CN102521560A - Instrument pointer image identification method of high-robustness rod - Google Patents
Instrument pointer image identification method of high-robustness rod Download PDFInfo
- Publication number
- CN102521560A CN102521560A CN2011103591300A CN201110359130A CN102521560A CN 102521560 A CN102521560 A CN 102521560A CN 2011103591300 A CN2011103591300 A CN 2011103591300A CN 201110359130 A CN201110359130 A CN 201110359130A CN 102521560 A CN102521560 A CN 102521560A
- Authority
- CN
- China
- Prior art keywords
- pointer
- image
- instrument
- dial plate
- template
- 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
Images
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
- Arrangements For Transmission Of Measured Signals (AREA)
Abstract
The invention discloses an instrument pointer image identification method of a high-robustness rod, mainly comprising two phases including offline calibration and online identification. The offline calibration phase comprises the following steps of: collecting images, pre-treating the images, marking and learning, calibrating an instrument position, learning instrument characteristics, selecting a pointer identification method and parameters, establishing a database and storing a calibrated result. The online identification phase comprises the following steps of: collecting the images, pre-treating the images, positioning an instrument, adjusting the gray level of the instrument images, extracting the pointer and calculating readings of the pointer. The instrument pointer image identification method of the high-robustness rod, provided by the invention, is high in robustness, and has stronger robustness to the translation, the rotation and the shielding of the instrument, instrument deformation caused by light changes and different viewing angles, the change of an imaging size caused by different distances, and the like; and the pointer readings can be more accurately identified and the stable work under an actual environment is realized.
Description
Technical field
The present invention relates to image processing field, particularly a kind of reading automatic identifying method of pointer instrument particularly, relates to a kind of high robust gauge pointer image-recognizing method.
Background technology
Pointer instrument is extensive application in industry, in electric system, commercial production, field of petrochemical industry, need monitor in real time or regularly patrol and examine instrument, to guarantee that parameter is normal, system's even running.Traditional monitoring, routine inspection mode, relying on artificial is a uninteresting loaded down with trivial details job to instrument monitoring, needs the manpower of labor, and monitoring result receives subjective influence easily, has a lot of hidden danger and inconvenience.
Along with the development of Flame Image Process, monitor, patrol and examine and become possibility for the people by computer generation.Its core is utilized image processing techniques exactly, obtains the reading of pointer instrument.Number of patent application is 200520093025.7; In the Chinese patent of patent " simulated pressure table digital reading device " by name; A kind of device that the simulated pressure table is carried out reading with machine vision replacement human eye is disclosed; But this method requires accurately to obtain the home position of pointer rotation in advance, thereby has largely limited its usable range.Number of patent application is 200910042108.6, in the Chinese patent of patent by name " a kind of reading pointer type represent be worth image processing method ", discloses a kind of reading pointer type and has represented the image processing method that is worth.This method is chosen three corresponding point on three scale marks of equal in length, and calculates the circle through 3, draws the center of circle.But this method is high to environmental requirement, under actual conditions, owing to light influence, video camera mistake Jiao etc. will be difficult to extract scale mark, therefore is difficult to locate instrument.Accurate location to the instrument dial plate is the important prerequisite of pointer instrument identification, but the said method of existing related patent U.S. Patent No. all is difficult to steady operation under physical condition.Except the accurate orientation problem of instrument, in the existing patent, all do not have to consider that therefore, robustness is relatively poor, has limited its use under actual environment for the occlusion issue of light problem, instrument yet.
Summary of the invention
The objective of the invention is to deficiency, a kind of gauge pointer image-recognizing method of high robust is provided to prior art.
According to an aspect of the present invention, a kind of high robust gauge pointer image-recognizing method is provided, it is characterized in that, comprise the step and the step in ONLINE RECOGNITION stage in the off-line calibration stage of carrying out successively, wherein:
The step in said off-line calibration stage comprises: adjust camera, make it aim at the instrument dial plate, carry out IMAQ (101); Institute's images acquired is carried out image pre-service (102); On image, get a zone and serve as a mark, carry out mark study (103); In instrument location position (104), meter performance study (105) is confirmed pointer recognition methods and pointer identification parameter (106), and is set up database and calibration result is preserved (107) afterwards;
The step in said ONLINE RECOGNITION stage comprises: carry out IMAQ (101) earlier; After institute's images acquired is carried out image pre-service (108); The image that is only comprised the dial plate zone through instrument location (109); Then carry out gradation of image adjustment (110), pointer extracting (111) is accomplished total indicator reading at last and is calculated (112).
Preferably; Said image pre-service (102), (108) comprise step: image is carried out color space conversion, decomposition; Obtain the image under tone space, saturation degree space, the gray space; Be designated as tone images, saturation degree image and gray level image respectively, and carry out figure image intensifying, medium filtering processing.
Preferably, said mark study (103) comprises step: select a color space, abundant zone of texture information in this space hypograph is served as a mark, to its foundation and study template, and calculate the average gray value in this template zone; Said instrument location position (104) comprises step: the homogeneous transformation matrix sequence of recorder dial plate profile, the calculation template region point sequence on to the homogeneous transformation matrix in dial plate zone, template region to homogeneous transformation matrix, template region and the instrument dial plate profile of pointer rotation center region.
Preferably, said meter performance study (105) comprises step: confirm the pointer characteristic of this instrument and the relation of instrument range and needle deflection angle, wherein, said pointer characteristic comprises pointer number and pointer thickness type.
Preferably, said definite pointer recognition methods and pointer identification parameter (106) comprise following arbitrary step:
-in image, draw out rectangle frame, make it comprise the position of current pointer; If said pointer thickness type is thick pointer, then central projection method is confirmed as the pointer recognition methods, in said rectangle frame, respectively on gray level image, tone images, saturation degree image, in the dial plate zone, to the projection of pointer rotation center; Perhaps
-in image, draw out rectangle frame, make it comprise the position of current pointer; If said pointer thickness type is the microdactylia pin, then the pointer recognition methods is confirmed as in the Hough conversion, in said rectangle frame, on gray level image, tone images, saturation degree image, in the dial plate zone, utilize Hough change detection straight line successively.
Preferably, the said database of setting up is preserved calibration result (107) and is comprised step: preserve following any or appoint a plurality of calibration results: the relation of homogeneous transformation matrix sequence, gauge pointer number, pointer thickness type, instrument range and the needle deflection angle of the average gray value in the template that generates by mark, the color space under the template, template zone, the template region point sequence on to the homogeneous transformation matrix in dial plate zone, template region to homogeneous transformation matrix, template region and the instrument dial plate profile of pointer rotation center region, the recognition methods of pointer, the identification parameter of pointer.
Preferably; Said instrument location (109) comprises step: carry out template matches in the image after image pre-service (108); Calculate the homogeneous transformation matrix of template and current marking image; And the homogeneous transformation matrix sequence of the point sequence of the template region of preserving in the binding data storehouse on to the homogeneous transformation matrix in dial plate zone and template region to homogeneous transformation matrix, template region and the instrument dial plate profile of pointer rotation center region can calculate the regional profile information of positional information, the dial plate in instrument dial plate zone, the positional information of pointer rotation center region.
Preferably, said gradation of image adjustment (110) comprises step: calculate the average gray value of marked region, then to the adjustment of entire image gray scale, make the average gray value in its template that equals to preserve in database zone.
Preferably; Said pointer extracting (111) comprises step: according to dial plate zone that has calculated and pointer rotation center region; Gauge pointer recognition methods and identification parameter in the binding data storehouse; In the dial plate zone, carry out central projection method or Hough conversion, calculate the deflection angle of pointer.
Preferably, said total indicator reading calculates (112) and comprises step: according to the deflection angle of the pointer that has calculated, the relation of instrument range in the binding data storehouse and needle deflection angle calculates the reading of pointer.
Compared with prior art, the present invention has following advantage:
(1), utilize image processing techniques, realized that the robotization of meter reading is handled, improved work efficiency, reduced artificial subjective error factor.
(2), utilization is based on the location of mark; Improved the accuracy of instrument dial plate location, and based on the location of mark to the visual angle of the variable in distance of light variation, camera and instrument, camera squint, the rotation of meter diagram picture, meter diagram picture block etc. all comparatively robust.
(3), illumination condition is changed insensitive handmarking, except selecting certain zone serves as a mark on the dial plate, can also near instrument, pasting thus improved the robustness of system.
(4), use profile to describe instrument dial plate zone, so the present invention is all applicable for the dial plate zone of arbitrary shape, and still has robustness preferably for the dial plate warpage that tilts to cause because of shooting angle.
Description of drawings
Fig. 1 is that hardware of the present invention is formed block scheme;
Fig. 2 is total program flow diagram of the present invention;
Fig. 3 is an off-line calibration stage program flow diagram of the present invention;
Fig. 4 is an ONLINE RECOGNITION stage program flow diagram of the present invention;
Fig. 5 is central projection method's process flow diagram (is example with maximal value central projection method).
Embodiment
As shown in Figure 2, according to high robust gauge pointer image-recognizing method provided by the invention, comprise the step and the step in ONLINE RECOGNITION stage in the off-line calibration stage of carrying out successively.The step in said off-line calibration stage comprises: adjust camera, make it aim at the instrument dial plate, carry out IMAQ (step 101); Institute's images acquired is carried out image pre-service (step 102); On image, get a zone and serve as a mark, carry out mark study (step 103); At instrument location position (step 104), meter performance study (step 105) is confirmed pointer recognition methods and pointer identification parameter (step 106), and is set up database calibration result is preserved (step 107) afterwards.The step in said ONLINE RECOGNITION stage comprises: carry out IMAQ (step 101) earlier; After institute's images acquired is carried out image pre-service (step 108); The image that is only comprised the dial plate zone through instrument location (step 109); Then carry out gradation of image adjustment (step 110), pointer extracting (step 111) is accomplished total indicator reading at last and is calculated (step 112).
Particularly; Said image pre-service (step 102, step 108) comprises step: image is carried out color space conversion, decomposition; Obtain the image under tone space, saturation degree space, the gray space; Be designated as tone images, saturation degree image and gray level image respectively, and carry out figure image intensifying, medium filtering processing.Said mark study (step 103) comprises step: select a color space, abundant zone of texture information in this space hypograph is served as a mark, to its foundation and study template, and calculate the average gray value in this template zone; Said instrument location position (step 104) comprises step: the homogeneous transformation matrix sequence of recorder dial plate profile, the calculation template region point sequence on to the homogeneous transformation matrix in dial plate zone, template region to homogeneous transformation matrix, template region and the instrument dial plate profile of pointer rotation center region.Said meter performance study (step 105) comprises step: confirm the pointer characteristic of this instrument and the relation of instrument range and needle deflection angle, wherein, said pointer characteristic comprises pointer number and pointer thickness type.
More particularly, be thick pointer or microdactylia pin according to said pointer thickness type, said definite pointer recognition methods and pointer identification parameter (step 106) comprise following arbitrary step:
-in image, draw out rectangle frame, make it comprise the position of current pointer; If said pointer thickness type is thick pointer, then central projection method is confirmed as the pointer recognition methods, in said rectangle frame, respectively on gray level image, tone images, saturation degree image, in the dial plate zone, to the projection of pointer rotation center; Perhaps
-in image, draw out rectangle frame, make it comprise the position of current pointer; If said pointer thickness type is the microdactylia pin, then the pointer recognition methods is confirmed as in the Hough conversion, in said rectangle frame, on gray level image, tone images, saturation degree image, in the dial plate zone, utilize Hough change detection straight line successively.
Wherein, the said database of setting up is preserved (step 107) with calibration result and is comprised step: preserve following any or appoint a plurality of calibration results: the relation of homogeneous transformation matrix sequence, gauge pointer number, pointer thickness type, instrument range and the needle deflection angle of the average gray value in the template that generates by mark, the color space under the template, template zone, the template region point sequence on to the homogeneous transformation matrix in dial plate zone, template region to homogeneous transformation matrix, template region and the instrument dial plate profile of pointer rotation center region, the recognition methods of pointer, the identification parameter of pointer.
Further; Said instrument location (step 109) comprises step: carry out template matches in the image after image pre-service (108); Calculate the homogeneous transformation matrix of template and current marking image; And the homogeneous transformation matrix sequence of the point sequence of the template region of preserving in the binding data storehouse on to the homogeneous transformation matrix in dial plate zone and template region to homogeneous transformation matrix, template region and the instrument dial plate profile of pointer rotation center region can calculate the regional profile information of positional information, the dial plate in instrument dial plate zone, the positional information of pointer rotation center region.Said gradation of image adjustment (step 110) comprises step: calculate the average gray value of marked region, then to the adjustment of entire image gray scale, make the average gray value in its template that equals to preserve in database zone.Said pointer extracting (step 111) comprises step: according to dial plate zone that has calculated and pointer rotation center region; Gauge pointer recognition methods and identification parameter in the binding data storehouse; In the dial plate zone; Carry out central projection method or Hough conversion, calculate the deflection angle of pointer.Said total indicator reading calculates (step 112) and comprises step: according to the deflection angle of the pointer that has calculated, the relation of instrument range in the binding data storehouse and needle deflection angle calculates the reading of pointer.
Further particularly, as shown in Figure 3, the said off-line calibration stage comprises the steps:
Step (1), instrument IMAQ;
Step (2), image pre-service;
Step (3), mark study;
Step (4), instrument location position;
Step (5), meter performance study;
The automatic selection of step (6), pointer recognition methods and pointer identification parameter;
The foundation of step (7), database and calibration result are preserved.
Wherein, said step (1) instrument IMAQ comprises following substep:
Step (1A), camera is aimed at the instrument dial plate, obtain image, send into computing machine.Preferably, use colour imagery shot, picture format is 8 RGB images.
Said step (2) image pre-service comprises following substep:
Step (2A), color of image space conversion, decomposition.Image transitions, decomposition with step (1A) obtains obtain the image under tone space, saturation degree space, the gray space, are designated as tone images, saturation degree image and gray level image respectively.
Step (2B), figure image intensifying.Comprise medium filtering etc. and remove the noise in the meter diagram picture.
The study of said step (3) mark comprises following substep:
Step (3A), the zone of on image, selecting a texture information to enrich serve as a mark.Preferably, above-mentioned image is chosen for the most tangible zone of contrast in tone images, saturation degree image or the gray level image.Preferably, above-mentioned zone can be selected as follows: near somewhere around instrument, and paste a texture information and enrich, contrast tangible card, this card is served as a mark.
Step (3B), marked region is learnt, set up marker template.
The average gray value of step (3C), calculating marked region.
Said step (4) instrument location position comprises following substep:
Step (4A), on image, through sketch out the central area of instrument dial plate zone outline, pointer rotation of mouse.
Step (4B), in image the search (3B) described marker template.Result and the described instrument dial plate of step (4A) zone outline according to template matches; Calculate instrument dial plate regional center; Thereby calculate the homogeneous transformation matrix of template center to instrument dial plate regional center; Be designated as
position according to mark; And the center of the described pointer rotation of step (4A); Calculate the homogeneous transformation matrix of template center to the pointer rotation center; Be designated as
position according to mark; And the point sequence of the described instrument dial plate of step (4A) zone outline; Calculate the homogeneous transformation matrix sequence of template center, be designated as
(wherein subscript i representes numbering a little) to the point sequence.
The study of said step (5) meter performance comprises following substep:
Pointer number, the pointer thickness type of step (5A), input instrument.
The range ability and the corresponding needle deflection angle of step (5B), input instrument.
The automatic selection of said step (6) pointer recognition methods and pointer identification parameter comprises following substep:
Step (6A), utilize mouse in image, to draw out rectangle frame, make it comprise the position of current pointer.
Step (6B), basis (5A) definite pointer thickness type are selected different recognition methodss.Thick pointer uses central projection method, and the microdactylia pin uses the Hough conversion.
Step (6C), in step (4A) in the determined instrument dial plate zone, in gray level image, tone images, saturation degree image, carry out central projection method's (projection centre is confirmed by step (4A)) or Hough conversion respectively.
Step (6D1), if adopt central projection method, successively on described each image space of step (6C), in the dial plate zone, to the projection of pointer rotation center.Specific practice is: in the dial plate zone, do ray, the starting point of ray is the pointer rotation center, and the terminal point of ray is on the regional outline of dial plate a bit, and the inclination angle of ray and horizontal direction is designated as θ.To each bar ray, calculate the accumulated value Sum (θ) of the grey scale pixel value on the ray.Because pointer and background there are differences, so pointer should be positioned on Sum (θ) maximal value or the pairing ray of minimum value.
Wherein, should guarantee that the pointer that recognizes drops in the rectangle frame that (6A) indicated, it is correct also promptly guaranteeing under this image space, using this recognition methods and the resulting recognition result of parameter.
Preferably, image space should be selected as follows: the Sum (θ) of the feasible pointer area that identifies is maximum with the absolute value of the deviation of the average gray value in whole dial plate zone.
Step (6D10), if adopt the Hough conversion, on described each image space of step (6C), in the dial plate zone, utilize Hough change detection straight line successively.
Wherein, should guarantee that the pointer that recognizes drops in the rectangle frame that step (6A) indicated, it is correct also promptly guaranteeing under this image space, using this recognition methods and the resulting recognition result of parameter.
Preferably, image space should be selected as follows: the pixel number that straight line comprised at the feasible pointer place of identifying reaches maximum.
Said step (7) database is set up and the calibration result preservation comprises following substep:
Step (7A), preservation Template Information comprise: the template that image space, mark generated under the template, the average gray value of marked region.
Step (7B), preservation location position information comprise: the mark region is to the homogeneous transformation matrix in dial plate zone, the homogeneous transformation matrix that the pointer rotation center region is arrived in the mark region, the profile in dial plate zone.
Step (7C), preservation meter performance comprise: the relation of gauge pointer number, pointer thickness type, instrument range and needle deflection angle.
Step (7D), preservation identification parameter comprise: the recognition methods of pointer, the identification parameter of pointer.
Further particularly, as shown in Figure 4, the said ONLINE RECOGNITION stage comprises the steps:
Step (8), instrument IMAQ;
Step (9), image pre-service;
Step (10), instrument location;
Step (11), gradation of image adjustment;
Step (12), pointer extracting;
Step (13), total indicator reading calculate.
Wherein, said step (9) image pre-service comprises following substep:
Step (9A), the described database of read step (7A) obtain the affiliated image space of template.
Step (9B), with the coloured image that step (8) is obtained, be transformed under the described image space of step (9A).
Said step (10) instrument location comprises following substep:
Step (10A), read step (7A), (7B) described database; Obtain template that mark generates, marked region homogeneous transformation matrix to the dial plate zone; Be the homogeneous transformation matrix of
described in the step (4B) marked region to the pointer rotation center region; Be the homogeneous transformation matrix sequence of
described in the step (4B) mark center, be
(wherein subscript i representes numbering a little) described in the step (4B) to the point sequence.
Step (10B), in the image that step (9B) is obtained; Carry out template matches; Obtain the mark center position; Be designated as Mark (Mark_x; Mark_y), and this transformation matrix representation template of transformation matrix
with respect to the transformation relation of the skew of current mark, rotation, convergent-divergent.
Step (10C), according to the information and the described transformation matrix of step (10A) of the template matches of step (10B) gained, the position of the positional information of reckoner disk area, pointer rotation center.
Note dial plate regional center coordinate be Pannel (Pannel_x, Pannel_y),
Pointer rotation center coordinate be Center (Center_x, Center_y),
The point sequence of the profile of description list disk area is Contour (Contour_x
i, Contour_y
i)
Then formula is following:
Thus, just can confirm dial plate zone and pointer rotation center.
The adjustment of said step (11) gradation of image comprises following substep:
Step (11A), the described data of database of read step (7A) are obtained the average gray value of marked region, are designated as G (A).This promptly thinks under the standard illumination condition, the standard grayscale value of marked region.
Step (11B), according to the information of the template matches of step (10B) gained, calculate the average gray value of marked region this moment, be designated as G (B).Promptly think under the real work situation actual grey value of marked region.
Step (11C), the image that step (9B) is obtained carry out the gray scale adjustment, make the described actual grey value of step (11B) equal the described standard grayscale value of step (11A).Pixel coordinate is that (x, gray values of pixel points y) is that (x y), supposes that image is 8bit to g, and then the maximum gradation value of pixel is 255 on the note image.Then the gray scale adjustment is undertaken by following formula:
Said step (12) pointer extracting comprises following substep:
Step (12A), the described database of read step (7D) obtain the recognition methods and the identification parameter of pointer.
Step (12B), in the described dial plate of step (10C) zone, according to (12A) described method and parameter detecting pointer, calculate the deflection angle of pointer.
Step (12C), central projection method extract pointer.With maximal value central projection method is example, in (10C) described dial plate zone, is starting point with the described pointer rotation center of step (10C), calculates grey scale pixel value sum on this ray, and the pairing ray of its maximal value is pointer place straight line.
Pointer is extracted in step (12D), Hough conversion.In the described dial plate of step (10C) zone, Hough change detection straight line, in detected many straight lines, the maximum straight line of point that comprises is pointer place straight line.
Said step (13) total indicator reading calculates and comprises following substep:
Step (13A), the described data of database of read step (7C) are obtained the relation of instrument range and needle deflection angle.
Step (13B), according to the described needle deflection angle of step (12B) and (13A) relation of described instrument range and needle deflection angle, calculate the reading of pointer.Computing formula is following:
Wherein, d is a total indicator reading, θ
*Be (12B) described needle deflection angle, θ
0The pairing angle of pointer during for zero graduation, θ
1The corresponding angle of pointer during for full scale, D is total range of instrument.
More than specific embodiment of the present invention is described.It will be appreciated that the present invention is not limited to above-mentioned specific implementations, those skilled in the art can make various distortion or modification within the scope of the claims, and this does not influence flesh and blood of the present invention.
Claims (10)
1. one kind high robust gauge pointer image-recognizing method is characterized in that, comprises the step and the step in ONLINE RECOGNITION stage in the off-line calibration stage of carrying out successively, wherein:
The step in said off-line calibration stage comprises: adjust camera, make it aim at the instrument dial plate, carry out IMAQ; Institute's images acquired is carried out the image pre-service; On image, get a zone and serve as a mark, carry out mark study; At the instrument location position, after the meter performance study, confirm pointer recognition methods and pointer identification parameter, and set up database calibration result is preserved;
The step in said ONLINE RECOGNITION stage comprises: carry out IMAQ earlier; After institute's images acquired is carried out the image pre-service, only comprised the regional image of dial plate through the instrument location, then carry out the gradation of image adjustment; Pointer extracting is accomplished the calculating of total indicator reading at last.
2. high robust gauge pointer image-recognizing method according to claim 1; It is characterized in that; Said image pre-service, comprise step: image is carried out color space conversion, decomposition; Obtain the image under tone space, saturation degree space, the gray space, be designated as tone images, saturation degree image and gray level image respectively, and carry out figure image intensifying, medium filtering processing.
3. high robust gauge pointer image-recognizing method according to claim 1 and 2; It is characterized in that; Said mark study comprises step: select a color space; Abundant zone of texture information in this space hypograph is served as a mark,, and calculate the average gray value in this template zone its foundation and study template; Said instrument location position comprises step: the homogeneous transformation matrix sequence of recorder dial plate profile, the calculation template region point sequence on to the homogeneous transformation matrix in dial plate zone, template region to homogeneous transformation matrix, template region and the instrument dial plate profile of pointer rotation center region.
4. according to each described high robust gauge pointer image-recognizing method in the claim 1 to 3; It is characterized in that; Said meter performance study comprises step: confirm the pointer characteristic of this instrument and the relation of instrument range and needle deflection angle; Wherein, said pointer characteristic comprises pointer number and pointer thickness type.
5. high robust gauge pointer image-recognizing method according to claim 4 is characterized in that, said definite pointer recognition methods and pointer identification parameter comprise following arbitrary step:
-in image, draw out rectangle frame, make it comprise the position of current pointer; If said pointer thickness type is thick pointer, then central projection method is confirmed as the pointer recognition methods, in said rectangle frame, respectively on gray level image, tone images, saturation degree image, in the dial plate zone, to the projection of pointer rotation center; Perhaps
-in image, draw out rectangle frame, make it comprise the position of current pointer; If said pointer thickness type is the microdactylia pin, then the pointer recognition methods is confirmed as in the Hough conversion, in said rectangle frame, on gray level image, tone images, saturation degree image, in the dial plate zone, utilize Hough change detection straight line successively.
6. according to each described high robust gauge pointer image-recognizing method in the claim 1 to 5; It is characterized in that the said database of setting up is preserved calibration result and comprised step: preserve following any or appoint a plurality of calibration results: the relation of homogeneous transformation matrix sequence, gauge pointer number, pointer thickness type, instrument range and the needle deflection angle of the average gray value in the template that generates by mark, the color space under the template, template zone, the template region point sequence on to the homogeneous transformation matrix in dial plate zone, template region to homogeneous transformation matrix, template region and the instrument dial plate profile of pointer rotation center region, the recognition methods of pointer, the identification parameter of pointer.
7. high robust gauge pointer image-recognizing method according to claim 6; It is characterized in that; Said instrument location comprises step: in the pretreated image of image, carry out template matches; Calculate the homogeneous transformation matrix of template and current marking image; And the homogeneous transformation matrix sequence of the point sequence of the template region of preserving in the binding data storehouse on to the homogeneous transformation matrix in dial plate zone and template region to homogeneous transformation matrix, template region and the instrument dial plate profile of pointer rotation center region can calculate the regional profile information of positional information, the dial plate in instrument dial plate zone, the positional information of pointer rotation center region.
8. according to claim 6 or 7 described high robust gauge pointer image-recognizing methods; It is characterized in that; Said gradation of image adjustment comprises step: the average gray value that calculates marked region; To the adjustment of entire image gray scale, make the average gray value in its template that equals to preserve in database zone then.
9. according to each described high robust gauge pointer image-recognizing method in the claim 6 to 8; It is characterized in that; Said pointer extracting comprises step: according to dial plate zone that has calculated and pointer rotation center region, gauge pointer recognition methods and identification parameter in the binding data storehouse are in the dial plate zone; Carry out central projection method or Hough conversion, calculate the deflection angle of pointer.
10. according to each described high robust gauge pointer image-recognizing method in the claim 6 to 9; It is characterized in that; Said total indicator reading calculates and comprises step: according to the deflection angle of the pointer that has calculated; The relation of instrument range in the binding data storehouse and needle deflection angle calculates the reading of pointer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110359130.0A CN102521560B (en) | 2011-11-14 | 2011-11-14 | Instrument pointer image identification method of high-robustness rod |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110359130.0A CN102521560B (en) | 2011-11-14 | 2011-11-14 | Instrument pointer image identification method of high-robustness rod |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102521560A true CN102521560A (en) | 2012-06-27 |
| CN102521560B CN102521560B (en) | 2014-11-19 |
Family
ID=46292471
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110359130.0A Active CN102521560B (en) | 2011-11-14 | 2011-11-14 | Instrument pointer image identification method of high-robustness rod |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102521560B (en) |
Cited By (29)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102722166A (en) * | 2012-06-29 | 2012-10-10 | 山东电力集团公司电力科学研究院 | Intelligent vision detection system and state detection method of transformer substation device |
| CN103207987A (en) * | 2013-02-28 | 2013-07-17 | 华北电力大学 | Indicating value identification method of dial instrument |
| CN103528617A (en) * | 2013-09-27 | 2014-01-22 | 中国人民解放军空军工程大学 | Cockpit instrument automatic recognizing and detecting method and device |
| CN104050446A (en) * | 2014-05-30 | 2014-09-17 | 浙江大学 | Meter pointer image identification method based on pointer width character |
| CN104298994A (en) * | 2014-02-28 | 2015-01-21 | 郑州金惠计算机系统工程有限公司 | Pointer-type meter reading identification method and device |
| CN104392206A (en) * | 2014-10-24 | 2015-03-04 | 南京航空航天大学 | Image processing method for automatic pointer-type instrument reading recognition |
| CN104484857A (en) * | 2014-12-26 | 2015-04-01 | 国网重庆市电力公司电力科学研究院 | Instrument data reading method and system |
| CN105091922A (en) * | 2014-05-04 | 2015-11-25 | 广东工业大学 | Virtual watch plate based pointer reading identifying method |
| CN105203148A (en) * | 2015-09-30 | 2015-12-30 | 湖北工业大学 | Visual detection method for automobile combination instrument |
| CN105303168A (en) * | 2015-10-14 | 2016-02-03 | 南京第五十五所技术开发有限公司 | Multi-view pointer type instrument identification method and device |
| CN105530456A (en) * | 2014-09-28 | 2016-04-27 | 深圳中兴力维技术有限公司 | Method, device and system for substation monitoring |
| 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 |
| CN106446913A (en) * | 2016-09-20 | 2017-02-22 | 山东省计量科学研究院 | Pointer position extracting method of multi-dial pointer type instrument |
| CN107135372A (en) * | 2017-03-29 | 2017-09-05 | 浙江大学 | A kind of instrument real-time monitoring system and method based on image recognition |
| CN107231521A (en) * | 2017-04-29 | 2017-10-03 | 安徽慧视金瞳科技有限公司 | Camera automatic positioning method is used in a kind of meter reading identification |
| CN107239742A (en) * | 2017-05-11 | 2017-10-10 | 安徽慧视金瞳科技有限公司 | A kind of gauge pointer scale value calculating method |
| CN107301412A (en) * | 2017-05-08 | 2017-10-27 | 浙江工业大学 | A kind of readings of pointer type meters method based on image procossing |
| CN107563263A (en) * | 2016-07-01 | 2018-01-09 | 珠海优特电力科技股份有限公司 | Video image identification method and system based on instrument |
| CN109034170A (en) * | 2018-07-18 | 2018-12-18 | 东南大学 | A kind of round readings of pointer type meters method of switch cubicle detection device |
| CN109063637A (en) * | 2018-07-30 | 2018-12-21 | 安徽慧视金瞳科技有限公司 | A kind of instrument piecemeal scale detection method |
| CN109086763A (en) * | 2018-07-23 | 2018-12-25 | 国网辽宁省电力有限公司信息通信分公司 | A kind of pointer instrument read method and device |
| CN109242906A (en) * | 2018-09-14 | 2019-01-18 | 威斯泰（苏州）智能科技有限公司 | A kind of target image recognizer and method that calculus is cumulative |
| CN109632055A (en) * | 2019-01-16 | 2019-04-16 | 杭州丹纳计量科技有限公司 | A kind of mechanical water meter dynamic vision measurement method based on intelligent terminal |
| CN109858474A (en) * | 2019-01-08 | 2019-06-07 | 北京全路通信信号研究设计院集团有限公司 | A kind of detection of transformer oil surface temperature controller and recognition methods |
| CN112001934A (en) * | 2019-05-27 | 2020-11-27 | 株式会社东芝 | Reading system, moving object, reading method, and storage medium |
| CN112270329A (en) * | 2020-10-30 | 2021-01-26 | 北京华维国创电子科技有限公司 | Accurate MARK point acquisition and identification algorithm based on multi-level algorithm fusion |
| CN113191352A (en) * | 2021-03-23 | 2021-07-30 | 浙江理工大学 | Water meter pointer reading identification method based on target detection and binary image detection |
| CN113255547A (en) * | 2021-06-03 | 2021-08-13 | 上海申瑞继保电气有限公司 | Electric power meter pointer image contour calculation method |
| CN113408519A (en) * | 2021-07-09 | 2021-09-17 | 深圳市铁越电气有限公司 | Method and system for reading pointer instrument based on template rotation matching |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1693852A (en) * | 2005-01-28 | 2005-11-09 | 华南理工大学 | Instrument pointer automatic detection identification method and automatic reading method |
| CN102176228A (en) * | 2011-01-28 | 2011-09-07 | 河海大学常州校区 | Machine vision method for identifying dial plate information of multi-pointer instrument |
-
2011
- 2011-11-14 CN CN201110359130.0A patent/CN102521560B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1693852A (en) * | 2005-01-28 | 2005-11-09 | 华南理工大学 | Instrument pointer automatic detection identification method and automatic reading method |
| CN102176228A (en) * | 2011-01-28 | 2011-09-07 | 河海大学常州校区 | Machine vision method for identifying dial plate information of multi-pointer instrument |
Cited By (43)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102722166A (en) * | 2012-06-29 | 2012-10-10 | 山东电力集团公司电力科学研究院 | Intelligent vision detection system and state detection method of transformer substation device |
| CN103207987A (en) * | 2013-02-28 | 2013-07-17 | 华北电力大学 | Indicating value identification method of dial instrument |
| CN103207987B (en) * | 2013-02-28 | 2016-05-18 | 华北电力大学 | A kind of registration recognition methods of pointer instrument |
| CN103528617A (en) * | 2013-09-27 | 2014-01-22 | 中国人民解放军空军工程大学 | Cockpit instrument automatic recognizing and detecting method and device |
| CN103528617B (en) * | 2013-09-27 | 2016-08-10 | 中国人民解放军空军工程大学 | A kind of cockpit instrument identifies and detection method and device automatically |
| CN104298994B (en) * | 2014-02-28 | 2017-12-05 | 郑州金惠计算机系统工程有限公司 | A kind of readings of pointer type meters recognition methods and device |
| CN104298994A (en) * | 2014-02-28 | 2015-01-21 | 郑州金惠计算机系统工程有限公司 | Pointer-type meter reading identification method and device |
| CN105091922A (en) * | 2014-05-04 | 2015-11-25 | 广东工业大学 | Virtual watch plate based pointer reading identifying method |
| CN105091922B (en) * | 2014-05-04 | 2019-09-20 | 广东工业大学 | A kind of pointer gauge Recognition of Reading method based on virtual dial plate |
| CN104050446B (en) * | 2014-05-30 | 2017-05-03 | 浙江大学 | Meter pointer image identification method based on pointer width character |
| CN104050446A (en) * | 2014-05-30 | 2014-09-17 | 浙江大学 | Meter pointer image identification method based on pointer width character |
| CN105530456A (en) * | 2014-09-28 | 2016-04-27 | 深圳中兴力维技术有限公司 | Method, device and system for substation monitoring |
| CN104392206B (en) * | 2014-10-24 | 2017-11-24 | 南京航空航天大学 | A kind of image processing method of readings of pointer type meters automatic identification |
| CN104392206A (en) * | 2014-10-24 | 2015-03-04 | 南京航空航天大学 | Image processing method for automatic pointer-type instrument reading recognition |
| WO2016101643A1 (en) * | 2014-12-26 | 2016-06-30 | 国网重庆市电力公司电力科学研究院 | Meter data read method and system |
| CN104484857B (en) * | 2014-12-26 | 2017-08-18 | 国网重庆市电力公司电力科学研究院 | A kind of instrumented data read method and system |
| CN104484857A (en) * | 2014-12-26 | 2015-04-01 | 国网重庆市电力公司电力科学研究院 | Instrument data reading method and system |
| CN105203148A (en) * | 2015-09-30 | 2015-12-30 | 湖北工业大学 | Visual detection method for automobile combination instrument |
| CN105303168A (en) * | 2015-10-14 | 2016-02-03 | 南京第五十五所技术开发有限公司 | Multi-view pointer type instrument identification method and device |
| 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 |
| CN107563263A (en) * | 2016-07-01 | 2018-01-09 | 珠海优特电力科技股份有限公司 | Video image identification method and system based on instrument |
| CN106446913A (en) * | 2016-09-20 | 2017-02-22 | 山东省计量科学研究院 | Pointer position extracting method of multi-dial pointer type instrument |
| CN107135372A (en) * | 2017-03-29 | 2017-09-05 | 浙江大学 | A kind of instrument real-time monitoring system and method based on image recognition |
| CN107231521A (en) * | 2017-04-29 | 2017-10-03 | 安徽慧视金瞳科技有限公司 | Camera automatic positioning method is used in a kind of meter reading identification |
| CN107301412A (en) * | 2017-05-08 | 2017-10-27 | 浙江工业大学 | A kind of readings of pointer type meters method based on image procossing |
| CN107301412B (en) * | 2017-05-08 | 2019-12-17 | 浙江工业大学 | Pointer instrument reading method based on image processing |
| CN107239742A (en) * | 2017-05-11 | 2017-10-10 | 安徽慧视金瞳科技有限公司 | A kind of gauge pointer scale value calculating method |
| CN107239742B (en) * | 2017-05-11 | 2020-06-09 | 安徽慧视金瞳科技有限公司 | Method for calculating scale value of instrument pointer |
| CN109034170A (en) * | 2018-07-18 | 2018-12-18 | 东南大学 | A kind of round readings of pointer type meters method of switch cubicle detection device |
| CN109086763B (en) * | 2018-07-23 | 2021-11-26 | 国网辽宁省电力有限公司信息通信分公司 | Pointer instrument reading method and device |
| CN109086763A (en) * | 2018-07-23 | 2018-12-25 | 国网辽宁省电力有限公司信息通信分公司 | A kind of pointer instrument read method and device |
| CN109063637A (en) * | 2018-07-30 | 2018-12-21 | 安徽慧视金瞳科技有限公司 | A kind of instrument piecemeal scale detection method |
| CN109242906A (en) * | 2018-09-14 | 2019-01-18 | 威斯泰（苏州）智能科技有限公司 | A kind of target image recognizer and method that calculus is cumulative |
| CN109858474B (en) * | 2019-01-08 | 2021-10-19 | 北京全路通信信号研究设计院集团有限公司 | Detection and identification method for transformer oil surface temperature controller |
| CN109858474A (en) * | 2019-01-08 | 2019-06-07 | 北京全路通信信号研究设计院集团有限公司 | A kind of detection of transformer oil surface temperature controller and recognition methods |
| CN109632055A (en) * | 2019-01-16 | 2019-04-16 | 杭州丹纳计量科技有限公司 | A kind of mechanical water meter dynamic vision measurement method based on intelligent terminal |
| CN112001934A (en) * | 2019-05-27 | 2020-11-27 | 株式会社东芝 | Reading system, moving object, reading method, and storage medium |
| CN112270329A (en) * | 2020-10-30 | 2021-01-26 | 北京华维国创电子科技有限公司 | Accurate MARK point acquisition and identification algorithm based on multi-level algorithm fusion |
| CN113191352A (en) * | 2021-03-23 | 2021-07-30 | 浙江理工大学 | Water meter pointer reading identification method based on target detection and binary image detection |
| CN113255547A (en) * | 2021-06-03 | 2021-08-13 | 上海申瑞继保电气有限公司 | Electric power meter pointer image contour calculation method |
| CN113255547B (en) * | 2021-06-03 | 2024-02-09 | 上海申瑞继保电气有限公司 | Electric power instrument pointer image contour calculation method |
| CN113408519A (en) * | 2021-07-09 | 2021-09-17 | 深圳市铁越电气有限公司 | Method and system for reading pointer instrument based on template rotation matching |
| CN113408519B (en) * | 2021-07-09 | 2023-07-18 | 深圳市铁越电气有限公司 | Method and system for pointer instrument reading based on template rotation matching |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102521560B (en) | 2014-11-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102521560B (en) | Instrument pointer image identification method of high-robustness rod | |
| CN107609557B (en) | Pointer instrument reading identification method | |
| CN112818988B (en) | Automatic identification reading method and system for pointer instrument | |
| EP2194725B1 (en) | Method and apparatus for correcting a depth image | |
| CN101033962B (en) | Measurement method for displacement of model experiment based on optics | |
| CN109558871B (en) | Pointer instrument reading identification method and device | |
| CN109409290B (en) | Thermometer verification reading automatic identification system and method | |
| CN106845514B (en) | Deep learning-based reading judgment method and device for pointer type dial plate | |
| CN110503144A (en) | A kind of pointer instrument recognition methods for crusing robot | |
| CN104050446A (en) | Meter pointer image identification method based on pointer width character | |
| CN112801094B (en) | Pointer instrument image inclination correction method | |
| CN109447062A (en) | Pointer-type gauges recognition methods based on crusing robot | |
| CN109389165A (en) | Oil level gauge for transformer recognition methods based on crusing robot | |
| CN109086763A (en) | A kind of pointer instrument read method and device | |
| CN108055532A (en) | Automate the method and apparatus of matching test card | |
| CN113688817A (en) | Instrument identification method and system for automatic inspection | |
| CN106485739B (en) | A kind of point set method for registering based on L2 distance | |
| CN116310263A (en) | Pointer type aviation horizon instrument indication automatic reading implementation method | |
| CN112990190A (en) | Method and device for identifying reading of pointer instrument | |
| CN109087265B (en) | Multi-camera image coordinate conversion method and device | |
| CN109784257A (en) | A kind of detection of transformer thermometer and recognition methods | |
| Ueda et al. | A two-stage deep learning-based approach for automatic reading of analog meters | |
| CN113269749B (en) | Strip position image data collection method and visual detection method thereof | |
| CN115115889A (en) | Instrument image analysis method and device | |
| CN109444150A (en) | Contactless crackle measurement method and its device |
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 |