CN109360289A - Merge the electric power meter detection method of crusing robot location information - Google Patents
Merge the electric power meter detection method of crusing robot location information Download PDFInfo
- Publication number
- CN109360289A CN109360289A CN201811148567.8A CN201811148567A CN109360289A CN 109360289 A CN109360289 A CN 109360289A CN 201811148567 A CN201811148567 A CN 201811148567A CN 109360289 A CN109360289 A CN 109360289A
- Authority
- CN
- China
- Prior art keywords
- image
- crusing robot
- instrument
- object candidate
- location information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- 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
- G06V10/42—Global feature extraction by analysis of the whole pattern, e.g. using frequency domain transformations or autocorrelation
- G06V10/431—Frequency domain transformation; Autocorrelation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/20—Scenes; Scene-specific elements in augmented reality scenes
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of electric power meter detection methods for merging crusing robot location information.The present invention is broadly divided into 4 steps: (1) crusing robot reaches designated position and obtains picture；(2) crusing robot location information is combined, coarse positioning is carried out to target instrument region using Fourier and phase coherent techniques；(3) it is accurately positioned using the target instrument region of machine learning, obtains multiple object candidate areas；(4) crusing robot location information is merged, the characteristic parameter of the multiple features fusions such as IoU, mutual information, vision Hash is calculated, candidate region is waited using the choice of parameters, obtains final goal.The present invention utilizes machine learning, can detect the instrument under a variety of illumination, attitudes vibration.
Description
Technical field
The present invention relates to object detection methods, and in particular to a kind of electric power meter for merging crusing robot location information
Detection method.
Background technique
Electric inspection process robot needs to realize autonomous localization and navigation in substation, the identification of field instrument registration, fills automatically
The basic functions such as electricity.Wherein the function of core is exactly the registration of the instrument and meter of detection and the live power equipment of identification, is such as discharged
The instrument registration such as counter, oil level gauge, voltmeter, thermometer, these instrument are substantially mechanical instrument, need robot according to
The reading of instrument registration is carried out by visual sensor.And the premise for accurately identifying instrument registration is accurately to detect instrument in visual pattern
The position of table, and most of instrument, in outdoor, most methods, are detected and are known using traditional image processing means at present
Not, in the case where illumination condition changes, detection effect is bad, as soon as a kind of general illumination condition just needs a group parameter, this is needed
It proposes a kind of more general detection and recognition methods, copes with the instrument Detection task under the conditions of different illumination, posture.
Summary of the invention
The object of the present invention is to provide a kind of electric power meter detection method for merging crusing robot location information, solutions
Robot location's not timing present in certainly existing instrument detection technique, target scale, angle change are big, and target is illuminated by the light influence
Greatly so as to cause the problem of detection inaccuracy.
Realize the technical solution of the object of the invention are as follows: a kind of electric power meter detection side for merging crusing robot location information
Method, specific steps are as follows:
Step 1 chooses one in inspection point bat using Instrument image data set training classifier, and for each inspection point
As template image, crusing robot reaches specified inspection point and obtains instrument picture to be detected the instrument taken the photograph image placed in the middle；
Step 2 treats the target gauge field detected in instrument picture using plum forests Fourier transformation and phase coherent techniques
Domain carries out coarse positioning；
Step 3 is accurately positioned target instrument region using the method for machine learning, and image to be detected is sent into and is instructed
The classifier practiced obtains several object candidate areas；
Step 4, the perceptual hash, mutual information and the friendship that calculate multiple candidate regions and three kinds of parameter indexes of ratio, screen target
Candidate region obtains final goal.
Preferably, perceptual hash, mutual information and the friendship of multiple candidate regions and three kinds of parameter indexes of ratio are calculated in step 4,
Screening object candidate area obtains the specific steps of final goal are as follows:
Step 4-1, each object candidate area and coarse positioning target instrument region are asked respectively and is handed over and than parameter IOU；
Step 4-2, the instrument area image in each object candidate area image and template image perception is done respectively to breathe out
It is uncommon to calculate, obtain perceptual hash index；
Step 4-3, the mutual information index of each object candidate area image and template image is calculated separately；
Step 4-4, by the friendship of each object candidate area and than three kinds of IOU, mutual information index and perceptual hash indexs
The confidence level that weighting finds out each object candidate area is done, the maximum object candidate area of confidence level is alternately detected into knot
Fruit；
If the IOU of step 4-5, alternative testing result meets while being less than given threshold thresholdIOU and (pHash+
1/I(G(X),H(Y))) be greater than threshold value thresholdA when, using step 2 determine coarse positioning target instrument region as final mesh
Mark, otherwise in case selecting testing result as final goal, pHash is alternative testing result perceptual hash index, I (G(X),H(Y))
For mutual information index.
Preferably, each object candidate area and coarse positioning target instrument region are asked and are handed over and than parameter IOU's by step 4-1
Formula are as follows:
In formula, C is coarse positioning target instrument region, niFor object candidate area.
Preferably, step 4-2 respectively does each object candidate area image and the instrument area image in template image
What perceptual hash calculated method particularly includes:
Object candidate area image and template image are zoomed into same size, carry out cosine transform, chooses cosine transform
The DC component of the part low frequency region in the image upper left corner afterwards, removal coordinate (0,0) obtains N-dimensional feature vector altogether, calculates
The Hamming distance of the feature vector of object candidate area image and template image, perceptually Hash index.
Preferably, the calculation formula of the mutual information index of each object candidate area image of step 4-3 and template image are as follows:
G(X)、H(Y)The respectively number of template image and candidate image gray-scale pixels, W, H are respectively candidate region image
It is wide, high.
Preferably, in step 4-4 confidence level calculation formula are as follows:
Confidence=1- (pHash+1/I (G(X),H(Y)))/(IOU+D)
In formula, I (G(X),H(Y)) it is mutual information index, pHash is perceptual hash index, and for IOU to hand over and than index, D is to set
Fixed constant.
Preferably, threshold value thresholdIOU value range 0.1~0.4, threshold value thresholdA value range 10~50.
Preferably, the classifier in step 1 is Adaboost classifier.
Compared with prior art, the present invention its remarkable advantage are as follows: 1) present invention is based on machine on the basis of classifier screening
Device people's location navigation is carried out using the frequency domain of image, carries out phase relevant calculation, subtracted using picture displacement index limits
The erroneous detection problem of missing inspection is lacked；2) present invention has merged robot localization information, using the existing location information of robot, as far as possible
So that position multiplicity is high (such as the positioning for being lower than 5cm), the variations such as scale, rotation are smaller, utilize phase phase on this basis
Put row rough detection into, substantially all tables can be examined；3) present invention utilizes machine learning, can detect that a variety of illumination, posture become
Instrument under changing.
Detailed description of the invention
Fig. 1 is overall procedure schematic diagram of the present invention.
Fig. 2 is training data schematic diagram.
Fig. 3 is the template image schematic diagram of acquisition.
Fig. 4 is the result figure of case study on implementation 1.
The result figure of Fig. 5 implementation case column 2.
Fig. 6 is flow diagram of the present invention.
Specific embodiment
As shown in figures 1 to 6, a kind of electric power meter detection method merging crusing robot location information, specific steps
Are as follows:
Step 1 chooses one in inspection point bat using Instrument image data set training classifier, and for each inspection point
As template image, crusing robot reaches specified inspection point and obtains instrument picture to be detected the instrument taken the photograph image placed in the middle；
Step 2 treats the target gauge field detected in instrument picture using plum forests Fourier transformation and phase coherent techniques
Domain carries out coarse positioning；
Step 3 is accurately positioned target instrument region using the method for machine learning, and image to be detected is sent into and is instructed
The classifier practiced obtains several object candidate areas；
Step 4, the perceptual hash, mutual information and the friendship that calculate multiple candidate regions and three kinds of parameter indexes of ratio, screen target
Candidate region obtains final goal.
Perceptual hash, mutual information and the friendship of multiple candidate regions and three kinds of ratio are calculated in further embodiment, in step 4
Parameter index, screening object candidate area obtain the specific steps of final goal are as follows:
Step 4-1, each object candidate area and coarse positioning target instrument region are asked respectively and is handed over and than parameter IOU；
Step 4-2, the instrument area image in each object candidate area image and template image perception is done respectively to breathe out
It is uncommon to calculate, obtain perceptual hash index；
Step 4-3, the mutual information index of each object candidate area image and template image is calculated separately；
Step 4-4, by the friendship of each object candidate area and than three kinds of IOU, mutual information index and perceptual hash indexs
The confidence level that weighting finds out each object candidate area is done, the maximum object candidate area of confidence level is alternately detected into knot
Fruit；
If the IOU of step 4-5, alternative testing result meets while being less than given threshold thresholdIOU and (pHash+
1/I(G(X),H(Y))) be greater than threshold value thresholdA when, using step 2 determine coarse positioning target instrument region as final mesh
Mark, otherwise in case selecting testing result as final goal, pHash is alternative testing result perceptual hash index, I (G(X),H(Y))
For mutual information index.
In further embodiment, each object candidate area and coarse positioning target instrument region are asked friendship simultaneously by step 4-1
Formula than parameter IOU are as follows:
In formula, C is coarse positioning target instrument region, niFor object candidate area.
In further embodiment, step 4-2 is respectively by the instrument in each object candidate area image and template image
Area image does perceptual hash calculating method particularly includes:
Object candidate area image and template image are zoomed into same size, carry out cosine transform, chooses cosine transform
The DC component of the part low frequency region in the image upper left corner afterwards, removal coordinate (0,0) obtains N-dimensional feature vector altogether, calculates
The Hamming distance of the feature vector of object candidate area image and template image, perceptually Hash index.
In further embodiment, the mutual information index of each object candidate area image of step 4-3 and template image
Calculation formula are as follows:
G(X)、H(Y)The respectively number of template image and candidate image gray-scale pixels, W, H are respectively candidate region image
It is wide, high.
In further embodiment, the calculation formula of confidence level in step 4-4 are as follows:
Confidence=1- (pHash+1/I (G(X),H(Y)))/(IOU+D)
In formula, I (G(X),H(Y)) it is mutual information index, pHash is perceptual hash index, and for IOU to hand over and than index, D is to set
Fixed constant.
In further embodiment, threshold value thresholdIOU value range 0.1~0.4, threshold value thresholdA value
Range 10~50.
In further embodiment, the classifier in step 1 is Adaboost classifier.
The present invention, which solves, solves deficiency present in existing instrument detection technique, such as: 1) robot location's not timing, mesh
Scale, angle change are big；2) it is big such as excessively bright, excessively dark etc. to be illuminated by the light influence for target.It is used to solve above-mentioned two problems
The technical solution that two kinds of detection methods combine: 1) robot existing location information is utilized.Since robot navigation positions skill
The accuracy of art, position error is less than 5cm, therefore the variations such as camera graphical rule, rotation for taking are smaller, on this basis
Rough detection is carried out using phase correlation；2) for the attitudes vibration of the illumination variation of image, detection target, using machine learning
Method is accurately detected.
Below with reference to embodiment, the present invention will be further described.
Embodiment 1
A kind of electric power meter detection method merging crusing robot location information, specific steps are as follows:
Step 1-1, the picture for collecting the live inspection point of crusing robot acquisition, therefrom extracts the image of ammeter, according to
Instrument, which is classified, is fabricated to the greyscale image data collection of 48*48, as shown in Fig. 2, image data set is sent into training in classifier；
Step 1-2, an inspection instrument image placed in the middle, such as Fig. 3 are respectively chosen from the picture that each inspection point takes
Shown, as detection template image；
Step 1-3, crusing robot reaches specified inspection point by location navigation, and navigation error 5cm shoots an image
A is detected for instrument；
The image A and current inspection point template image B of crusing robot shooting are carried out plum forests Fourier transformation by step 2
With phase relevant calculation, the image shift index (- 20,30) of abscissa and ordinate is found out, the instrument in image A is calculated
Estimated location (728,891,201,198)；
Step 3 is accurately positioned image to be detected using trained machine learning Adaboost classifier in advance,
Multiple object candidate areas are obtained, if Fig. 4 is the target area that classifier screens, wherein Fig. 1 classifies according to Adaboost
Device obtains three candidate frames；
Step 4, the IOU for calculating separately each candidate frame, mutual information and pHash and confidence level, numerical value in candidate frame
One behavior perceptual hash pHash, the second behavior mutual information, third behavior IOU, fourth line are confidence level Confidence.；
Step 4-1, the instrument estimated location (region C) in the candidate region that classifier obtains and step 2 asks friendship and than joining
Number IOU respectively obtain three and hand over and than parameter index (0.7,0.0,0.0), shown in calculation such as formula (1):
Step 4-2, perceptual hash pHash index is calculated.The candidate region image A and interception template image that classifier obtains
Middle instrument area image B does perceptual hash pHash calculating.Perceptual hash pHash calculating is that two pictures are zoomed to 32*32
Size carries out cosine transform, the region of the 8*8 in the image upper left corner after choosing cosine transform, the direct current point of removal coordinate (0,0)
The Hamming distance that the feature vector of image A and image B is calculated to 63 dimensional feature vectors is measured, perceptually Hash pHash refers to
Mark, the index being calculated are as shown in Figure 4.
Step 4-3, mutual information index is calculated.The mutual of candidate region image and template image is calculated using formula (2)~(5)
Information index:
Wherein X, Y are gray-scale pixel values, G(X)、H(Y)For the number of template image and candidate image gray-scale pixels..
Step 4-4, by the friendship of each candidate region and than tri- kinds of IOU, mutual information, perceptual hash pHash indexs according to public affairs
Formula (6) does the confidence level that weighting finds out the candidate region, and wherein D is a constant.
Confidence=1- (pHash+1/I (G(X),H(Y)))/(IOU+D) (6)
It is sorted from large to small according to the confidence level of all candidate regions, the maximum region of confidence level is found out, as final area
Domain.
Embodiment 2
A kind of electric power meter detection method merging crusing robot location information, specific steps are as follows:
Step 1, crusing robot reach specified inspection point by location navigation, and navigation error 5cm shoots an image A
It is detected for instrument；
The image A and current inspection point template image B of crusing robot shooting are carried out plum forests Fourier transformation by step 2
With phase relevant calculation, the picture displacement offset target (- 12,33) of abscissa and ordinate is found out, is calculated in image A
Instrument estimated location (984,502,103,86)；
Step 3 is accurately positioned image to be detected using trained machine learning Adaboost classifier in advance,
It is as shown in Figure 5 to obtain multiple object candidate areas；1 candidate frame (having number) is obtained according to Adaboost classifier, is free of
The box of numerical value is that step 2 obtains.
Step 4, IOU, mutual information and the pHash and confidence level for calculating each candidate frame, numerical value the first row in candidate frame
For pHash, the second behavior mutual information, third behavior IOU, fourth line confidence.
Step 4-1, the candidate region that classifier obtains and instrument estimated location ask friendship and than parameter IOU, such as formula (7) institute
Show, result is numerical value fourth line in Fig. 5 box:
Step 4-2, perceptual hash pHash index is calculated.The candidate region image A and interception template image that classifier obtains
Middle instrument area image B does perceptual hash pHash calculating.Perceptual hash pHash calculating is that two pictures are zoomed to 32*32
Size carries out cosine transform, the region of the 8*8 in the image upper left corner after choosing cosine transform, the direct current point of removal coordinate (0,0)
The Hamming distance that the feature vector of image A and image B is calculated to 63 dimensional feature vectors is measured, perceptually Hash pHash index
The index being calculated is as shown in Figure 5.
Step 4-3, mutual information index is calculated.The mutual of candidate region image and template image is calculated using formula (8)-(11)
Information index.
Wherein X, Y are gray-scale pixel values, and G_ ((X)), H_ ((Y)) are the number of template image and candidate image gray-scale pixels
Mesh.
Step 4-4, by the friendship of each candidate region and than tri- kinds of IOU, mutual information, perceptual hash pHash indexs according to public affairs
Formula (12) does the confidence level that weighting finds out the candidate region, and wherein D is a constant.
Confidence=1- (pHash+1/I (G(X),H(Y)))/(IOU+D) (12)
Since IOU index is less than threshold value thresholdIOU (0.3), therefore abandon the alternative detection zone of Adaboost classifier
Domain, the estimation region that 2 obtain that takes steps are exported as detection zone result.
Claims (8)
1. a kind of electric power meter detection method for merging crusing robot location information, which is characterized in that specific steps are as follows:
Step 1 chooses one in inspection point shooting using Instrument image data set training classifier, and for each inspection point
As template image, crusing robot reaches specified inspection point and obtains instrument picture to be detected instrument image placed in the middle；
Step 2, using plum forests Fourier transformation and phase coherent techniques treat detection instrument picture in target instrument region into
Row coarse positioning；
Step 3 is accurately positioned target instrument region using the method for machine learning, and image to be detected feeding was trained
Classifier, obtain several object candidate areas；
Step 4, the perceptual hash, mutual information and the friendship that calculate multiple candidate regions and three kinds of parameter indexes of ratio, screen target candidate
Region obtains final goal.
2. the electric power meter detection method of fusion crusing robot location information according to claim 1, which is characterized in that
Perceptual hash, mutual information and the friendship of multiple candidate regions and three kinds of parameter indexes of ratio are calculated in step 4, screen object candidate area
Obtain the specific steps of final goal are as follows:
Step 4-1, each object candidate area and coarse positioning target instrument region are asked respectively and is handed over and than parameter IOU；
Step 4-2, the instrument area image in each object candidate area image and template image is made into perceptual hash meter respectively
It calculates, obtains perceptual hash index；
Step 4-3, the mutual information index of each object candidate area image and template image is calculated separately；
Step 4-4, add by the friendship of each object candidate area and than three kinds of IOU, mutual information index and perceptual hash indexs
Power finds out the confidence level of each object candidate area, by the maximum object candidate area of confidence level alternately testing result；
If the IOU of step 4-5, alternative testing result meets while being less than given threshold thresholdIOU and (pHash+1/I
(G(X), H(Y))) be greater than threshold value thresholdA when, using step 2 determine coarse positioning target instrument region as final goal, it is no
Then in case selecting testing result as final goal, pHash is alternative testing result perceptual hash index, I (G(X), H(Y)) it is mutual trust
Cease index.
3. the electric power meter detection method of fusion crusing robot location information according to claim 2, which is characterized in that
Each object candidate area and coarse positioning target instrument region are sought the formula handed over and than parameter IOU by step 4-1 are as follows:
In formula, C is coarse positioning target instrument region, niFor object candidate area.
4. the electric power meter detection method of fusion crusing robot location information according to claim 2, which is characterized in that
Instrument area image in each object candidate area image and template image is done the tool of perceptual hash calculating by step 4-2 respectively
Body method are as follows:
Object candidate area image and template image are zoomed into same size, carry out cosine transform, after choosing cosine transform
The part low frequency region in the image upper left corner, the DC component of removal coordinate (0,0) obtain N-dimensional feature vector altogether, calculate target
The Hamming distance of the feature vector of candidate region image and template image, perceptually Hash index.
5. the electric power meter detection method of fusion crusing robot location information according to claim 2, which is characterized in that
The calculation formula of the mutual information index of each object candidate area image of step 4-3 and template image are as follows:
G(X)、H(Y)The respectively number of template image and candidate image gray-scale pixels, W, H are respectively that candidate region image is wide, high.
6. the electric power meter detection method of fusion crusing robot location information according to claim 2, which is characterized in that
The calculation formula of confidence level in step 4-4 are as follows:
Confidence=1- (pHash+1/I (G(X), H(Y)))/(IOU+D)
In formula, I (G(X), H(Y)) it is mutual information index, pHash is perceptual hash index, and for IOU to hand over and than index, D is setting
Constant.
7. the electric power meter detection method of fusion crusing robot location information according to claim 2, which is characterized in that
Threshold value thresholdIOU value range 0.1~0.4, threshold value thresholdA value range 10~50.
8. the electric power meter detection method of fusion crusing robot location information according to claim 1, which is characterized in that
Classifier in step 1 is Adaboost classifier.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811148567.8A CN109360289B (en) | 2018-09-29 | 2018-09-29 | Power meter detection method fusing inspection robot positioning information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811148567.8A CN109360289B (en) | 2018-09-29 | 2018-09-29 | Power meter detection method fusing inspection robot positioning information |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109360289A true CN109360289A (en) | 2019-02-19 |
| CN109360289B CN109360289B (en) | 2021-09-28 |
Family
ID=65348090
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811148567.8A Active CN109360289B (en) | 2018-09-29 | 2018-09-29 | Power meter detection method fusing inspection robot positioning information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109360289B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110298339A (en) * | 2019-06-27 | 2019-10-01 | 北京史河科技有限公司 | A kind of instrument disk discrimination method, device and computer storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104268528A (en) * | 2014-09-28 | 2015-01-07 | 深圳市科松电子有限公司 | Method and device for detecting crowd gathered region |
| CN105260412A (en) * | 2015-09-24 | 2016-01-20 | 东方网力科技股份有限公司 | Image storage method and device, and image retrieval method and device |
| CN106951930A (en) * | 2017-04-13 | 2017-07-14 | 杭州申昊科技股份有限公司 | A kind of instrument localization method suitable for Intelligent Mobile Robot |
| CN107092905A (en) * | 2017-03-24 | 2017-08-25 | 重庆邮电大学 | A kind of instrument localization method to be identified of electric inspection process robot |
| CN107610162A (en) * | 2017-08-04 | 2018-01-19 | 浙江工业大学 | A kind of three-dimensional multimode state medical image autoegistration method based on mutual information and image segmentation |
| CN107665348A (en) * | 2017-09-26 | 2018-02-06 | 山东鲁能智能技术有限公司 | A kind of digit recognition method and device of transformer station's digital instrument |
| CN107958073A (en) * | 2017-12-07 | 2018-04-24 | 电子科技大学 | A kind of Color Image Retrieval based on particle swarm optimization algorithm optimization |
| CN108446584A (en) * | 2018-01-30 | 2018-08-24 | 中国航天电子技术研究院 | A kind of unmanned plane scouting video image target automatic testing method |
-
2018
- 2018-09-29 CN CN201811148567.8A patent/CN109360289B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104268528A (en) * | 2014-09-28 | 2015-01-07 | 深圳市科松电子有限公司 | Method and device for detecting crowd gathered region |
| CN105260412A (en) * | 2015-09-24 | 2016-01-20 | 东方网力科技股份有限公司 | Image storage method and device, and image retrieval method and device |
| CN107092905A (en) * | 2017-03-24 | 2017-08-25 | 重庆邮电大学 | A kind of instrument localization method to be identified of electric inspection process robot |
| CN106951930A (en) * | 2017-04-13 | 2017-07-14 | 杭州申昊科技股份有限公司 | A kind of instrument localization method suitable for Intelligent Mobile Robot |
| CN107610162A (en) * | 2017-08-04 | 2018-01-19 | 浙江工业大学 | A kind of three-dimensional multimode state medical image autoegistration method based on mutual information and image segmentation |
| CN107665348A (en) * | 2017-09-26 | 2018-02-06 | 山东鲁能智能技术有限公司 | A kind of digit recognition method and device of transformer station's digital instrument |
| CN107958073A (en) * | 2017-12-07 | 2018-04-24 | 电子科技大学 | A kind of Color Image Retrieval based on particle swarm optimization algorithm optimization |
| CN108446584A (en) * | 2018-01-30 | 2018-08-24 | 中国航天电子技术研究院 | A kind of unmanned plane scouting video image target automatic testing method |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110298339A (en) * | 2019-06-27 | 2019-10-01 | 北京史河科技有限公司 | A kind of instrument disk discrimination method, device and computer storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109360289B (en) | 2021-09-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107229930B (en) | Intelligent identification method for numerical value of pointer instrument | |
| CN102521560B (en) | Instrument pointer image identification method of high-robustness rod | |
| CN109447061A (en) | Reactor oil level indicator recognition methods based on crusing robot | |
| CN105335973B (en) | Apply to the visual processing method of strip machining production line | |
| CN108319949A (en) | Mostly towards Ship Target Detection and recognition methods in a kind of high-resolution remote sensing image | |
| CN109100741A (en) | A kind of object detection method based on 3D laser radar and image data | |
| CN107025648A (en) | A kind of board failure infrared image automatic testing method | |
| CN109583324A (en) | A kind of pointer meters reading automatic identifying method based on the more box detectors of single-point | |
| CN109389165A (en) | Oil level gauge for transformer recognition methods based on crusing robot | |
| CN108520514B (en) | Consistency detection method for electronic elements of printed circuit board based on computer vision | |
| CN105913415A (en) | Image sub-pixel edge extraction method having extensive adaptability | |
| CN106056597B (en) | Object visual detection method and device | |
| CN104977313A (en) | Method and device for detecting and identifying X-ray image defects of welding seam | |
| CN103927758B (en) | Saliency detection method based on contrast ratio and minimum convex hull of angular point | |
| CN109447062A (en) | Pointer-type gauges recognition methods based on crusing robot | |
| CN107862690A (en) | The circuit board element localization method and positioner of a kind of feature based Point matching | |
| CN107016353B (en) | A kind of integrated method and system of variable resolution target detection and identification | |
| CN106157308A (en) | Rectangular target object detecting method | |
| CN107578021A (en) | Pedestrian detection method, apparatus and system based on deep learning network | |
| JP2011002919A (en) | Method and device for detecting object | |
| CN108960115A (en) | Multi-direction Method for text detection based on angle point | |
| CN110096980A (en) | Character machining identifying system | |
| CN109507198A (en) | Mask detection system and method based on Fast Fourier Transform (FFT) and linear Gauss | |
| CN103854278A (en) | Printed circuit board image registration method based on shape context of mass center of communicated region | |
| CN114092728A (en) | Pointer instrument intelligent identification method and system based on deep learning |
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 |