CN106446913A - Pointer position extracting method of multi-dial pointer type instrument - Google Patents
Pointer position extracting method of multi-dial pointer type instrument Download PDFInfo
- Publication number
- CN106446913A CN106446913A CN201610832328.9A CN201610832328A CN106446913A CN 106446913 A CN106446913 A CN 106446913A CN 201610832328 A CN201610832328 A CN 201610832328A CN 106446913 A CN106446913 A CN 106446913A
- Authority
- CN
- China
- Prior art keywords
- image
- dial plate
- pointer
- computer
- circle
- 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
- 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
- G06V10/443—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 by matching or 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
- G06V10/20—Image preprocessing
- G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
-
- 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
- G06V10/34—Smoothing or thinning of the pattern; Morphological operations; Skeletonisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/02—Recognising information on displays, dials, clocks
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
A pointer position extracting method of a multi-dial pointer type instrument comprises the steps of a), inputting an acquired original image into a computer; b), converting the acquired original image into a gray-scale image; c), performing binarization processing, and obtaining an instrument edge in the image by means of a Canny edge detection algorithm; d), performing detection circularizing processing on the image in the instrument edge by means of a computer through a Hough gradient method; e), establishing a square outside the excircle of a small dial by means of the computer; and f), acquiring a linear segment of the image in a square area by means of the computer through a Hough line transform algorithm. If a large pointer of a large dial is intersected with the excircle of a small dial, and distances between two intersection points (namely a point D and a point E) and a circle center O of a pointer seat of the small dial in the square area is 0.9-1.1 times of the radius of the excircle of the small dial, a line segment is deleted and finally one linear segment is obtained in the square area, wherein the line segment is the position of a small pointer in the image, thereby effectively preventing an error of mis-determining the large pointer as the small pointer.
Description
Technical field
The present invention relates to a kind of pointer position extracting method of many hand of dial formula instrument.
Background technology
Pointer instrument is to apply one of very universal measuring instrumentss in current production process, and not only species is many, and
Quantity is big.At present, the domestic calibration operation to this kind of instrument partly more adopts artificial observation instrumental panel total indicator reading, this tradition
The impact of other subjective factorss such as the operating experience of the examined personnel of method and the mental status, has high labor intensive, production
Efficiency is low, detection error is big, poor reliability the problems such as.In addition, under the relatively more severe working environment such as high temperature, radiation, field,
It is normally unsuitable for the reading of personal monitoring's pointer instrument, in this case, image recognition has obtained increasing application,
And key technology therein is exactly the extraction of gauge pointer position.Traditional Meter recognition is primarily directed to the instrument of single pointer
Carry out, some instrument dial plates are distributed with dial plate of different sizes and pointer（As dial gauge）But, many pointers are mutually intersected
The situation that fork overlaps, using traditional identifying schemes with regard to the little pointer in None- identified instrument, the position of little pointer cannot be extracted.
Content of the invention
The present invention is in order to overcome the shortcomings of above technology, there is provided in a kind of many hand of dial formula instrument, little pointer position can
With the accurate method extracted.
The present invention overcomes its technical problem be employed technical scheme comprise that：
A kind of pointer position extracting method of many hand of dial formula instrument, including：
A) the dial plate region of pointer instrument is carried out with original image collection, and the original image input computer by collection；
B) using computer, original image is carried out with gray processing process and original image is converted to gray level image by coloured image；
C) using computer, binary conversion treatment is carried out to gray level image, obtain the instrument in image using Canny edge detection algorithm
Table edge；
d）Using computer, by Hough gradient method, to instrument, intramarginal image carries out detecting circular process, detects in dial plate
The K circular center of circle and radius, a maximum circle of radius is defined as that big dial plate is cylindrical, and remaining each circle is orientated as little
Dial plate is cylindrical, and the center of circle of indicator seat of the little dial plate center of circle cylindrical with corresponding little dial plate is mutually concentric；
E) set up square outside little dial plate is cylindrical using computer, little dial plate cylindrical in this square, described pros
The length of side of shape is more than or equal to the outer diameter of a circle of little dial plate；
F) using computer, scaling method is become by Hough line and align Image Acquisition straightway in square region, will be less than little dial plate
The straightway of 2/3 radius removes, if a straightway has two intersection points with little dial plate is cylindrical, and two intersection points and this square region
The distance in the center of circle of indicator seat of the little dial plate in domain is 0.9-1.1 times of little dial plate exradius, then be judged to this straight line
The outline of Duan Wei great indicator, and remove, finally give the straight line section in square area.
For the definition reducing picture noise and improve image, the image in above-mentioned steps b) is converted into gray level image
Afterwards, also include by greyscale image transitions for Mat type Image image, computer carried out using OpenCV function pair Image image
The step of smooth and Edge contrast.
It is irregular indicator or the situation of wide indicator to be applied to little indicator, to finally giving after being additionally included in step f)
Straight line section in square area calculates the distance in the center of circle to the indicator seat of little dial plate for two end points, with little table
The center of circle of the indicator seat of disk is defined as the tip of pointer apart from big end point position, and computer calculates the table of little dial plate
The center of circle of the needle stand straightway that be connected to form most advanced and sophisticated with pointer.
The invention has the beneficial effects as follows：Utilize Hough gradient after the gradation of image process to collection, binary conversion treatment
Method determines the indicator seat center of circle and the straightway obtaining in little dial plate using Hough line change algorithm, if the big indicator of big dial plate
Cylindrical with little dial plate intersects, and two intersection points be D point with E point and this square area in the indicator seat of little dial plate the center of circle
The distance of O is 0.9-1.1 times of little dial plate exradius, then delete this straightway, after removing big indicator, permissible
Obtain final straight line section in square area, this straightway is the little indicator position in image, is therefore prevented effectively from
In traditional detection method, little dial plate is cylindrical is divided into two parts to lead to the mistake that big indicator is mistakenly identified as little indicator by big indicator
Occur.
Brief description
Fig. 1 is the little pointer position schematic diagram of the little dial plate of the present invention.
Specific embodiment
1 the present invention will be further described below in conjunction with the accompanying drawings.
A kind of pointer position extracting method of many hand of dial formula instrument, including：
A) the dial plate region of pointer instrument is carried out with original image collection, and the original image input computer by collection.b)
Using computer, original image is carried out with gray processing process and original image is converted to gray level image by coloured image, due to colour
Image is usually rgb format, and that is, each pixel is by R（Red）、G（Green）、B（Blue）Three component compositions, true for 24
Colour, each component greatly and is difficult to process it is therefore necessary to change coloured image with 1 byte representation, memory data output
For gray level image.Gray level image only comprises monochrome information, does not comprise color information, therefore reduces the data processing of computer
Amount.C) using computer, binary conversion treatment is carried out to gray level image, obtain the instrument in image using Canny edge detection algorithm
Table edge, greyscale image transitions is conducive to improving processing speed for bianry image, meets requirement of real-time.Bianry image refers to
Entire image only has the image of 0,255 two gray values, there are not other gray values between them.d）Led to using computer
Intramarginal image carries out detecting circular process to instrument to cross Hough gradient method, detect the K circular center of circle in dial plate with
Radius, it is cylindrical that a maximum circle of radius is defined as big dial plate, and remaining each circle orientates that little dial plate is cylindrical as, little dial plate
The center of circle O of the indicator seat center of circle cylindrical with corresponding little dial plate is mutually concentric.E) set up outside little dial plate is cylindrical using computer
Square, little dial plate cylindrical in this square, the described foursquare length of side is more than or equal to the outer diameter of a circle of little dial plate.f)
Using computer, scaling method is become by Hough line and align Image Acquisition straightway in square region, will be less than little dial plate 2/3 radius
Straightway remove, if a straightway is with little dial plate is cylindrical two intersection points, and in two intersection points and this square area
The distance of the center of circle O of the indicator seat of little dial plate is 0.9-1.1 times of little dial plate exradius, then be judged to that this straightway is big
The outline of indicator, and remove, finally give the straight line section in square area.Less than little dial plate 2/3 radius
Straight line may be considered the scale value on little dial plate, therefore removes it.If as shown in figure 1, the big indicator of big dial plate simultaneously
Cylindrical with little dial plate intersects, and two intersection points be D point with E point and this square area in the indicator seat of little dial plate the center of circle
The distance of O is 0.9-1.1 times of little dial plate exradius, then delete this straightway, after removing big indicator, permissible
Obtain final straight line section in square area, this straightway is the little indicator position in image, is therefore prevented effectively from
In traditional detection method, little dial plate is cylindrical is divided into two parts to lead to the mistake that big indicator is mistakenly identified as little indicator by big indicator
Occur.
After image in above-mentioned steps b) is converted into gray level image, also include by greyscale image transitions for Mat type Image
Image, computer carries out smooth and Edge contrast step using OpenCV function pair Image image.Because OpenCV is processed
Picture is Mat type matrix, therefore first image is entered row format conversion, and computer first carries out smothing filtering, Gu Mingsi to image
It is exactly that the part more than burr of a ripple or a certain frequency is removed on adopted signal, be exactly briefly low-pass filtering.It is reacted to
It is exactly noise reduction and image blurring process on image（Because high frequency has reacted details, so removing details to obtain fuzzy profile）,
Edge due to image is typically in HFS, so this smothing filtering will result in edge blurry, i.e. image and background
Will not make a clear distinction between good and evil but blurred transition.It is thus desirable to carrying out image sharpening again to fuzzy image, the edge of image is made to become
Clearly.The purpose that image sharpening is processed is the Jing Guoping in order that the details of the edge of image, contour line and image is apparent from
The basic reason that sliding image thickens is because that image receives average or integral operation, and therefore it can be carried out with inverse fortune
Calculating (such as differentiating) can be so that image be apparent from.Differentiate the rate of change being to seek signal, micro- by Fourier transform
Divide property to understand, differentiate and there is stronger high fdrequency components effect.To consider from frequency domain, image blurring essence is because it
High fdrequency components are attenuated, and therefore can make image clearly with high pass filter.By increasing the place to image smoothing and sharpening
Reason, further increases the definition in subsequent step and accuracy.
As shown in Figure 1, if in the case that little indicator is irregular indicator such as wide indicator, detect one of step f) is straight
Line segment is AB line segment or CB line segment in Fig. 1, and it is the outer contour of little indicator, therefore further, is additionally may included in step
Rapid f) after to calculate two end points to the straight line section finally giving in square area be the A point shown in Fig. 1 and B point
Or the distance of C point and the center of circle O to the indicator seat of little dial plate for the B point, with the center of circle O of the indicator seat of little dial plate apart from big end point
Position is defined as the tip of pointer, i.e. B point position in Fig. 1, computer calculate the center of circle O of the indicator seat of little dial plate with
The straightway being connected to form of pointer tip B point, OB straightway is the position of pointer in image, is therefore applied to and irregularly refers to
Pin, the position of wide pointer are extracted, and further increase the precision of pointer extracting.
Claims (3)
1. a kind of pointer position extracting method of many hand of dial formula instrument is it is characterised in that comprise the steps：
A) the dial plate region of pointer instrument is carried out with original image collection, and the original image input computer by collection；
B) using computer, original image is carried out with gray processing process and original image is converted to gray level image by coloured image；
C) using computer, binary conversion treatment is carried out to gray level image, obtain the instrument in image using Canny edge detection algorithm
Table edge；
d）Using computer, by Hough gradient method, to instrument, intramarginal image carries out detecting circular process, detects in dial plate
The K circular center of circle and radius, a maximum circle of radius is defined as that big dial plate is cylindrical, and remaining each circle is orientated as little
Dial plate is cylindrical, the center of circle of the indicator seat of little dial plate（O）The center of circle cylindrical with corresponding little dial plate is mutually concentric；
E) set up square outside little dial plate is cylindrical using computer, little dial plate cylindrical in this square, described pros
The length of side of shape is more than or equal to the outer diameter of a circle of little dial plate；
F) using computer, scaling method is become by Hough line and align Image Acquisition straightway in square region, will be less than little dial plate
The straightway of 2/3 radius removes, if a straightway has two intersection points with little dial plate is cylindrical, and two intersection points and this square region
The center of circle of the indicator seat of the little dial plate in domain（O）Distance be 0.9-1.1 times of little dial plate exradius, then be judged to that this is straight
Line segment is the outline of big indicator, and removes, and finally gives the straight line section in square area.
2. many hand of dial formula instrument according to claim 1 pointer position extracting method it is characterised in that：Described step
Rapid b) in image be converted into gray level image after, also include by greyscale image transitions for Mat type Image image, computer profit
Carry out smooth and Edge contrast step with OpenCV function pair Image image.
3. many hand of dial formula instrument according to claim 1 pointer position extracting method it is characterised in that：Also include
After step f), two end points is calculated to the indicator of little dial plate to the straight line section finally giving in square area
The center of circle of seat（O）Distance, the center of circle with the indicator seat of little dial plate（O）It is defined as pointer apart from big end point position
Tip, computer calculates the center of circle of the indicator seat of little dial plate（O）The most advanced and sophisticated straightway being connected to form with pointer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610832328.9A CN106446913A (en) | 2016-09-20 | 2016-09-20 | Pointer position extracting method of multi-dial pointer type instrument |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610832328.9A CN106446913A (en) | 2016-09-20 | 2016-09-20 | Pointer position extracting method of multi-dial pointer type instrument |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106446913A true CN106446913A (en) | 2017-02-22 |
Family
ID=58165680
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610832328.9A Pending CN106446913A (en) | 2016-09-20 | 2016-09-20 | Pointer position extracting method of multi-dial pointer type instrument |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106446913A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108519734A (en) * | 2018-03-26 | 2018-09-11 | 广东乐芯智能科技有限公司 | A kind of system of determining surface pointer position |
| WO2019184642A1 (en) * | 2018-03-26 | 2019-10-03 | 广东乐芯智能科技有限公司 | Method and system for determining positions of watch face hands |
| CN114332077A (en) * | 2022-03-02 | 2022-04-12 | 国网江西省电力有限公司电力科学研究院 | RANSAC-based light dial plate identification method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102521560A (en) * | 2011-11-14 | 2012-06-27 | 上海交通大学 | Instrument pointer image identification method of high-robustness rod |
| 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 |
-
2016
- 2016-09-20 CN CN201610832328.9A patent/CN106446913A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102521560A (en) * | 2011-11-14 | 2012-06-27 | 上海交通大学 | Instrument pointer image identification method of high-robustness rod |
| 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 |
Non-Patent Citations (4)
| Title |
|---|
| 于向阳: "机器视觉在仪表监控识别系统中的应用", 《中国优秀硕士学位论文全文数据库》 * |
| 倪刚: "指针式电力仪表图像嵌入式识别技术的研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
| 董国龄: "基于图像处理技术的水表自动检定及管理系统的设计", 《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》 * |
| 辛凤鸣: "指针式仪表自动检定算法研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108519734A (en) * | 2018-03-26 | 2018-09-11 | 广东乐芯智能科技有限公司 | A kind of system of determining surface pointer position |
| WO2019184642A1 (en) * | 2018-03-26 | 2019-10-03 | 广东乐芯智能科技有限公司 | Method and system for determining positions of watch face hands |
| CN114332077A (en) * | 2022-03-02 | 2022-04-12 | 国网江西省电力有限公司电力科学研究院 | RANSAC-based light dial plate identification method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106228170A (en) | A kind of pointer position extracting method of pointer instrument | |
| CN110580480B (en) | Meter reading identification method based on image processing | |
| CN102999886B (en) | Image Edge Detector and scale grating grid precision detection system | |
| CN115082418B (en) | Precise identification method for automobile parts | |
| CN110288561A (en) | Refractory brick surface scratch recognition methods based on frequency filtering enhancing | |
| CN106228161B (en) | A kind of pointer-type dial plate automatic reading method | |
| CN112837290B (en) | Crack image automatic identification method based on seed filling algorithm | |
| CN108960237B (en) | Reading identification method for pointer type oil level indicator | |
| CN106845545B (en) | Image recognition method and device in automatic digital electric meter calibration device | |
| CN107220645A (en) | Water meter recognition methods based on dynamic image pro cess | |
| CN106446913A (en) | Pointer position extracting method of multi-dial pointer type instrument | |
| CN107145890A (en) | A kind of pointer dashboard automatic reading method under remote various visual angles environment | |
| CN113298776B (en) | Method for detecting appearance defects of metal closed water pump impeller | |
| Maji et al. | An automated method for counting and characterizing red blood cells using mathematical morphology | |
| CN103226829A (en) | Image edge detection method based on edge enhancement operator | |
| CN115994907B (en) | Intelligent processing system and method for comprehensive information of food detection mechanism | |
| CN109886931A (en) | Gear ring of wheel speed sensor detection method of surface flaw based on BP neural network | |
| CN110866547B (en) | Automatic classification system and method for traditional Chinese medicine decoction pieces based on multiple features and random forests | |
| CN104835155A (en) | Fractal-based early-stage breast cancer calcification point computer auxiliary detection method | |
| CN107561736A (en) | LCD defect inspection methods based on Fourier and Hough transform | |
| CN114155226A (en) | Micro defect edge calculation method | |
| CN111815542B (en) | Tree annual ring image medulla positioning and annual ring measuring method | |
| CN106446903B (en) | A kind of instrument band switch location recognition method | |
| CN110146027B (en) | Method for measuring thickness of celadon glaze layer of SD-OCT image | |
| CN111815580A (en) | Image edge identification method and small module gear module detection method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170222 |