CN110298285A - A kind of Meter recognition method, apparatus and computer storage medium - Google Patents
A kind of Meter recognition method, apparatus and computer storage medium Download PDFInfo
- Publication number
- CN110298285A CN110298285A CN201910548176.3A CN201910548176A CN110298285A CN 110298285 A CN110298285 A CN 110298285A CN 201910548176 A CN201910548176 A CN 201910548176A CN 110298285 A CN110298285 A CN 110298285A
- Authority
- CN
- China
- Prior art keywords
- identified
- instrument
- region
- calibration
- color
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of Meter recognition method, apparatus and computer storage mediums, which comprises treats identifier table section and carries out color calibration, obtains color calibration region；Obtain the Instrument image to be identified for corresponding to the instrument region to be identified；The color calibration information corresponded in the Instrument image to be identified is obtained according to the color calibration region；The Instrument image to be identified is split according to the color calibration information, obtains target Instrument image.The step of embodiment of the present invention is by first tentatively carrying out color calibration to instrument region, then the method for being partitioned into target Instrument image is identified from entire Instrument image to be identified according to the color region demarcated, much complex ambient noise can be simply and effectively excluded, so that facilitating Meter recognition machine can be improved recognition accuracy to instrument.
Description
Technical field
The present invention relates to image identification technical field more particularly to a kind of Meter recognition method, apparatus and computer to store
Medium.
Background technique
In the production and living of modern society, digital instrument is the measuring instrumentss for being widely used in various fields, price
Cheaply, enormous amount, it is many kinds of.The accurate digital instrument pair and industrial production of reading has highly important effect.
In many enterprises, there is a large amount of instrument, if completing to read by people, such workload is very big, no
It is only at high cost, and error rate is also higher.Therefore the prior art usually reduces the labor of worker by instrument automatic identification technology
Fatigue resistance, while improving the degree of automation of meter reading.But existing Meter recognition machine is usually because of instrument sample to be identified
Formula is excessive, and background environment is too complicated and error rate is higher when leading to automatic identification, causes many inconvenience to production and living.
Therefore, Meter recognition machine how is improved to ask the recognition accuracy of instrument as current technology urgently to be resolved
Topic.
Summary of the invention
The embodiment of the present invention creatively provides a kind of instrument to effectively overcome drawbacks described above present in the prior art
Recognition methods, device and computer storage medium, which comprises treat identifier table section and carry out color calibration, obtain
Color calibration region；Obtain the Instrument image to be identified for corresponding to the instrument region to be identified；According to the color calibration area
Domain obtains the color calibration information corresponded in the Instrument image to be identified；According to the color calibration information to described wait know
Other Instrument image is split, and obtains target Instrument image.
In an embodiment, the identifier table section for the treatment of carries out color calibration, obtains color calibration region packet
It includes: the choosing of circumferential profile frame being carried out to the instrument region to be identified, obtains frame favored area to be identified；To the frame constituency to be identified
Domain carries out color mark, obtains color calibration region.
In an embodiment, described obtained according to the color calibration region corresponds to the Instrument image to be identified
In color calibration information include: according to the color space vector in the color calibration region extract correspond to the instrument to be identified
Calibration region contour in table image；The Instrument image to be identified is retrieved according to the calibration region contour, is obtained
Demarcate area coordinate.
It is described that the Instrument image to be identified is divided according to the color calibration information in an embodiment
It cuts, obtaining target Instrument image includes: to be determined to correspond in the Instrument image to be identified according to the calibration area coordinate
Target instrument region；The Instrument image to be identified is split according to target instrument region, obtains target meter diagram
Picture.
Another aspect of the present invention provides a kind of Meter recognition device, and described device includes: color calibration module, for treating
Identifier table section carries out color calibration, obtains color calibration region；Image collection module, for obtaining corresponding to described wait know
The Instrument image to be identified in other instrument region；Calibration obtains module, corresponds to institute for obtaining according to the color calibration region
State the color calibration information in Instrument image to be identified；Image segmentation module is used for according to the color calibration information to described
Instrument image to be identified is split, and obtains target Instrument image.
In an embodiment, the color calibration module includes: frame-choosing unit, for the gauge field to be identified
Domain carries out the choosing of circumferential profile frame, obtains frame favored area to be identified；Coloring unit, for the frame favored area to be identified into
Row color mark obtains color calibration region.
In an embodiment, it includes: calibration contours extract unit that the calibration, which obtains module, for according to the face
The color space vector that colour code determines region extracts the calibration region contour corresponded in the Instrument image to be identified；Demarcate coordinate
Extraction unit obtains calibration region and sits for being retrieved according to the calibration region contour to the Instrument image to be identified
Mark.
In an embodiment, it includes: target area extraction unit that described image, which divides module, for according to the mark
Determine area coordinate and determines the target instrument region corresponded in the Instrument image to be identified；Target image cutting unit, is used for
The Instrument image to be identified is split according to target instrument region, obtains target Instrument image.
Another aspect of the present invention provides a kind of computer readable storage medium, and the storage medium includes one group of computer can
It executes instruction, when executed for executing Meter recognition method described in any of the above embodiments.
The embodiment of the present invention is to solve existing Meter recognition machine because instrument pattern to be identified is excessive, and background environment is too multiple
Miscellaneous and error rate higher problem when leading to automatic identification, provides a kind of Meter recognition method, apparatus and computer storage is situated between
Matter is to improve Meter recognition machine to the recognition accuracy of instrument.The embodiment of the present invention first treats identifier table section and carries out color
Calibration, by tentatively by instrument approximate region calibrate come method can effectively exclude more jamming pattern, obtain less make an uproar
The color calibration region of sound；Then the meter diagram to be identified for not removing ambient noise for corresponding to instrument region to be identified is first obtained
Picture, acquisition modes here can be by manually taking pictures, scanning or machine shooting, the modes such as import；Then according to above-mentioned face
Colour code determines the corresponding color calibration information obtained in Instrument image to be identified in region, and wherein color calibration information includes to correspond to
The colouring information etc. in color calibration region；Finally identification Instrument image is treated according to color calibration information obtained to be divided
It cuts, i.e., identifies the target being partitioned into entire Instrument image to be identified according to information such as the colors for corresponding to color calibration region
Instrument image.The step of embodiment of the present invention is by first tentatively carrying out color calibration to instrument region, then according to being demarcated
Color region identifies the method for being partitioned into target Instrument image from entire Instrument image to be identified, can simply and effectively exclude
Much complex ambient noise, so that facilitating Meter recognition machine can be improved recognition accuracy to instrument.
Detailed description of the invention
Fig. 1 is a kind of implementation process schematic diagram of Meter recognition method provided by one embodiment of the invention；
Fig. 2 is a kind of a kind of schematic diagram in instrument region to be identified provided by one embodiment of the invention；
Fig. 3 is a kind of a kind of schematic diagram in color calibration region provided by one embodiment of the invention；
Fig. 4 is a kind of a kind of schematic diagram of color calibration information provided by one embodiment of the invention；
Fig. 5 is a kind of a kind of schematic diagram of target Instrument image provided by one embodiment of the invention；
Fig. 6 is a kind of composed structure schematic diagram of Meter recognition device provided by one embodiment of the invention.
Specific embodiment
To keep the purpose of the present invention, feature, advantage more obvious and understandable, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is only
It is only a part of the embodiment of the present invention, and not all embodiments.Based on the embodiments of the present invention, those skilled in the art are not having
Every other embodiment obtained under the premise of creative work is made, shall fall within the protection scope of the present invention.
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means specific features, structure, material or spy described in conjunction with this embodiment or example
Point is included at least one embodiment or example of the invention.Moreover, particular features, structures, materials, or characteristics described
It may be combined in any suitable manner in any one or more of the embodiments or examples.In addition, without conflicting with each other, this
The technical staff in field can be by the spy of different embodiments or examples described in this specification and different embodiments or examples
Sign is combined.
In addition, term " first ", " second " are used for descriptive purposes only and cannot be understood as indicating or suggesting relative importance
Or implicitly indicate the quantity of indicated technical characteristic." first " is defined as a result, the feature of " second " can be expressed or hidden
It include at least one this feature containing ground.In the description of the present invention, the meaning of " plurality " is two or more, unless otherwise
Clear specific restriction.
Fig. 1 to Fig. 5 is please referred to, on the one hand the embodiment of the present invention provides a kind of Meter recognition method, method includes:
Step 101, it treats identifier table section and carries out color calibration, obtain color calibration region；
Step 102, the Instrument image to be identified for corresponding to instrument region to be identified is obtained；
Step 103, the color calibration information corresponded in Instrument image to be identified is obtained according to color calibration region；
Step 104, identification Instrument image is treated according to color calibration information to be split, obtain target Instrument image.
The embodiment of the present invention is to solve existing Meter recognition machine because instrument pattern to be identified is excessive, and background environment is too multiple
Miscellaneous and error rate higher problem when leading to automatic identification provides a kind of Meter recognition method to improve Meter recognition machine pair
The recognition accuracy of instrument.The embodiment of the present invention first passes through step 101 and treats identifier table section progress color calibration, by first
Instrument approximate region is calibrated the method come by step can effectively exclude more jamming pattern, obtain the color calibration of less noise
Region, such as Fig. 3；Then it is first obtained by step 102 and does not remove the to be identified of ambient noise corresponding to instrument region to be identified
Instrument image, acquisition modes here can be by manually taking pictures, scanning or machine shooting, the modes such as import；Then pass through step
Rapid 103 according to the corresponding color calibration information obtained in Instrument image to be identified in above-mentioned color calibration region, such as Fig. 4, wherein
Color calibration information includes the colouring information etc. corresponding to color calibration region；Finally by step 104 according to obtained
Color calibration information is treated identification Instrument image and is split, i.e., is known according to information such as the colors for corresponding to color calibration region
The target Instrument image not being partitioned into entire Instrument image to be identified, such as Fig. 5.The embodiment of the present invention passes through first tentatively to instrument
Region carries out the step of color calibration, is then identified and is divided from entire Instrument image to be identified according to the color region demarcated
The method of target Instrument image out can simply and effectively exclude much complex ambient noise, to facilitate Meter recognition machine
Device can be improved the recognition accuracy to instrument.
In an embodiment, treats identifier table section and carry out color calibration, obtaining color calibration region includes:
It treats identifier table section and carries out the choosing of circumferential profile frame, obtain frame favored area to be identified；
Color mark is carried out to frame favored area to be identified, obtains color calibration region.
In the embodiment of the present invention, treating the method that identifier table section carries out color calibration includes: first tentatively to carry out instrument
The choosing of circumferential profile frame, removes more ambient noise, obtains frame favored area to be identified；Wherein, for wait know in frame favored area to be identified
Other instrument covers meter reading when to avoid subsequent progress color mark, can be only using hollow frame favored area as to be identified
Frame favored area, in order to subsequent machine recognition reading；Then color mark is carried out to obtained frame favored area to be identified, so that this
Part colours calibration region can have preferable environment-identification, so that facilitating Meter recognition machine can be improved to instrument
Recognition accuracy.
In an embodiment, the color calibration corresponded in Instrument image to be identified is obtained according to color calibration region
Information includes:
Correspond to Instrument image to be identified by exposure mask function to extract according to the color space vector in color calibration region
In calibration region contour；
Identification Instrument image is treated according to calibration region contour to be retrieved, and calibration area coordinate is obtained.
In the embodiment of the present invention, after being shot under different light or environment due to the calibration color in color calibration region
It can show different color effects, therefore, the embodiment of the present invention will shoot the resulting BGR for being only capable of representing fixed color
(Blue-Green-Red, bluish-green red) image be converted into color space vector HSV (Hue, Saturation, Value,
Tone, saturation degree, lightness) format-pattern, more extensive color gamut can be expressed, so that color calibration region is complete
It identifies.And exposure mask function can be extracted by the color space vector in color calibration region corresponding to instrument to be identified
Calibration area coordinate in table image, function call are as follows:
When use blue as demarcate color when, color space ranges of vectors can be set as [110,50,50]-[130,
255,255], then corresponding order is,
Hsv=cv2.cvtColor (img, cv2.COLOR_BGR2HSV)
Lower_blue=np.array ([110,50,50])
Upper_blue=np.array ([130,255,255])
Mask=cv2.inRange (hsv, lower_blue, upper_blue)
To obtain the binary image that can show calibration region contour as shown in Figure 4；Then further according to acquired
Calibration region contour treat identification Instrument image carry out profile retrieval, just obtained corresponding to calibration region contour calibration area
Domain coordinate, retrieval functions are as follows:
Img, contours, hierarchy=cv2.findContours (mask, cv2.RETR_EXTERNAL,
cv2.CHAIN_APPROX_NONE)
In this way, calibration region just effectively be located out by the present invention in Instrument image to be identified, facilitate into one
The segmentation to calibration region is walked, to obtain the higher target Instrument image of accuracy rate.
In an embodiment, identification Instrument image is treated according to color calibration information and is split, target instrument is obtained
Table image includes:
The target instrument region corresponded in Instrument image to be identified is determined according to calibration area coordinate；
Identification Instrument image is treated according to target instrument region to be split, and obtains target Instrument image.
In the embodiment of the present invention, the target gauge field corresponded in Instrument image to be identified is determined according to calibration area coordinate
Domain is specifically as follows, and obtains the top left corner pixel point coordinate in calibration area coordinate and demarcates the width and height in region；Root
Rectangular target instrument region is determined according to top left corner pixel point coordinate obtained and width and height, and according to mesh obtained
Mark instrument region is split, and obtains target instrument region.Action command is as follows:
X, y are top left corner pixel point coordinate, and w is the width of rectangle, and h is the height of rectangle,
for cnt in range(len(contours)):
X, y, w, h=cv2.boundingRect (contours [cnt])
Cv2.imwrite (' %s.jpg ' %cnt, img [y+10:y+h-10, x+10:x+w-10])
It is split relative to directly identification Instrument image is treated according to calibration area coordinate, passes through characteristic coordinates point in this way
And width and height establish the mode of rectangular target areas, can obtain more convenient for cutting and the target instrument region saved.
Fig. 6 is a kind of composed structure schematic diagram of Meter recognition device provided by one embodiment of the invention, please refers to figure
6, another aspect of the present invention provides a kind of Meter recognition device, and device includes:
Color calibration module 201 carries out color calibration for treating identifier table section, obtains color calibration region；
Image collection module 202, for obtaining the Instrument image to be identified for corresponding to instrument region to be identified；
Calibration obtains module 203, for obtaining the color corresponded in Instrument image to be identified according to color calibration region
Calibration information；
Image segmentation module 204 is split for treating identification Instrument image according to color calibration information, obtains target
Instrument image.
The embodiment of the present invention is to solve existing Meter recognition machine because instrument pattern to be identified is excessive, and background environment is too multiple
Miscellaneous and error rate higher problem when leading to automatic identification provides a kind of Meter recognition method to improve Meter recognition machine pair
The recognition accuracy of instrument.The embodiment of the present invention first passes through color calibration module 201 and treats identifier table section progress color mark
It is fixed, by tentatively by instrument approximate region calibrate come method can effectively exclude more jamming pattern, obtain less noise
Color calibration region；Then it is first obtained by image collection module 202 and does not remove background corresponding to instrument region to be identified
The Instrument image to be identified of noise, acquisition modes here can be by manually taking pictures, scanning or machine shooting, the side such as import
Formula；Then module 203 is obtained according to the corresponding face obtained in Instrument image to be identified in above-mentioned color calibration region by calibration
Color calibration information, wherein color calibration information includes the colouring information etc. corresponding to color calibration region；Finally by image
Segmentation module 204 is treated identification Instrument image according to color calibration information obtained and is split, i.e., according to corresponding to color
The information such as the color in region are demarcated to identify the target Instrument image being partitioned into entire Instrument image to be identified.The present invention is implemented
The step of example is by first tentatively carrying out color calibration to instrument region, then according to the color region demarcated from entire to be identified
The method for being partitioned into target Instrument image is identified in Instrument image, can simply and effectively exclude much complex ambient noise, from
And facilitates Meter recognition machine and can be improved recognition accuracy to instrument.
In an embodiment, color calibration module 201 includes:
Frame-choosing unit carries out the choosing of circumferential profile frame for treating identifier table section, obtains frame favored area to be identified；
Coloring unit obtains color calibration region for carrying out color mark to frame favored area to be identified.
In the embodiment of the present invention, color calibration module 201 includes: to first pass through frame-choosing unit tentatively to carry out instrument circumferential profile
Frame choosing, removes more ambient noise, obtains frame favored area to be identified；Wherein, it is instrument to be identified in frame favored area to be identified, is
Avoid covering meter reading when subsequent progress color mark, can only using hollow frame favored area as frame favored area to be identified,
In order to subsequent machine recognition reading；Then color mark is carried out to obtained frame favored area to be identified by coloring unit
Note enables this part colours calibration region to have preferable environment-identification, so that facilitating Meter recognition machine can mention
Recognition accuracy of the height to instrument.
In an embodiment, calibration obtains module 203 and includes:
Contours extract unit is demarcated, corresponds to instrument to be identified for extracting according to the color space vector in color calibration region
Calibration region contour in table image；
Coordinate extraction unit is demarcated, is retrieved, is marked for treating identification Instrument image according to calibration region contour
Determine area coordinate.
In the embodiment of the present invention, after being shot under different light or environment due to the calibration color in color calibration region
It can show different color effects, therefore, the embodiment of the present invention will shoot the resulting BGR image for being only capable of representing fixed color
It is converted into the HSV format-pattern with color space vector, more extensive color gamut can be expressed, thus by color calibration
Region completely identifies.And exposure mask function can extract correspondence by the color space vector in color calibration region
Calibration area coordinate in Instrument image to be identified, wherein the function call of calibration contours extract unit is as follows:
When use blue as demarcate color when, color space ranges of vectors can be set as [110,50,50]-[130,
255,255], then corresponding order is,
Hsv=cv2.cvtColor (img, cv2.COLOR_BGR2HSV)
Lower_blue=np.array ([110,50,50])
Upper_blue=np.array ([130,255,255])
Mask=cv2.inRange (hsv, lower_blue, upper_blue)
To obtain the binary image that can show calibration region contour as shown in Figure 4；Then it is sat again by calibration
Mark extraction unit is treated identification Instrument image according to obtained calibration region contour and is retrieved, and has just obtained corresponding to calibration
The calibration area coordinate of region contour, retrieval functions are as follows:
Img, contours, hierarchy=cv2.findContours (mask, cv2.RETR_EXTERNAL,
cv2.CHAIN_APPROX_NONE)
In this way, calibration region just effectively be located out by the present invention in Instrument image to be identified, facilitate into one
The segmentation to calibration region is walked, to obtain the higher target Instrument image of accuracy rate.
In an embodiment, image segmentation module 204 includes:
Target area extraction unit, for determining the target corresponded in Instrument image to be identified according to calibration area coordinate
Instrument region；
Target image cutting unit is split for treating identification Instrument image according to target instrument region, obtains mesh
Mark Instrument image.
In the embodiment of the present invention, target area extraction unit is determined according to calibration area coordinate corresponds to meter diagram to be identified
Target instrument region as in is specifically as follows, and obtains the upper left corner picture in calibration area coordinate by target area extraction unit
Vegetarian refreshments coordinate and the width and height for demarcating region；It is true according to top left corner pixel point coordinate obtained and width and height
Determine rectangular target instrument region, and be split by target image cutting unit according to target instrument obtained region, is obtained
To target instrument region.Action command is as follows:
X, y are top left corner pixel point coordinate, and w is the width of rectangle, and h is the height of rectangle,
for cnt in range(len(contours)):
X, y, w, h=cv2.boundingRect (contours [cnt])
Cv2.imwrite (' %s.jpg ' %cnt, img [y+10:y+h-10, x+10:x+w-10])
It is split relative to directly identification Instrument image is treated according to calibration area coordinate, passes through characteristic coordinates point in this way
And width and height establish the mode of rectangular target areas, can obtain more convenient for cutting and the target instrument region saved.
Another aspect of the present invention provides a kind of computer readable storage medium, and storage medium includes that one group of computer is executable
Instruction, when executed for executing any one of above-mentioned Meter recognition method.
The embodiment of the present invention is to solve existing Meter recognition machine because instrument pattern to be identified is excessive, and background environment is too multiple
Miscellaneous and error rate higher problem when leading to automatic identification provides a kind of Meter recognition method to improve Meter recognition machine pair
The recognition accuracy of instrument.The embodiment of the present invention, which first passes through, treats identifier table section progress color calibration, by tentatively by instrument
Table approximate region, which calibrates the method come, can effectively exclude more jamming pattern, obtain the color calibration region of less noise；
Then correspond to the Instrument image to be identified for not removing ambient noise in instrument region to be identified, acquisition side here by obtaining
Formula can by manually take pictures, scan or machine shooting, import etc. modes；Then by according to above-mentioned color calibration region pair
The color calibration information in Instrument image to be identified should be obtained, wherein color calibration information includes corresponding to color calibration region
Colouring information etc.；It finally treats identification Instrument image according to color calibration information obtained to be split, i.e., according to correspondence
The target Instrument image being partitioned into entire Instrument image to be identified is identified in information such as the colors in color calibration region.This hair
The step of bright embodiment is by first tentatively carrying out color calibration to instrument region, then according to the color region demarcated from entire
The method for being partitioned into target Instrument image is identified in Instrument image to be identified, can simply and effectively be excluded much complex background and be made an uproar
Sound, so that facilitating Meter recognition machine can be improved recognition accuracy to instrument.
More than, only a specific embodiment of the invention, but scope of protection of the present invention is not limited thereto, and it is any to be familiar with
Those skilled in the art in the technical scope disclosed by the present invention, can easily think of the change or the replacement, and should all cover
Within protection scope of the present invention.Therefore, protection scope of the present invention should be subject to the protection scope in claims.
Claims (9)
1. a kind of Meter recognition method, which is characterized in that the described method includes:
It treats identifier table section and carries out color calibration, obtain color calibration region；
Obtain the Instrument image to be identified for corresponding to the instrument region to be identified；
The color calibration information corresponded in the Instrument image to be identified is obtained according to the color calibration region；
The Instrument image to be identified is split according to the color calibration information, obtains target Instrument image.
2. being obtained the method according to claim 1, wherein the identifier table section for the treatment of carries out color calibration
Include: to color calibration region
The choosing of circumferential profile frame is carried out to the instrument region to be identified, obtains frame favored area to be identified；
Color mark is carried out to the frame favored area to be identified, obtains color calibration region.
3. the method according to claim 1, wherein described obtained according to the color calibration region corresponds to institute
The color calibration information stated in Instrument image to be identified includes:
The calibration area corresponded in the Instrument image to be identified is extracted according to the color space vector in the color calibration region
Domain profile；
The Instrument image to be identified is retrieved according to the calibration region contour, obtains calibration area coordinate.
4. according to the method described in claim 3, it is characterized in that, it is described according to the color calibration information to described to be identified
Instrument image is split, and is obtained target Instrument image and is included:
The target instrument region corresponded in the Instrument image to be identified is determined according to the calibration area coordinate；
The Instrument image to be identified is split according to target instrument region, obtains target Instrument image.
5. a kind of Meter recognition device, which is characterized in that described device includes:
Color calibration module carries out color calibration for treating identifier table section, obtains color calibration region；
Image collection module, for obtaining the Instrument image to be identified for corresponding to the instrument region to be identified；
Calibration obtains module, for obtaining the color corresponded in the Instrument image to be identified according to the color calibration region
Calibration information；
Image segmentation module obtains mesh for being split according to the color calibration information to the Instrument image to be identified
Mark Instrument image.
6. device according to claim 5, which is characterized in that the color calibration module includes:
Frame-choosing unit obtains frame favored area to be identified for carrying out the choosing of circumferential profile frame to the instrument region to be identified；
Coloring unit obtains color calibration region for carrying out color mark to the frame favored area to be identified.
7. device according to claim 5, which is characterized in that the calibration obtains module and includes:
Contours extract unit is demarcated, for extracting corresponding to described according to the color space vector in the color calibration region wait know
Calibration region contour in other Instrument image；
Calibration coordinate extraction unit is obtained for being retrieved according to the calibration region contour to the Instrument image to be identified
To calibration area coordinate.
8. device according to claim 7, which is characterized in that described image divides module and includes:
Target area extraction unit corresponds in the Instrument image to be identified for being determined according to the calibration area coordinate
Target instrument region；
Target image cutting unit is obtained for being split according to target instrument region to the Instrument image to be identified
To target Instrument image.
9. a kind of computer readable storage medium, which is characterized in that the storage medium includes a group of computer-executable instructions,
The described in any item Meter recognition methods of 1-4 are required for perform claim when executed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910548176.3A CN110298285A (en) | 2019-06-24 | 2019-06-24 | A kind of Meter recognition method, apparatus and computer storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910548176.3A CN110298285A (en) | 2019-06-24 | 2019-06-24 | A kind of Meter recognition method, apparatus and computer storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110298285A true CN110298285A (en) | 2019-10-01 |
Family
ID=68028598
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910548176.3A Pending CN110298285A (en) | 2019-06-24 | 2019-06-24 | A kind of Meter recognition method, apparatus and computer storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110298285A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111222425A (en) * | 2019-12-26 | 2020-06-02 | 新绎健康科技有限公司 | Method and device for positioning facial features |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000246874A (en) * | 1999-03-03 | 2000-09-12 | Dainippon Printing Co Ltd | Method and apparatus for correcting color |
| CN101498592A (en) * | 2009-02-26 | 2009-08-05 | 北京中星微电子有限公司 | Reading method and apparatus for pointer instrument |
| CN106529556A (en) * | 2016-11-16 | 2017-03-22 | 国家电网公司 | Visual inspection system for instrument indicator lamp |
| CN106557751A (en) * | 2016-11-22 | 2017-04-05 | 中国石油大学(华东) | A kind of pointer pressure automatic reading method with color configured information |
| CN109447061A (en) * | 2018-09-29 | 2019-03-08 | 南京理工大学 | Reactor oil level indicator recognition methods based on crusing robot |
-
2019
- 2019-06-24 CN CN201910548176.3A patent/CN110298285A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000246874A (en) * | 1999-03-03 | 2000-09-12 | Dainippon Printing Co Ltd | Method and apparatus for correcting color |
| CN101498592A (en) * | 2009-02-26 | 2009-08-05 | 北京中星微电子有限公司 | Reading method and apparatus for pointer instrument |
| CN106529556A (en) * | 2016-11-16 | 2017-03-22 | 国家电网公司 | Visual inspection system for instrument indicator lamp |
| CN106557751A (en) * | 2016-11-22 | 2017-04-05 | 中国石油大学(华东) | A kind of pointer pressure automatic reading method with color configured information |
| CN109447061A (en) * | 2018-09-29 | 2019-03-08 | 南京理工大学 | Reactor oil level indicator recognition methods based on crusing robot |
Non-Patent Citations (4)
| Title |
|---|
| ALVIN93: ""Opencv轮廓检测"", 《HTTPS://BLOG.CSDN.NET/ALVIN93/ARTICLE/DETAILS/80940989》 * |
| TIGER_JIANG: ""使用opencv进行简单的手势检测[by Python]"", 《HTTPS://WWW.BBSMAX.COM/A/LPDOLABYZ3/》 * |
| 臧雪: "巡检机器人自主仪表识别视觉系统的设计", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 隐士2018: ""cv2.boundingRect(cnt)以及cv2.rectangle"", 《HTTPS://DEVELOPER.ALIYUN.COM/ARTICLE/541348》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111222425A (en) * | 2019-12-26 | 2020-06-02 | 新绎健康科技有限公司 | Method and device for positioning facial features |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103518224B (en) | Method for analysing microbial growth | |
| CN103186904B (en) | Picture contour extraction method and device | |
| CN105976354B (en) | Color and gradient based component positioning method and system | |
| Berezhnoy et al. | Computer analysis of van Gogh’s complementary colours | |
| CN108921804A (en) | Distort the bearing calibration of file and picture | |
| US8542917B2 (en) | System and method for identifying complex tokens in an image | |
| US8559714B2 (en) | Post processing for improved generation of intrinsic images | |
| CN103839283A (en) | Area and circumference nondestructive measurement method of small irregular object | |
| US5446543A (en) | Method and apparatus for extracting a pattern of color from an object using a neural network | |
| US20120072454A1 (en) | Apparatus and method for identifying the creator of a work of art | |
| CN110569774B (en) | An Automatic Digitization Method of Line Chart Image Based on Image Processing and Pattern Recognition | |
| US8249342B1 (en) | Color analytics for a digital image | |
| Bhookya et al. | Yield estimation of chilli crop using image processing techniques | |
| CN115880297B (en) | Quilt cover dyeing quality assessment method based on machine vision | |
| CN108564631A (en) | Car light light guide acetes chinensis method, apparatus and computer readable storage medium | |
| US20030012430A1 (en) | Process of identifying the sky in an image and an image obtained using this process | |
| CN109507198A (en) | Mask detection system and method based on Fast Fourier Transform (FFT) and linear Gauss | |
| US3745037A (en) | Method for producing diagrammed work surfaces on which are created artistic renderings of multi-color images | |
| Guarneri et al. | 3D remote colorimetry and watershed segmentation techniques for fresco and artwork decay monitoring and preservation | |
| CN111429462A (en) | License plate positioning method based on edge detection and mathematical morphology | |
| CN109993202A (en) | A kind of line chirotype shape similarity judgment method, electronic equipment and storage medium | |
| CN110044485B (en) | Image type fabric color measuring method | |
| CN112990064A (en) | Dial pointer identification method based on color segmentation and probability model | |
| CN110298285A (en) | A kind of Meter recognition method, apparatus and computer storage medium | |
| CN113781476B (en) | Textile dyeing quality assessment method and system based on image processing |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191001 |