CN102609712B - Reading method of round-like pointer instrument used for mobile robot - Google Patents
Reading method of round-like pointer instrument used for mobile robot Download PDFInfo
- Publication number
- CN102609712B CN102609712B CN201210043415.8A CN201210043415A CN102609712B CN 102609712 B CN102609712 B CN 102609712B CN 201210043415 A CN201210043415 A CN 201210043415A CN 102609712 B CN102609712 B CN 102609712B
- Authority
- CN
- China
- Prior art keywords
- ellipse
- image
- pointer
- mobile robot
- oval
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a reading method of a round-like pointer instrument used for a mobile robot. A target instrument template is extracted manually, and the reading method comprises the following steps: A. carrying out graying and binaryzation on a target image collected by the mobile robot; B. firstly extracting a dial area of the image obtained in the step A, extracting a boundary image of an equipment by utilizing a Canny algorithm, taking a boundary point to fit an oval, filtering the obtained oval, and rejecting a non-oval image which is fitted into the oval so as to obtain the dial image; C. extracting a pointer in an area obtaining the dial image, so as to obtain the angle of the pointer; and D. extracting the relation of a pointer angle and a reading corresponding to the target instrument in a database, and calculating the reading corresponding to the angle obtained in the step C. The method provided by the invention has robustness in the positioning precision of the robot, can be used for accurately reading the instrument, is small in calculation amount, and is simple in constructing a prophase database.
Description
Technical field
The present invention relates to a kind of Computer Image Processing method, specifically, relate to a kind of method of the pointer instrument automatic reading based on mobile robot's image acquisition; Similar round is by the circular and oval set forming.
Background technology
In transformer station, instrument is in occupation of important effect, particularly important to the reading of all kinds of instrument, how to realize the reading of all kinds of instrument is concerned to the accuracy of data acquisition, the key factor that collection difficulty is little and cost is low.In present stage, the collection of all kinds of Instrument images and identification, mostly rely on fixing camera to gather, in this acquisition mode, the position of dial plate region in gathered image fixed, easily identification, but a large amount of equipment for fixing camera need to be installed, and this is not only subject to the impact that capital consumption is larger, and be subject to substation equipment safe distance restriction.
And Intelligent Mobile Robot, can effectively address the above problem, but compared with fixing camera, due to the impact of robot motion's stop precision and cloud platform rotation error, make the position of instrument in the image collecting unfixing, need further be improved, by the extraction of dial plate effectively to address this problem, thereby extended the application of image recognition technology in commercial production.
Image recognition technology has a very wide range of applications in commercial production especially operation power, production, and the meter reading based on image processing is its application direction important in electric system.The communications equipment room of electric system and transformer station are provided with a lot of equipment instrument, as: conservator table, oil thermometer, SF6 switch tensimeter, voltage table, reometer etc.At present, mostly utilize long-range fixed video supervisory system to realize to the monitoring of these instrument, instrument numerical value is fetched realization by gauge pointer is monitored in real time and intercepted image scene by manual read, can distinguish accurately although do like this, but install and substation safety distance problem foregoing position, especially need additional manpower support, this scheme is no longer advocated by current.
Utilize mobile robot and image processing techniques, instrument is independently taken, and definite pointer position, automatically calculate total indicator reading, the state of real-time recording unit, this has important realistic meaning to guaranteeing the production safety of electric power enterprise, is also the trend of unattended operation transformer station in future
Oval as an important shape facility, in actual life, exist in a large number, in a lot of fields, requirement can detect the elliptic contour in image rapidly.In actual life, the perspective projection of voluminous object is ellipse, and ellipse fitting is the condition precedent of follow-up object identification and measurement, and oval robust is accurately extracted in the meaning that has particular importance in image recognition and computer vision.
China CN1581209A application for a patent for invention has openly proposed a kind of circle/ellipse/square instrument monitoring method in transformer station based on contour of object, its basic scheme is that overhead control end computer based is in the circle/ellipse/rectangular search having based on contour of object, the long-range table detection/read module of detection algorithm, by the crusing robot in WLAN (wireless local area network) control transformer station, according to the predefined route of patrolling and examining, within the scope of the search volume of setting, lock the locus of instrument by search subroutine, then obtain instrument several picture with visible image capturing head, beam back overhead control end by WLAN (wireless local area network) and carry out data processing to reach the object of remote monitoring.The method still needs accurately to control the accurate location of crusing robot, namely depends on the location of crusing robot under perfect condition, to guarantee the accuracy of image acquisition, the searching of dial plate is not had to robustness, and too high to systematically controlling cost.
As further improvement, Chinese CN101620682A application has openly proposed a kind of readings of pointer type meters method, and the method is set up the scale question blank of different instrument, forms database; Acquisition instrument image, input Instrument image also carries out pre-service to it, to extract effective identified region of instrument dial plate; From effective identified region of instrument dial plate, utilize pointer and scale mark stroke feature, extract pointer and the scale mark of instrument; According to the pointer and the scale mark that extract, the scale mark of extraction is mated with scale question blank corresponding in scale lookup table database, carry out reading and automatically identify.This scheme has larger improvement with respect to aforesaid CN1581209A, but this method can only be adapted to fixing control point, and the searching of dial plate is not had to robustness, under complicated outdoor environment, cannot realize the accurate location to instrument.Especially will set up loaded down with trivial details database, previous work amount is very large.
Summary of the invention
Therefore, the object of the present invention is to provide a kind of similar round readings of pointer type meters method for mobile robot, the positioning precision of robot is had to robustness, can realize the accurate reading to instrument, and calculated amount is little, Primary Stage Data storehouse is set up simple.
In order to realize goal of the invention of the present invention, the technical scheme adopting is:
A kind of similar round readings of pointer type meters method for mobile robot of this invention, manual extraction target instrument template, the corresponding relation of target-marking meter dial and angle, deposits database in, and then this number reading method comprises the following steps:
A. target image gray processing mobile robot being gathered, binaryzation;
B. the dial plate region of first extraction step A institute synthetic image: utilize Canny algorithm extraction equipment boundary image; Get frontier point, and then fitted ellipse; The ellipse obtaining is filtered, weed out and be synthesized to oval non-oval image, obtain dial plate image;
C. in acquisition dial plate image-region, extract pointer, and then obtain the angle of pointer;
D. between the pointer angle of extracting corresponding target instrument in database and registration, be related to the registration that calculation procedure C gained angle is corresponding.
The invention has the beneficial effects as follows that just gathering individual equipment drawing by robot looks like to add template base in previous work, corresponding Instrument image is by manual extraction, and carry out manual identification and mark according to the Instrument image obtaining, and can be by set relevant readings parameter take it as benchmark, previous work amount is very little.Stop precision under robot manipulating task condition is little on gauge pointer reading impact on template ellipse, only robot is very high not in place far away, to such an extent as to ellipticity is too little, the stop precision of robot can meet the application of this method in theory, it is not fairly obvious that instrument is obtained morphological differences in image in robot, template that previous work obtains can meet automatically reading of registration, and the precision that robot is stopped has robustness.
Than existing mobile robot's meter reading method, the reading accuracy of this programme, in the situation that calculated amount reduces, has kept higher reading accuracy.
The above-mentioned similar round readings of pointer type meters method for mobile robot, while getting frontier point, if the frontier point number that can be extracted is greater than threshold value
t c , utilize least square fitting ellipse; Otherwise, reselect border, can leach like this some and should, by the figure of matching, not reduce calculated amount.
The above-mentioned similar round readings of pointer type meters method for mobile robot, the method that the ellipse obtaining is filtered is first to calculate the frontier point of extraction and the distance of institute's fitted ellipse, and set a threshold value
t d the frontier point that is less than this threshold value is match point, all match points are defined as qualified rates of fitting with the ratio of the frontier point that extracts, and calculate the ellipticity of gained ellipse, thereby in the time being filtered oval ellipticity and being greater than the ellipticity threshold value of setting and qualified rates of fitting and being greater than the qualified rates of fitting threshold value of setting, record gained ellipse, accordingly can the most noise of filtering.
The above-mentioned similar round readings of pointer type meters method for mobile robot, in order to obtain more accurate dial plate image, the step B carrying out repeatedly obtains multiple dial plate images, then multiple dial plate images of gained are carried out to the optimum oval step of selecting, using the optimum ellipse obtaining as final dial plate image.
The above-mentioned similar round readings of pointer type meters method for mobile robot, the optimum oval step of selecting is, first calculate the area of the corresponding fitted ellipse of all gained dial plate images, take the ellipse area of manual extraction corresponding to corresponding target instrument template as desirable ellipse area, travel through the absolute value of the difference of all gained dial plate images and desirable ellipse area, be defined as difference in areas, get the fitted ellipse of difference in areas minimum as optimum oval.
The above-mentioned similar round readings of pointer type meters method for mobile robot, described ellipticity threshold value is 0.85, qualified rates of fitting threshold value is 0.8.
The above-mentioned similar round readings of pointer type meters method for mobile robot, also comprise parameter for a certain target instrument template and the number reading method testing procedure before mobile robot's operation, by sneak into other figures in test ellipse, can checking divide to detect tests ellipse, and this can be for debugging some threshold value.
Accompanying drawing explanation
Fig. 1 is the ellipse in two dimensional surface.
Fig. 2 is according to a kind of similar round readings of pointer type meters method flow diagram for mobile robot of the present invention.
Embodiment
A kind of method for the pointer instrument automatic reading on mobile robot, first should set up template base, according to mobile robot's the object of patrolling and examining, set and patrol and examine route, and individual equipment drawing of object pointer formula instrument to collect of patrolling and examining on object is looked like to add template base.Manual extraction target instrument template, the corresponding relation of target-marking meter dial and angle, deposits database in, further, can represent evenly and the instrument of non-homogeneous scale type, to mate different pointer angles and the relation of reading.
Just generating after template to be that instrument is carried out reading and operated for corresponding pointer, and reading operation steps is as follows:
Step1: input image to be identified, gather associated picture by mobile robot and send into corresponding recognition means, obtain the dial plate region of image to be identified by ellipse fitting algorithm;
Step2: then his-and-hers watches disk area image is realized pointer detection through Image Pretreatment Algorithm;
Step3: pointer angle is inquired about, obtain the registration of instrument.
In described step1, the step of ellipse fitting algorithm is as follows:
A1. oval method for expressing:
Referring to Figure of description 1, in two dimensional surface coordinate system, ellipse generally can represent by two kinds of forms, and a kind of general algebra form of elliptic curve equation represents, suc as formula (1)
Another more intuitive mode is to represent with the geometric parameter of plane coordinate system, i.e. elliptical center position
, major semi-axis and minor semi-axis
, the corner of major axis
, α, the arbitrary ellipse in β two dimensional surface can be unique definite by these 5 parameters, and the geometric meaning of parameter is as shown in Figure 1.The parameter of two kinds of representations can be used formula (2)-Shi (5) conversion.
A2. least square fitting ellipse
The elliptic equation of supposing general type is listed suc as formula (1), is the implicit equation of two multiplication of vectors by ellipse representation
Wherein
for (A, B, C, D, E, F), i.e. coefficient vector,
, because error,
point
non-vanishing, will
think a little
to implicit equation
algebraic distance, according to the principle of least square, curve fit problem can be by by algebraic distance quadratic sum
Minimize to realize.By extremum principle, make f (A, B, C, D, E) value for minimum, must have
Can obtain thus a system of linear equations, then application solves the algorithm (as complete pivot Gaussian elimination) of system of linear equations, in conjunction with constraint condition, just can be in the hope of equation coefficient A, and B, C, D, E, the value of F, finally obtains elliptic equation.
A3. the ellipse fitting based on border and screening:
While carrying out ellipse fitting by least square method, only can obtain having the ellipse of least error, not consider the edge fitting degree that gained is oval and original, cause the ellipse that obtains and actually have a relatively large deviation, arithmetic result is often also unsatisfactory.In the case, evaluating screening by the ellipse that matching is obtained is necessary.This programme has proposed a kind of least square ellipse fitting algorithm based on border.
The border adopting when matching with obtain oval matching degree, we are called qualified rates of fitting, are the important indicators of weighing gained fitted ellipse.First traversal needs borderline all sample points of matching, asks for each point to the distance between the ellipse that matching obtains, if be less than certain threshold value
t d , claim that this sample point is match point (match_points), and count with num_matchPt, travel through completely, ask for the total number (num_edgePt) for the match point of this fitted ellipse, qualified rates of fitting is
Describe an ellipse and it is highly important that parameter is its major semi-axis (a) and minor semi-axis (b), screen matching gained ellipse by definition ellipticity (ellipseRatio) and elliptical area product moment.
Ellipticity definition:
The condition of ellipticity has been described matching gained ellipse and has been trended towards round degree, by determining oval size to the restriction of major and minor axis.
The definition of difference in areas:
Wherein,
for desirable ellipse area,
for matching gained ellipse area.The final ellipse of selecting difference in areas minimum, if do not specify oval size, selects area larger oval as optimum oval.
Threshold value
t d can select according to test, distinguish under the situation of non-required fitted ellipse and required fitted ellipse, those skilled in the art easily obtains accordingly.
Desirable ellipse area, shows in template, the pointer instrument in template is carried out to manual extraction, and this accuracy is indubitable, and the ellipse area obtaining take this is reference ellipsoid, and desirable oval, corresponding area is desirable ellipse area.
More specifically, concrete operations are as follows:
The target image that mobile robot is gathered carries out gray processing, and binaryzation utilizes Canny algorithm to extract boundary image;
Then get border for the boundary image obtaining, the sample that obtains, if the number of frontier point is greater than threshold value
t c , utilize least square fitting ellipse; Otherwise, reselect border, until obtain required fitted ellipse;
Then just can follow according to the fitted ellipse digital simulation rate and the ellipticity that obtain.Calculate qualified rates of fitting and ellipticity according to formula (10) (11).
Be greater than ellipticity threshold value if meet ellipticity, qualified rates of fitting is greater than qualified rates of fitting threshold value, records the oval E of gained;
Calculate again oval E and desirable elliptical area product moment.Poor according to formula (12) reference area;
According to the fitted ellipse circulation execution step 2 of original acquisition) ~ step 5), the ellipse of selecting difference in areas minimum is optimum oval.
Above-mentioned concrete operations see Figure of description 2.
Referring to Figure of description 2, it is a kind of scheme shown in Fig. 2 that said method characterizes more accurately, and the judgement of Fig. 2 conditional shows, may just can obtain needed optimum ellipse by an ellipse fitting, also may have many times and just can hit.Certainly; the possibility of hit at first time is smaller; so; therefore; have border points extraction repeatedly, also have the checking of qualified rates of fitting repeatedly, in theory again in the flow process shown in Fig. 2; step after the set of border shows as the execution number of times of step above from front to back, conventionally can be more than the execution number of times of later step.
In described step Step2, gauge pointer is depended in the pre-service of image, gauge pointer is different from as wall clock pointer, do not emphasize ornamental, there is linear distinguishing feature, therefore, extraction those skilled in the art by means of at present popular rectilinear figure element easily obtains pointer in the situation that obtaining dial plate, and main step is as follows:
B. straight line Morphology Algorithm is carried out in the dial plate region of the target image through gray scale chemical combination binaryzation, realize the angle extraction of pointer after refinement, the extraction comparison of straight line key element is simple, does not repeat them here.
Ellipse fitting experiment is take VS2005 as platform, at Pentium (R) 2.66GHz, on the PC of 1G internal memory, realizes.By more synthetic images, be mainly oval and other images of being sneaked into, then test by ellipse fitting, can simulate the reasonable ellipse of effect, do not disturbed by other image, can obtain all ellipse figures.
Experiment showed, in addition if do not filtered, robot gathers the image that may contain other in conservator chart picture in transformer station and have garden or similar round border, such as conservator edge, also can simulate corresponding ellipse.By to ellipticity, qualified rates of fitting, the conditions such as oval size are carried out controlled filter, and the fitted ellipse region after screen is just in time the image of the conservator table section extracted.
In composograph, ellipticity is selected less, can detect more flat ellipse, and qualified rates of fitting is larger, all elliptical shapes in can detected image.Instrument image is taken to obtain to by robot, and ellipticity threshold value is general selects 0.85, and qualified rates of fitting threshold value selects 0.8, has reasonable recognition capability.Method elapsed time, between 10-25ms, meets the requirement of real-time.
By reference to the accompanying drawings the specific embodiment of the present invention is described although above-mentioned; but not limiting the scope of the invention; one of ordinary skill in the art should be understood that; on the basis of technical scheme of the present invention, those skilled in the art do not need to pay various modifications that creative work can make or distortion still in protection scope of the present invention.
Claims (5)
1. for mobile robot's a similar round readings of pointer type meters method, it is characterized in that, manual extraction target instrument template, the corresponding relation of target-marking meter dial and angle, deposits database in, and then this number reading method comprises the following steps:
A. target image gray processing mobile robot being gathered, binaryzation;
B. the dial plate region of first extraction step A institute synthetic image: utilize Canny algorithm extraction equipment boundary image; Get frontier point, and then fitted ellipse; The ellipse obtaining is filtered, weed out and be synthesized to oval non-oval image, obtain dial plate image;
C. in acquisition dial plate image-region, extract pointer, and then obtain the angle of pointer;
D. between the pointer angle of extracting corresponding target instrument in database and registration, be related to the registration that calculation procedure C gained angle is corresponding;
While getting frontier point, if the frontier point number that can be extracted is greater than threshold value TC, utilize least square fitting ellipse; Otherwise, reselect border;
The method that the ellipse obtaining is filtered is first to calculate the frontier point of extraction and the distance of institute's fitted ellipse, and set a threshold value
t d the frontier point that is less than this threshold value is match point, all match points are defined as qualified rates of fitting with the ratio of the frontier point that extracts, and calculate the ellipticity of gained ellipse, thereby in the time being filtered oval ellipticity and being greater than the ellipticity threshold value of setting and qualified rates of fitting and being greater than the qualified rates of fitting threshold value of setting, record gained ellipse.
2. the similar round readings of pointer type meters method for mobile robot according to claim 1, it is characterized in that, the step B carrying out repeatedly obtains multiple dial plate images, then multiple dial plate images of gained are carried out to the optimum oval step of selecting, using the optimum ellipse obtaining as final dial plate image.
3. the similar round readings of pointer type meters method for mobile robot according to claim 2, it is characterized in that, the optimum oval step of selecting is, first calculate the area of the corresponding fitted ellipse of all gained dial plate images, take the ellipse area of manual extraction corresponding to corresponding target instrument template as desirable ellipse area, the absolute value that travels through the difference of all gained dial plate images and desirable ellipse area, is defined as difference in areas, gets the fitted ellipse of difference in areas minimum as optimum oval.
4. according to the arbitrary described similar round readings of pointer type meters method for mobile robot of claims 1 to 3, it is characterized in that, described ellipticity threshold value is 0.85, and qualified rates of fitting threshold value is 0.8.
5. the similar round readings of pointer type meters method for mobile robot according to claim 1, it is characterized in that, also comprise parameter for a target instrument template and the number reading method testing procedure before mobile robot's operation, by sneak into other figures in test ellipse, can checking divide to detect tests ellipse.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210043415.8A CN102609712B (en) | 2012-02-24 | 2012-02-24 | Reading method of round-like pointer instrument used for mobile robot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210043415.8A CN102609712B (en) | 2012-02-24 | 2012-02-24 | Reading method of round-like pointer instrument used for mobile robot |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102609712A CN102609712A (en) | 2012-07-25 |
| CN102609712B true CN102609712B (en) | 2014-07-09 |
Family
ID=46527071
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210043415.8A Expired - Fee Related CN102609712B (en) | 2012-02-24 | 2012-02-24 | Reading method of round-like pointer instrument used for mobile robot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102609712B (en) |
Families Citing this family (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103810488B (en) * | 2012-11-09 | 2017-08-08 | 阿里巴巴集团控股有限公司 | Image characteristic extracting method, image search method, server, terminal and system |
| CN103148881B (en) * | 2013-02-06 | 2015-05-13 | 广东电网公司电力科学研究院 | Method and device for automatically reading pointer instrument by video frequency |
| CN106228161B (en) * | 2016-07-18 | 2019-04-19 | 电子科技大学 | A kind of pointer-type dial plate automatic reading method |
| CN106446903B (en) * | 2016-09-20 | 2019-04-23 | 山东省计量科学研究院 | A kind of instrument band switch location recognition method |
| CN106650697B (en) * | 2016-12-30 | 2019-11-15 | 亿嘉和科技股份有限公司 | A kind of meter dial recognition methods |
| CN107388988A (en) * | 2017-07-27 | 2017-11-24 | 成都畅达通检测技术股份有限公司 | Tunnel oval-shaped degree determination method and system |
| CN109034170B (en) * | 2018-07-18 | 2022-02-11 | 东南大学 | Reading method for circular pointer type instrument of switch cabinet detection device |
| CN109359604B (en) * | 2018-10-23 | 2022-03-11 | 大连航佳机器人科技有限公司 | Method for identifying instrument under shadow interference facing inspection robot |
| CN111222364B (en) * | 2018-11-23 | 2023-08-18 | 中移物联网有限公司 | Meter reading method and meter reading device |
| CN109910610B (en) * | 2019-02-18 | 2020-09-29 | 武汉格瑞恩电子仪表有限公司 | Combined instrument |
| CN109993154B (en) * | 2019-04-19 | 2020-02-07 | 山东大学 | Intelligent identification method for single-pointer sulfur hexafluoride instrument of transformer substation |
| CN110097062B (en) * | 2019-04-26 | 2021-06-04 | 亿嘉和科技股份有限公司 | Pointer identification method using pointer center point |
| CN110084251A (en) * | 2019-04-29 | 2019-08-02 | 北京史河科技有限公司 | A kind of Meter recognition method, apparatus and storage medium |
| CN111598097B (en) * | 2020-05-08 | 2023-03-24 | 福建省海峡智汇科技有限公司 | Instrument position and reading identification method and system based on robot vision |
| CN114332077A (en) * | 2022-03-02 | 2022-04-12 | 国网江西省电力有限公司电力科学研究院 | RANSAC-based light dial plate identification method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1581209A (en) * | 2004-05-21 | 2005-02-16 | 清华大学 | Converting station circular/ellipsoid/square instrument monitoring method based object profile |
| CN101620682A (en) * | 2008-06-30 | 2010-01-06 | 汉王科技股份有限公司 | Method and system for automatically identifying readings of pointer type meters |
-
2012
- 2012-02-24 CN CN201210043415.8A patent/CN102609712B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1581209A (en) * | 2004-05-21 | 2005-02-16 | 清华大学 | Converting station circular/ellipsoid/square instrument monitoring method based object profile |
| CN101620682A (en) * | 2008-06-30 | 2010-01-06 | 汉王科技股份有限公司 | Method and system for automatically identifying readings of pointer type meters |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102609712A (en) | 2012-07-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102609712B (en) | Reading method of round-like pointer instrument used for mobile robot | |
| Deng et al. | Automatic indoor construction process monitoring for tiles based on BIM and computer vision | |
| CN107229930B (en) | Intelligent identification method for numerical value of pointer instrument | |
| CN106338706B (en) | A kind of methods, devices and systems of electric energy metering device global error detection | |
| CN103927507A (en) | Improved multi-instrument reading identification method of transformer station inspection robot | |
| CN104977584A (en) | Convective weather approach prediction method and system | |
| US9361702B2 (en) | Image detection method and device | |
| CN108228798A (en) | The method and apparatus for determining the matching relationship between point cloud data | |
| CN109308447A (en) | The method of equipment operating parameter and operating status is automatically extracted in remote monitoriong of electric power | |
| CN109190473A (en) | The application of a kind of " machine vision understanding " in remote monitoriong of electric power | |
| EP2913148B1 (en) | Method and system for analyzing process monitoring data | |
| CN105930837A (en) | Transformer station instrument equipment image recognition method based on autonomous routing inspection robot | |
| CN111598942A (en) | Method and system for automatically positioning electric power facility instrument | |
| US20210216906A1 (en) | Integrating simulated and real-world data to improve machine learning models | |
| CN102663418B (en) | An image set modeling and matching method based on regression model | |
| CN103954334A (en) | Fully automatic image pickup type water meter verification system and operating method thereof | |
| CN112183906A (en) | Machine room environment prediction method and system based on multi-model combined model | |
| KR20240005367A (en) | Location tracking system in factory and method | |
| CN117292076A (en) | Dynamic three-dimensional reconstruction method and system for local operation scene of engineering machinery | |
| Troncoso-Pastoriza et al. | Use of BIM data as input and output for improved detection of lighting elements in buildings | |
| CN116055521A (en) | Inspection system and image recognition method for electric inspection robot | |
| CN110851978A (en) | Camera position optimization method based on visibility | |
| CN102721878A (en) | Intelligentized automatic checkout device of quartz crystal oscillator instrument | |
| CN111578938B (en) | Target object positioning method and device | |
| CN104134213A (en) | Target positioning method and device in digital image |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20140709 Termination date: 20150224 |
|
| EXPY | Termination of patent right or utility model |