Λ (Ny + Nh) < 1920) ^ (Ny = y +2 × n), where n is an integer;
③ Nw = w；
④ Nh = h；
uniformly regulating the images of the samples with the size (w, h) into 64 × 64 image small blocks, and then calculating HOG + LBP characteristics of the image small blocks to form 8100-dimensional vector characteristics which serve as the training data set;
5) bringing the training data set into a kd tree for training to obtain a model file for identifying the standard area, and meanwhile, calculating the corner point characteristics of the standard area;
6) acquiring an image to be identified of a patrol point through the intelligent robot;
7) selecting a plurality of small blocks of the image to be recognized at intervals of 32 pixels from the image to be recognized;
8) bringing the set of the small blocks of the image to be recognized into the trained kd tree, and selecting the small block of the image to be recognized with the minimum vector distance with the standard area from the set of the small blocks of the image to be recognized as the most similar image block by a nearest neighbor search method;
9) matching the corner features of the most similar image blocks with the corner features of the standard area to obtain a mapping relation between the most similar image blocks and the standard area;
10) determining an ammeter area in the image to be identified according to the mapping relation;
11) graying and binarizing the ammeter area in the step 10) to obtain a processed binarized image;
12) finding a line segment in the binary image through Hough line segment fitting, and selecting the line segment with the shortest distance from the pointer rotation center as an ammeter scale pointer;
13) determining the electric meter numerical value of the inspection point through the slope and the scale set of the electric meter scale pointer;
before shooting in the step 1), setting the magnification factor, the focusing value, the horizontal angle of a holder, the vertical angle of the holder and the position coordinate of the inspection point of a camera corresponding to the inspection point; and 6) before the image to be identified is obtained, moving to a corresponding inspection point according to the position coordinate of the inspection point, and setting according to the amplification factor, the focusing value, the horizontal angle of the holder and the vertical angle of the holder set in the step 1).
2. The inspection point electric meter value identification method based on the intelligent robot according to claim 1, characterized in that: the standard image has a size of 1920 pixels by 1080 pixels.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710998464.XA CN107729906B (en) | 2017-10-24 | 2017-10-24 | Intelligent robot-based inspection point ammeter numerical value identification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710998464.XA CN107729906B (en) | 2017-10-24 | 2017-10-24 | Intelligent robot-based inspection point ammeter numerical value identification method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107729906A CN107729906A (en) | 2018-02-23 |
| CN107729906B true CN107729906B (en) | 2021-11-02 |
Family
ID=61213646
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710998464.XA Active CN107729906B (en) | 2017-10-24 | 2017-10-24 | Intelligent robot-based inspection point ammeter numerical value identification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107729906B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108616723A (en) * | 2018-04-20 | 2018-10-02 | 国网江苏省电力有限公司电力科学研究院 | A kind of video routing inspection system for GIL piping lanes |
| CN110378257B (en) * | 2019-07-04 | 2023-12-19 | 山东巧思智能科技有限公司 | Artificial intelligent model whole process automation system |
| CN112562112A (en) * | 2020-11-16 | 2021-03-26 | 深圳市长龙铁路电子工程有限公司 | Automatic inspection method and system |
| CN113450384B (en) * | 2021-06-11 | 2023-12-29 | 力源电力设备股份有限公司 | Pointer type meter physical information reading method based on coding mark information |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3998215B1 (en) * | 2007-03-29 | 2007-10-24 | 国立大学法人山口大学 | Image processing apparatus, image processing method, and image processing program |
| CN103034838A (en) * | 2012-12-03 | 2013-04-10 | 中国人民解放军63963部队 | Special vehicle instrument type identification and calibration method based on image characteristics |
| CN103927507A (en) * | 2013-01-12 | 2014-07-16 | 山东鲁能智能技术有限公司 | Improved multi-instrument reading identification method of transformer station inspection robot |
| CN104899609A (en) * | 2015-06-19 | 2015-09-09 | 四川大学 | Image registration-based mechanical meter identification method |
| CN105913095A (en) * | 2016-05-17 | 2016-08-31 | 杭州申昊科技股份有限公司 | Instrument recognition method for transformer substation patrol inspection robot |
| CN105930837A (en) * | 2016-05-17 | 2016-09-07 | 杭州申昊科技股份有限公司 | Transformer station instrument equipment image recognition method based on autonomous routing inspection robot |
| CN105975979A (en) * | 2016-04-22 | 2016-09-28 | 浙江大学 | Instrument detection method based on machine vision |
| CN106709452A (en) * | 2016-12-23 | 2017-05-24 | 浙江大学 | Instrument position detection method based on intelligent inspection robot |
| CN106778823A (en) * | 2016-11-22 | 2017-05-31 | 国网湖北省电力公司宜昌供电公司 | A kind of readings of pointer type meters automatic identifying method |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104463240B (en) * | 2013-09-23 | 2017-12-22 | 深圳市朗驰欣创科技有限公司 | A kind of instrument localization method and device |
| US9958474B2 (en) * | 2015-09-14 | 2018-05-01 | Denso International America, Inc. | Self-calibrating indicating device |
-
2017
- 2017-10-24 CN CN201710998464.XA patent/CN107729906B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3998215B1 (en) * | 2007-03-29 | 2007-10-24 | 国立大学法人山口大学 | Image processing apparatus, image processing method, and image processing program |
| CN103034838A (en) * | 2012-12-03 | 2013-04-10 | 中国人民解放军63963部队 | Special vehicle instrument type identification and calibration method based on image characteristics |
| CN103927507A (en) * | 2013-01-12 | 2014-07-16 | 山东鲁能智能技术有限公司 | Improved multi-instrument reading identification method of transformer station inspection robot |
| CN104899609A (en) * | 2015-06-19 | 2015-09-09 | 四川大学 | Image registration-based mechanical meter identification method |
| CN105975979A (en) * | 2016-04-22 | 2016-09-28 | 浙江大学 | Instrument detection method based on machine vision |
| CN105913095A (en) * | 2016-05-17 | 2016-08-31 | 杭州申昊科技股份有限公司 | Instrument recognition method for transformer substation patrol inspection robot |
| CN105930837A (en) * | 2016-05-17 | 2016-09-07 | 杭州申昊科技股份有限公司 | Transformer station instrument equipment image recognition method based on autonomous routing inspection robot |
| CN106778823A (en) * | 2016-11-22 | 2017-05-31 | 国网湖北省电力公司宜昌供电公司 | A kind of readings of pointer type meters automatic identifying method |
| CN106709452A (en) * | 2016-12-23 | 2017-05-24 | 浙江大学 | Instrument position detection method based on intelligent inspection robot |
Non-Patent Citations (3)
| Title |
|---|
| A Robust Pointer Meter Reading Recognition Method for Substation Inspection Robot;Jia-Wei Gao 等;《2017 International Conference on Robotics and Automation Sciences》;20170829;43-47 * |
| 基于视觉显著性的指针式仪表读数识别算法;张文杰 等;《计算机辅助设计与图形学学报》;20151231;第27卷(第12期);2282-2295 * |
| 巡检机器人中的指针式仪表读数识别系统;许丽 等;《仪器仪表学报》;20170731;第38卷(第7期);1782-1790 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107729906A (en) | 2018-02-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107729906B (en) | Intelligent robot-based inspection point ammeter numerical value identification method | |
| CN111445517A (en) | Robot vision end positioning method and device and computer readable storage medium | |
| CN110363798B (en) | Method for generating remote sensing image interpretation sample set | |
| CN114241364A (en) | Method for quickly calibrating foreign object target of overhead transmission line | |
| CN110108712A (en) | Multifunctional visual sense defect detecting system | |
| LU500407B1 (en) | Real-time positioning method for inspection robot | |
| CN105548216A (en) | Visual detection method for appearance of semi-finished battery | |
| CN113985830A (en) | Feeding control method and device for sealing nail, electronic equipment and storage medium | |
| CN110738205A (en) | machine vision-based battery cell positive and negative electrode identification equipment and method | |
| CN111784655A (en) | Underwater robot recovery positioning method | |
| CN112163517A (en) | Underwater imaging fish net damage identification method and system based on deep learning | |
| CN113706455B (en) | Rapid detection method for damage of 330kV cable porcelain insulator sleeve | |
| CN116978834B (en) | Intelligent monitoring and early warning system for wafer production | |
| CN114227717A (en) | Intelligent inspection method, device, equipment and storage medium based on inspection robot | |
| CN113744269B (en) | Method and device for detecting welding quality of cylindrical battery cell, electronic equipment and storage medium | |
| CN111652069A (en) | Target identification and positioning method of mobile robot | |
| Jiang et al. | Mobile robot gas source localization via top-down visual attention mechanism and shape analysis | |
| CN115984759A (en) | Substation switch state identification method and device, computer equipment and storage medium | |
| CN115902977A (en) | Transformer substation robot double-positioning method and system based on vision and GPS | |
| CN110322508B (en) | Auxiliary positioning method based on computer vision | |
| CN110310239B (en) | Image processing method for eliminating illumination influence based on characteristic value fitting | |
| CN107356232A (en) | A kind of vision detection system image processing method | |
| CN114022763A (en) | Foreign matter detection method and device for high-voltage overhead line and readable storage medium | |
| CN112767433A (en) | Automatic deviation rectifying, segmenting and identifying method for image of inspection robot | |
| CN110414470B (en) | Inspection method based on terahertz and visible light |
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 |