CN115471650A - Gas pressure instrument reading method, device, equipment and medium - Google Patents
Gas pressure instrument reading method, device, equipment and medium Download PDFInfo
- Publication number
- CN115471650A CN115471650A CN202211034647.7A CN202211034647A CN115471650A CN 115471650 A CN115471650 A CN 115471650A CN 202211034647 A CN202211034647 A CN 202211034647A CN 115471650 A CN115471650 A CN 115471650A
- Authority
- CN
- China
- Prior art keywords
- image
- area
- dial
- pointer
- 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
- 238000000034 method Methods 0.000 title claims abstract description 99
- 238000012937 correction Methods 0.000 claims abstract description 31
- 230000011218 segmentation Effects 0.000 claims abstract description 29
- 238000012545 processing Methods 0.000 claims description 36
- 238000004422 calculation algorithm Methods 0.000 claims description 20
- 230000015654 memory Effects 0.000 claims description 17
- 238000001514 detection method Methods 0.000 claims description 16
- 238000012216 screening Methods 0.000 claims description 15
- 238000003708 edge detection Methods 0.000 claims description 9
- 230000001965 increasing effect Effects 0.000 claims description 9
- 238000009825 accumulation Methods 0.000 claims description 8
- 238000003702 image correction Methods 0.000 claims description 5
- 238000003709 image segmentation Methods 0.000 claims description 5
- 230000008569 process Effects 0.000 description 29
- 230000009466 transformation Effects 0.000 description 18
- 238000007689 inspection Methods 0.000 description 17
- 238000000605 extraction Methods 0.000 description 13
- 229910018503 SF6 Inorganic materials 0.000 description 10
- 238000010586 diagram Methods 0.000 description 10
- SFZCNBIFKDRMGX-UHFFFAOYSA-N sulfur hexafluoride Chemical compound FS(F)(F)(F)(F)F SFZCNBIFKDRMGX-UHFFFAOYSA-N 0.000 description 10
- 230000009467 reduction Effects 0.000 description 9
- 230000008030 elimination Effects 0.000 description 7
- 238000003379 elimination reaction Methods 0.000 description 7
- 238000001914 filtration Methods 0.000 description 7
- 239000011159 matrix material Substances 0.000 description 7
- 238000006243 chemical reaction Methods 0.000 description 6
- 238000012549 training Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 239000013598 vector Substances 0.000 description 5
- 230000009471 action Effects 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000002372 labelling Methods 0.000 description 3
- 229960000909 sulfur hexafluoride Drugs 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 230000002708 enhancing effect Effects 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000002146 bilateral effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 239000012212 insulator Substances 0.000 description 1
- 230000000873 masking effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000007670 refining Methods 0.000 description 1
- 239000010979 ruby Substances 0.000 description 1
- 229910001750 ruby Inorganic materials 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
Images
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
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/98—Detection or correction of errors, e.g. by rescanning the pattern or by human intervention; Evaluation of the quality of the acquired patterns
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/19—Recognition using electronic means
-
- 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)
- Quality & Reliability (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a gas pressure meter reading method, a device, equipment and a medium, wherein the method comprises the following steps: acquiring a meter image of a meter to be detected, wherein the meter image comprises a plurality of dial areas with different radiuses; extracting characteristic points of an instrument image to be detected, and performing matching error correction on the characteristic points and the acquired template instrument image to obtain a corrected instrument image; dividing a dial area of the correction instrument image to obtain a divided instrument image; determining parameter information of each dial plate area based on the segmentation instrument image, and judging the dial plate areas belonging to concentric circles in each dial plate area based on the parameter information; generating a mask image based on a dial plate area belonging to a concentric circle in the dial plate area, and identifying a pointer and a scale area from the dial plate area belonging to the concentric circle based on the mask image; and reading identification is carried out based on the distance relation between the pointer and the scale area, and a reading result is obtained. According to the invention, through the scheme, the dial image is rapidly detected, the pointer and the scale mark area are segmented, and the meter reading is realized.
Description
Technical Field
The invention relates to the technical field of instrument image recognition in the power industry, in particular to a gas pressure instrument reading method, a gas pressure instrument reading device, gas pressure instrument reading equipment and a gas pressure instrument reading medium.
Background
Sulfur hexafluoride (SF 6) is widely used as a good gas insulator in gas insulation of electrical equipment in power stations. The environment in the booster station is complex, various instruments and meters are numerous, and the positions and the readings of the instrument pointers are difficult to observe and accurately identify by manual work every day. The intelligent instrument identification method based on image vision is accurate and efficient, and therefore a manual identification method is replaced.
At present, common intelligent instrument identification methods can be divided into two categories, namely a traditional image processing method and a deep learning method. The intelligent instrument identification method for traditional image processing mainly comprises the steps of preprocessing and enhancing features of an image, and then detecting an instrument area and a pointer by using a least square method or Hough transformation and other methods; the intelligent instrument identification method based on deep learning is to detect the instrument through a convolutional neural network and extract a pointer by combining a traditional image processing method. The traditional instrument identification method for image processing is characterized by manually setting characteristics, so that the robustness is poor; the method for recognizing the instrument based on deep learning is based on a large amount of image data to carry out learning training, the method needs to collect a large amount of data sets and manually mark the data sets, is labor-consuming and time-consuming, and simultaneously, the model training needs to be based on hardware equipment with higher performance, so that the application of the method is limited.
In addition, the SF6 instrument and other types of pointer instruments have obvious difference on the characteristics of dial images, a semi-annular C-shaped area exists in the middle of the SF6 instrument panel, and the annular C-shaped area, the pointer area and the scale area form a connected domain, which brings difficulty to the division of pointers and scale marks, and causes that other types of pointer instrument reading identification methods cannot be directly applied to the SF6 instrument images for reading.
Disclosure of Invention
In view of this, embodiments of the present invention provide a method, an apparatus, a device and a medium for reading a gas pressure meter, so as to solve the problem that the existing reading method is difficult to meet the requirement of automatic identification of reading of images of a sulfur hexafluoride pressure gauge in practical applications.
According to a first aspect, the present invention provides a gas pressure gauge reading method comprising: acquiring a meter image of a meter to be detected, wherein the meter image comprises a plurality of dial areas with different radiuses; extracting characteristic points of an instrument image to be detected, and performing matching error correction on the characteristic points and the acquired template instrument image to obtain a correction instrument image; carrying out dial plate area segmentation on the image of the correction instrument to obtain a segmented instrument image; determining parameter information of each dial plate area based on the segmentation instrument image, and judging the dial plate areas belonging to concentric circles in each dial plate area based on the parameter information; generating a mask image based on a dial plate area belonging to a concentric circle in the dial plate area, and identifying a pointer and a scale area from the dial plate area belonging to the concentric circle based on the mask image; and reading identification is carried out based on the distance relation between the pointer and the scale area, and a reading result is obtained.
According to the embodiment of the invention, the dial area of the correcting instrument image is segmented, the parameter information of the dial area is determined, and the pointer and the scale area in the dial area are identified based on the parameter information of the dial area, so that the reading result is obtained through the distance relationship between the pointer and the scale area, and the automatic reading of the pointer instrument in the power routing inspection is realized. In the process of automatic reading of the pointer instrument in the power inspection, the heavy data set labeling, model training and other processes are avoided, so that the inspection efficiency is improved; and due to the low implementation cost, the algorithm deployment and application are more convenient, and the universality of the method is improved.
With reference to the first aspect, in a first embodiment of the first aspect, the performing region segmentation on the calibration gauge image includes: and detecting the dial circle radius of the correction instrument image, generating a mask image of the dial area according to the dial circle radius, and performing AND operation on the mask image and the correction instrument image to obtain a segmentation instrument image.
Through the embodiment, the dial plate circle radius of the instrument image is detected and corrected, and the mask image is generated based on the dial plate circle radius, so that the segmented instrument image is acquired. In the process, the image of the dial area is obtained through the AND operation of the mask image and the correction instrument image, a data basis is provided for the subsequent identification of the pointer and the scale area in the dial area, and the realization of the automatic reading of the pointer instrument in the power inspection is further ensured.
With reference to the first aspect, in a second implementation manner of the first aspect, the parameter information includes a dynamic minimum circle radius, a dynamic maximum circle radius, a minimum distance between circle centers, and a circle center accumulation threshold, and the determining the parameter information of each dial area based on the split instrument image includes: fuzzifying the character information in each dial area to obtain a plurality of first dial area images; carrying out binarization processing on the images of the first dial areas to obtain image edge points of the first dial areas; and determining parameter information of each dial plate area based on the image edge point of each first dial plate area.
Through the embodiment, the segmentation instrument image is subjected to smooth noise reduction and gray level processing, so that the character information in each instrument area is blurred, and the edge characteristic information of the image is enhanced. And performing binarization processing on the first dial area images to obtain image edge points, thereby determining parameter information of each dial area and providing a data basis for subsequently judging the dial areas belonging to the concentric circles in each dial area. .
With reference to the second implementation manner of the first aspect, in a third implementation manner of the first aspect, the binarizing processing the image of each first dial area to obtain an image edge point of each first dial area includes: carrying out binarization processing on the images of the first dial areas through a self-adaptive threshold segmentation algorithm; traversing all connected regions of the first dial plate region images after binarization processing by using a contour searching method, and performing area screening to remove image isolated points to obtain a plurality of second dial plate region images; and detecting the image of each second dial area through a Canny edge detection algorithm to obtain the image edge point of each first dial area.
In the embodiment, the binarization processing is performed on the first dial area image, all connected areas of the first dial area image after the binarization processing are traversed by the contour searching method, and isolated points are removed, so that the factors irrelevant to the edge image are filtered, the edge characteristics are enhanced, the image edge points of the first dial area are obtained through the edge detection algorithm, and a data basis is provided for the subsequent automatic reading of the pointer instrument in the power routing inspection.
With reference to the second implementation manner of the first aspect or the third implementation manner of the first aspect, in a fourth implementation manner of the first aspect, the determining, based on the parameter information, a dial area belonging to a concentric circle in each dial area includes: based on the minimum circle radius and the maximum circle radius, respectively increasing progressively according to corresponding preset step lengths and detecting corresponding concentric circle and circle characteristic information; the circle feature information includes: circle center coordinates and circle radius information; and screening based on the image edge points, the circle center coordinates and the circle radius information, and eliminating circles with the circle radii smaller than a preset threshold value to obtain a dial plate area belonging to concentric circles.
Through the mode, the parameter information of the side keys is converted through Hough, the minimum circle radius and the maximum circle radius are increased progressively according to the corresponding preset step length respectively, the elimination of the circles which are detected repeatedly is realized by screening based on the edge points, the coordinates of the circle center and the circle radius information of the images, the unqualified circles are further eliminated aiming at the circle radii, and the dial plate area corresponding to the concentric circles in the dial plate images is detected through improved Hough conversion.
With reference to the fourth embodiment of the first aspect, in the fifth embodiment of the first aspect, the identifying the pointer and the scale area from the dial area belonging to the concentric circles based on the mask image includes: performing AND operation on the mask image and the first dial area to divide a pointer and a scale area; converting the pointer and the scale area into a rectangular area, and counting the number of pixels in the rectangular area; and identifying the positions of the nearest adjacent scale marks on the left and right of the pointer and the position of the pointer according to the number of the pixels.
Through the embodiment, the mask image is generated according to the corresponding radius of each concentric circle region in the dial plate region, the pointer and the scale region are divided through the mask image, the position of the nearest scale mark on the left and right of the pointer and the position of the pointer are identified through image conversion, and a data basis is provided for automatic reading of the pointer instrument in the follow-up power inspection.
With reference to the fifth implementation manner of the first aspect, in the sixth implementation manner of the first aspect, the reading identification based on the distance relationship between the pointer and the scale area includes: identifying the scale value through a character identification algorithm, and calculating the indicating value of the pointer through the following formula:
wherein R represents the index value of the pointer, v l And v r Respectively representing the values corresponding to the nearest neighbor scale marks at the left and right sides of the pointer line; d is a radical of l And d r Representing the distance of the nearest neighboring graduation lines at the left and right of the pointer from the pointer.
In the above embodiment, the reading recognition is realized by the values of the index value of the pointer and the values corresponding to the nearest neighboring scale marks on the left and right of the pointer line, and the distance between the nearest neighboring scale marks on the left and right of the pointer and the pointer.
According to a second aspect, embodiments of the present invention provide a gas pressure gauge reading device, comprising: the device comprises an image acquisition module, a data acquisition module and a data processing module, wherein the image acquisition module is used for acquiring an instrument image of a detection instrument to be detected, and the instrument image comprises a plurality of dial areas with different radiuses; the image correction module is used for extracting the characteristic points of the instrument image and carrying out error correction matching on the characteristic points and a preset template instrument image to obtain a corrected instrument image; the image segmentation module is used for carrying out region segmentation on the basis of the correction instrument image to obtain a segmentation instrument image; the concentric circle detection module is used for determining parameter information of each dial plate area based on the segmentation instrument image and judging the dial plate areas belonging to the concentric circles in each dial plate area based on the parameter information; the pointer and scale mark identification module generates a mask image based on the dial plate area belonging to the concentric circle in the dial plate area, and identifies the pointer and the scale area from the dial plate area belonging to the concentric circle based on the mask image; and the meter reading module is used for reading identification based on the distance relationship between the pointer and the scale area to obtain a reading result.
By the scheme, the problem of automatic reading of the pointer instrument in the power inspection is solved, and the inspection efficiency is improved; the heavy data set labeling, model training and other processes in the deep learning method are avoided, the universality of the method is improved, the implementation cost is low, and the algorithm deployment and application are more convenient.
According to a third aspect, there is provided an electronic device comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing therein computer instructions, and the processor executing the computer instructions to perform the method of reading a gas pressure meter according to the first aspect or any embodiment of the first aspect.
According to a fourth aspect, there is provided a computer readable storage medium having stored thereon computer instructions for causing a computer to perform the gas pressure gauge reading method of the first aspect or any of the embodiments of the first aspect.
It can be known that the electronic device or the computer-readable storage medium provided above are all used for executing the corresponding method provided above, and therefore, the beneficial effects achieved by the method can refer to the beneficial effects in the corresponding method, and are not described herein again.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a flowchart showing a specific example of a gas pressure meter reading method according to the present embodiment;
FIG. 2 is a diagram showing a detection result of a specific example of a gas pressure meter reading method in the present embodiment;
FIG. 3 is a diagram illustrating the results of a tilt correction of a meter image for a specific example of a gas pressure meter reading method in accordance with the present embodiment;
FIG. 4 is a schematic view of a segmented dial area image of a specific example of a gas pressure meter reading method in the present embodiment;
fig. 5 is a schematic diagram of a binarized dial area image of a specific example of a gas pressure meter reading method in the embodiment;
FIG. 6 is a schematic diagram of an edge image obtained by edge detection according to a specific example of the gas pressure meter reading method in the present embodiment;
FIG. 7 is a schematic view of concentric circles of a dial plate of a specific example of a gas pressure meter reading method in the present embodiment;
FIG. 8 is a schematic diagram of a pointer and a scale area of a specific example of a gas pressure meter reading method in the present embodiment;
FIG. 9 is a schematic diagram of an image of a specific example of a reading method of the gas pressure meter in the embodiment after being expanded into a rectangle through an annular scale area;
FIG. 10 is a schematic structural diagram of a gas pressure meter reading device in the present embodiment;
fig. 11 is a schematic structural diagram of the electronic device in this embodiment.
Detailed Description
The technical solutions of the present invention will be described clearly and completely with reference to the accompanying drawings, and it should be understood that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, belong to the protection scope of the present invention.
Some of the terms of art appearing in the present embodiment are described below:
SF6: sulfur hexafluoride.
A booster station: refers to an overall system that voltage-transforms the charge passing through.
ORB (organized Fast and Rotated Brief), can be used to quickly create feature vectors for key points in an image. First, a specific region is searched from the image, and the feature vectors can be used for identifying an object in the image as a key point. Wherein key points, such as corner points, for example, have the characteristic that the pixel values change sharply from light to dark. The ORB will then compute a corresponding feature vector for each keypoint. The feature vector created by the ORB algorithm contains only 1 and 0, called binary feature vector.
OpenCV is a cross-platform computer vision and machine learning software library issued based on apache2.0 licensing (open source), which can run on Linux, windows, android, and MacOS operating systems. The method is light and efficient, is composed of a series of C functions and a small number of C + + classes, provides interfaces of languages such as ython, ruby, MATLAB and the like, and realizes a plurality of general algorithms in the aspects of image processing and computer vision.
Hough transformation: the method is used for detecting curves which can be described by certain functional relation in shapes of straight lines, circles, parabolas, ellipses and the like in the images, and the method is successfully applied to many fields of image analysis, pattern recognition and the like. The principle is as follows: the method comprises the steps of transforming curves (including straight lines) in an image space into a parameter space, and determining description parameters of the curves by detecting extreme points in the parameter space, so as to extract regular curves in the image.
Masking: the processed image (either totally or partially occluded) is used with the selected image, graphic or object to control the area or process of image processing.
For the convenience of understanding of the embodiments of the present invention, the following description will be further explained with reference to specific embodiments, which are not to be construed as limiting the embodiments of the present invention.
The embodiment provides a gas pressure meter reading method, as shown in fig. 1, the method mainly includes:
s101, acquiring a meter image of a meter to be detected, wherein the meter image comprises a plurality of dial areas with different radiuses.
Specifically, acquiring the instrument image to be detected refers to collecting and storing image range information as the instrument image of the instrument to be detected according to the type of the instrument to be detected.
Illustratively, the instrument image includes a plurality of dial areas of different radii, which means that, as shown in fig. 2, in the booster station SF6 gas pressure instrument, a semi-annular C-shaped area exists in the middle of the dial, and the annular C-shaped area, the pointer area, and the scale area constitute a connected domain, and the annular C-shaped area, the pointer area, and the scale area can be understood as concentric circles around the center of the dial, that is, a plurality of dial areas of different radii included in the instrument image.
And S102, extracting the characteristic points of the instrument image to be detected, and performing matching error correction on the characteristic points and the acquired template instrument image to obtain a correction instrument image.
Specifically, the process of extracting the characteristic points of the instrument image to be detected and performing matching error correction with the acquired template instrument image to obtain the corrected instrument image comprises the following steps: ORB feature point extraction is carried out on an image of the instrument to be detected based on an ORB image detection method, and a feature extraction result is obtained; based on the feature extraction result, carrying out feature point mismatching and removing on the feature extraction result; and calculating a perspective transformation matrix based on the feature extraction result after the feature points are mistakenly matched and removed, and performing perspective transformation on the image to be detected based on the calculated perspective transformation matrix to obtain a corrected instrument image.
Specifically, the ORB-based image detection method extracts ORB feature points of an image of a meter to be detected, and obtaining a feature extraction result means that ORB key points are obtained in a FAST corner extraction manner, and a BRIEF (binary robust index element feature) descriptor is obtained by describing a surrounding image region based on the key points, so that the feature points are extracted. The image detection method based on ORB is a mature prior art for carrying out ORB feature point extraction on an image of a meter to be detected, and is not described again.
Specifically, based on the feature extraction result, the feature point mismatching and removing of the feature extraction result refers to matching the extracted feature point with the template instrument image and removing the feature point with the mismatching.
Exemplarily, matching the extracted feature points with the template instrument image, and eliminating the feature points with wrong matching refers to calculating a nearest neighbor point qi and a second nearest neighbor point qi +1 of pi, and considering that pi and qi are successfully matched in one direction when the ratio of the Euclidean distance between pi and qi and the Euclidean distance between pi and qi +1 is smaller than a preset ratio threshold. And similarly, calculating the nearest neighbor point and the next nearest neighbor point of qi in the instrument image to be detected, and if the corresponding feature point of qi which is successfully matched in a one-way mode is pi, determining that the feature point is correctly matched. Otherwise, the feature points are mismatched. Wherein pi represents a feature point in the ith instrument image to be detected, qi represents a feature point in the ith template instrument image, qi +1 represents a feature point in the (i + 1) th template instrument image, and i is a natural number.
By carrying out feature point mismatching elimination on the feature extraction result, the elimination of the feature points is realized, the matching point pairs with high matching quality are reserved, and a data basis is provided for subsequently obtaining the image of the correction instrument.
Specifically, a perspective transformation matrix is calculated based on the feature extraction result after the feature points are mistakenly matched and removed, and perspective transformation is performed on the image to be detected based on the calculated perspective transformation matrix to obtain a corrected instrument image. In practical application, based on the feature extraction result after the feature point is subjected to mismatching and eliminated, the perspective transformation matrix is calculated by using a GetPerspectiveTransform function to obtain an image perspective transformation matrix based on OpenCV. In practical application, the perspective transformation is performed on the image to be detected based on the calculated perspective transformation matrix, and the obtained corrected instrument image can realize the image perspective transformation by using a warPerspectral function based on OpenCV.
Illustratively, as shown in fig. 2, the process of detecting and correcting the inclination of the SF6 meter by ORB feature matching is exemplarily shown, and both ends of the connecting line in the figure can be understood as a matched point pair. As shown in fig. 3, the result shown in fig. 3 is a corrected instrument image obtained by extracting the feature points of the instrument image to be detected, and performing matching error correction on the extracted feature points and the acquired template instrument image.
And S103, carrying out dial area segmentation on the correction instrument image to obtain a segmented instrument image.
Specifically, the step of performing dial area division on the correction instrument image to obtain the divided instrument image is to obtain a divided dial area image through the mask image and the instrument image.
And S104, determining parameter information of each dial area based on the image of the division instrument, and judging the dial areas belonging to concentric circles in each dial area based on the parameter information.
Specifically, the parameter information of each dial area includes a dynamic minimum circle radius, a dynamic maximum circle radius, a circle center minimum distance, and a circle center accumulation threshold. The judgment of the dial areas belonging to the concentric circles in each dial area based on the parameter information means that the edge area of the communication area formed by the scale area, the pointer and the C-shaped semi-ring area is detected based on the parameter information of each dial area, so that the judgment of the dial areas belonging to the concentric circles in each dial area is realized.
Exemplarily, taking the booster station SF6 gas pressure instrument as an example, the concentric dial areas include an annular C-shaped area, a pointer area and a scale area, that is, the concentric dial areas refer to a plurality of dial areas with different radii contained in the instrument image.
And S105, generating a mask image based on the dial area belonging to the concentric circle in the dial area, and identifying the pointer and the scale area from the dial area belonging to the concentric circle based on the mask image.
Specifically, generating the mask image based on the dial area belonging to the concentric circles in the dial area means generating the mask image based on each radius of the concentric circles in the dial area belonging to the concentric circles.
Specifically, identifying the pointer and the scale area from the dial area belonging to the concentric circle based on the mask image means that the generated mask image and the dial area of the concentric circle are subjected to and operation to determine corresponding images of the pointer and the scale area; and carrying out coordinate transformation on the corresponding image based on the pointer and the corresponding image of the scale area, so as to realize the identification of the pointer and the scale area.
And S106, reading identification is carried out based on the distance relation between the pointer and the scale area, and a reading result is obtained.
Specifically, performing reading recognition based on the distance relationship between the pointer and the scale area means performing reading recognition based on the distance between the pointer and the nearest scale line and the value corresponding to the nearest scale line.
In an optional embodiment, the dial area of the correcting instrument image is segmented, the parameter information of the dial area is determined, and the pointer and the scale area in the dial area are identified based on the parameter information of the dial area, so that the reading result is obtained through the distance relationship between the pointer and the scale area, and the automatic reading of the pointer instrument in the power patrol is realized. In the process of automatic reading of the pointer instrument in the power inspection, the heavy data set labeling, model training and other processes are avoided, so that the inspection efficiency is improved; and due to the low implementation cost, the algorithm is more convenient to deploy and apply, and the universality of the method is improved.
In an optional embodiment, in step S103, the process of performing region segmentation on the calibration gauge image mainly includes: and detecting the dial circle radius of the correcting instrument image, generating a mask image of the dial area according to the dial circle radius, and performing AND operation on the mask image and the correcting instrument image to obtain a segmentation instrument image.
Specifically, the dial circle radius of the image of the detection rectification meter refers to the dial circle radius of the image of the detection rectification meter through Hough transformation. It should be understood that the detection of the radius of the circular image in the image by Hough transform belongs to a more mature technology, and is not described in detail herein.
Specifically, generating a mask image of a dial area according to the dial circle radius, and performing and operation on the mask image and the correction instrument image to obtain the division instrument image means that the mask image covering the dial area in the correction instrument image is formed according to the determined dial circle radius, so that the division instrument image is obtained through the and operation on the mask image and the correction instrument image. The division instrument image is an image including a dial area. As shown in fig. 4, the dial image is divided by removing the meter outline from the corrected meter image.
In the above embodiment, the dial plate circle radius of the instrument image is detected and corrected through Hough transformation, and the mask image is generated based on the dial plate circle radius, so that the acquisition of the instrument image is segmented. In the process, the image of the dial area is obtained through the AND operation of the mask image and the correction instrument image, a data basis is provided for the subsequent identification of the pointer and the scale area in the dial area, and the realization of the automatic reading of the pointer instrument in the power inspection is further ensured.
In an optional embodiment of the present invention, the parameter information includes a dynamic minimum circle radius, a dynamic maximum circle radius, a circle center minimum distance, and a circle center accumulation threshold. In step S104, the process of determining the parameter information of each dial area based on the split meter image mainly includes:
(1) And performing fuzzification processing on the character information in each dial area to obtain a plurality of first dial area images.
Specifically, the process of blurring character information in each dial area to obtain a plurality of first dial area images includes: based on the segmentation instrument image, performing smooth noise reduction on the segmentation instrument image to obtain a noise-reduced processing image; and graying the processed images based on the processed images to obtain a plurality of first dial area images.
Specifically, the smooth noise reduction of the divided instrument image means that scale characters and some dial isolated features in the dial image are eroded by the smooth noise reduction. The smooth noise reduction of the segmentation instrument image can be realized by edge-preserving filtering. Wherein, the edge-preserving filtering may be bilateral filtering or mean-shift filtering.
In practical application, a mean shift filtering method is usually used to implement smooth noise reduction on a segmented instrument image. Illustratively, smooth noise reduction of segmented instrument images by means of mean-shift filtering may be achieved by calling pyrMeanShiftFiltering function in OpenCV on the images.
Specifically, based on the processed image, graying the processed image refers to filtering color information by performing image graying through a weighted average value, and enhancing edge feature information of the image. The term "grayscaling the image by the weighted average value" means that the weighted average value is used to adjust the weighting coefficients of the R, G, and B channels of the processed image, thereby grayscaling the processed image. In practical applications, the weighting factor for the R channel is usually 0.299, the weighting factor for the g channel is 0.578, and the weighting factor for the b channel is 0.114. It should be understood that the setting of the weighting coefficients may be adjusted according to the actual operating conditions, and is not particularly limited.
In the embodiment, the segmentation instrument image is subjected to smooth noise reduction and graying processing to realize fuzzification processing on the character information in each dial area, so that the edge characteristic information of the image is enhanced, and a data basis is provided for subsequently judging the dial areas belonging to the concentric circles in each dial area.
(2) And carrying out binarization processing on the images of the first dial areas to obtain image edge points of the first dial areas.
Specifically, the binarization processing is performed on each first dial area image to obtain the image edge point of each first dial area, which means that the image edge point of each first dial area is obtained by performing area threshold calculation, isolated point removal and edge detection on each first dial area image.
(3) And determining the parameter information of each dial area based on the image edge points of each first dial area.
Specifically, determining the parameter information of each dial area based on the image edge points of each first dial area means detecting the image edge points of each first dial area through Hough transformation, and determining the parameter information of each dial area. Wherein the parameter information includes: the minimum circle radius, the maximum circle radius, the minimum distance of the circle center and the circle center accumulation threshold value.
In the above embodiment, the segmentation instrument image is subjected to smooth noise reduction and graying processing, so that the character information in each instrument area is blurred, and thus the edge feature information of the image is enhanced. And the edge points of the images are obtained by carrying out binarization processing on the images of the first dial plate areas, so that the parameter information of the dial plate areas is determined, and a data basis is provided for the subsequent judgment of the dial plate areas belonging to the concentric circles in the dial plate areas.
In an optional embodiment of the present invention, the process of performing binarization processing on the image of each first dial area in the above step to obtain an image edge point of each first dial area specifically includes:
(1) And carrying out binarization processing on the images of the first dial areas through a self-adaptive threshold segmentation algorithm.
Specifically, the binarization processing of the first dial area images through the adaptive threshold segmentation algorithm means that each local part of the picture is processed to obtain a threshold, the area is segmented by the threshold, and each area is processed by a different threshold, so that the segmentation effect of the image is more accurate, and the problem that acquired images have uneven brightness information due to brightness or lamplight and the like is solved. In practical application, the binarization processing of each first dial area image can be realized based on a threshold segmentation algorithm in OpenCV.
(2) Traversing all connected regions of the first dial area images after binarization processing by using a contour searching method, and carrying out area screening to remove image isolated points to obtain a plurality of second dial area images.
Specifically, traversing all connected regions of each first dial plate region image after binarization processing by a contour search method refers to detecting the connected regions of each first dial plate region image after binarization processing, wherein the connected regions include: each isolated point region, an annular C-shaped region, a pointer region and a scale region. In practical application, each isolated point region refers to a region corresponding to characters in a meter image of a meter to be detected.
Specifically, area screening is carried out to remove the image isolated points, and the plurality of second dial area images are obtained, namely the area occupied by the image isolated points is smaller than the area occupied by other connected areas, so that the image isolated points are screened and removed according to the area. In practical application, all connected regions of the first dial region images after traversing binarization processing can be realized based on a connected components within OpenCV, and area screening is performed to remove isolated points of the images. Illustratively, as shown in fig. 5, the dial image is obtained by removing isolated points based on each first dial area image.
(3) And detecting the images of the second dial areas by a Canny edge detection algorithm to obtain the image edge points of the first dial areas.
Specifically, detecting images of each second dial area through a Canny edge detection algorithm to obtain image edge points of each first dial area means calculating the amplitude and the direction of the gradient of the images of each second dial area; based on the amplitude and the direction of the gradient, the edge is restrained and refined through a non-maximum value, and a narrow boundary image is obtained; and judging edge points based on the narrow boundary image, the preset maximum gray gradient value and the preset minimum gray gradient value, and taking the edge points as image edge points of each first dial area. Wherein gradient directions are classified into four categories: vertical, horizontal, and two diagonal lines.
Specifically, based on the magnitude and direction of the gradient, the refining edge is suppressed by the non-maximum value to obtain the narrow-boundary image, which means that each pixel is checked to see whether the gradient of the point is the maximum of surrounding points with the same gradient direction, so as to remove the points on the non-boundary, and thus, the remaining points are taken as the narrow-boundary image.
Specifically, based on the narrow-boundary image, the preset maximum gray gradient value and the preset minimum gray gradient value, the edge point is determined to be a true edge point when the gray gradient of the point in the narrow-boundary image is higher than the preset maximum gray gradient value, and the boundary lower than the preset minimum gray gradient value is discarded. If so, it is checked whether the point is connected to an edge point that is determined to be true, and if so, it is considered to be an edge point. Illustratively, as shown in fig. 6, an edge image is obtained through Canny edge detection based on each second dial area image.
In the embodiment, the binaryzation processing is performed on the first dial plate area image, all the connected areas of the first dial plate area image subjected to the binaryzation processing are traversed through the contour searching method, isolated points are removed, factors irrelevant to the edge image are filtered, edge characteristics are enhanced, the edge point of the image of each first dial plate area is obtained through an edge detection algorithm, and a data basis is provided for automatic reading of a pointer instrument in follow-up power routing inspection.
In an optional embodiment, the parameter information includes a minimum circle radius, a maximum circle radius, a minimum distance between circle centers, and a circle center accumulation threshold, and in step S104, the process of determining, based on the parameter information, a dial area belonging to a concentric circle in each dial area includes:
(1) And based on the minimum circle radius and the maximum circle radius, respectively increasing the steps according to the corresponding preset step length and detecting the corresponding concentric circle and circle characteristic information. The circle feature information includes: circle center coordinates and circle radius information.
In practical application, the detected minimum circle radius is increased in sequence according to steps 5 and 6 or other data, and the detected maximum circle radius is increased once according to steps 10 and 12 or other values. When the maximum circle radius is increased to half of the dial image size, the increment of the minimum circle radius and the maximum circle radius is stopped, and corresponding concentric circles are detected from the minimum to the maximum in sequence according to the change of the radii.
Specifically, the determination method of the circle feature information includes: calculating the local gradient of the image edge point of each first dial area; accumulating in an accumulation plane along a straight line where the gradient of each edge point is located, and selecting the edge point of the local maximum value in the accumulation plane as a candidate circle center according to a preset threshold; for each candidate circle center having a point associated with the candidate circle center, calculating a distance between the candidate circle center and the point associated with the candidate circle center; and selecting an optimal value as the circle radius information based on the calculated distance.
In practical applications, the calculating of the local gradient of the image edge point of each first dial area may be implemented by a sobel operator (Sobeloperator). The preset threshold may be set according to an actual working condition, which is not specifically limited in this application.
(2) And screening based on the image edge points, the circle center coordinates and the circle radius information, and eliminating circles with the circle radii smaller than a preset threshold value to obtain a dial plate area belonging to concentric circles.
Specifically, based on the image edge point and the detected corresponding concentric circle and circle characteristic information, determining the relation between the circle center and the circle radius; and screening information based on the determined relation between the circle center and the circle radius, and rejecting circles with the circle radius smaller than a preset threshold value to obtain a dial area belonging to a concentric circle.
Specifically, determining the relationship between the center of a circle and the radius of the circle based on the image edge points and the detected corresponding concentric circles and circle feature information means constructing an equation of the circle in a cartesian coordinate system according to the coordinates of the center of the circle and the radius of the circle, and determining the relationship between the center of a circle and the radius of the circle for the image edge points according to the equation of the circle.
Specifically, the equation of a circle in a cartesian coordinate system can be expressed as follows:
(x-a) 2 +(y-b) 2 ＝r 2 ，
wherein x and y represent coordinates of points on a circle, coordinates of the center of the circle are (a, b), and the radius of the circle is r.
Specifically, the relationship between the center of a circle and the radius of the circle for the edge point of the image can be expressed as follows:
a＝x 0 -rcosθ，
b＝y 0 -rsinθ，
wherein the coordinates of the edge points on the circle are (x) 0 ,y 0 ) And θ represents a variable.
In practical applications, for each edge point (x) 0 ,y 0 ) All circles passing through the edge point can be drawn in a three-dimensional rectangular coordinate system, and a three-dimensional curve is finally obtained. If the curves obtained by two different edge points intersect in the space, the two edge points are represented on the same circle. And more curves meet at a point, meaning that the circle represented by the intersection is composed of more edge points. The manner of determining the number of curve intersection points may be implemented to determine that the candidate circle center has the number of dots forming with the candidate circle center in the above embodiments, thereby determining whether the candidate circle center has more than a preset number of dots forming a circle with the candidate circle center.
Specifically, the information screening based on the determined relationship between the circle center and the circle radius refers to screening a circle composed of the minimum circle radius, the maximum circle radius, and the circle feature information, which are respectively increased in size according to the corresponding step length, according to the determined relationship between the circle center and the circle radius, and screening a circle conforming to the relationship between the determined circle center and the circle radius, thereby eliminating a circle subjected to repeated detection.
Specifically, the elimination of the circle having the circle radius smaller than the preset threshold value to obtain the dial area belonging to the concentric circle refers to the elimination of the circle having the circle radius smaller than the preset threshold value to obtain the dial area belonging to the concentric circle based on the circle meeting the relationship between the circle center and the circle radius after screening, thereby achieving the elimination of the unsatisfactory circle. The preset threshold represents a threshold of a radius of the circle, and may be set according to an actual working condition, which is not specifically limited in the present application. Exemplarily, as shown in fig. 7, the detection result of the concentric circles of the dial is determined based on the edge points of the image, the coordinates of the center of the circle, and the radius of the circle.
Through the mode, the side key parameter information is converted through Hough, the minimum circle radius and the maximum circle radius are increased according to the corresponding preset step length respectively, repeated detection circle elimination is achieved through screening based on the image edge point, the circle center coordinate and the circle radius information, unsatisfactory circles are further eliminated according to the circle radius, and improved Hough conversion is achieved for detecting the dial plate area corresponding to the concentric circles in the dial plate image.
In an alternative embodiment, the step of generating the mask image based on the dial plate areas belonging to the concentric circles in the dial plate area means that the mask image is generated according to the corresponding radius of each concentric circle area in the dial plate area belonging to the concentric circles in the dial plate area. In practical applications, as shown in FIG. 7, the mask image is generated based on the radius r2 and the radius r 4. It can be seen that the mask image generated from the radius r2 and the radius r4 can divide the image representing the pointer and the scale area from the concentric circles of the dial.
In an optional embodiment, in step S105, the process of identifying the pointer and the scale area from the dial area belonging to the concentric circle based on the mask image specifically includes:
(1) And performing AND operation on the mask image and the first dial area to divide the pointer and the scale area.
Illustratively, as shown in fig. 8, it is the result of the pointer and scale area segmented from the meter image to be detected.
(2) And converting the pointer and the scale area into a rectangular area, and counting the number of pixels in the rectangular area.
Specifically, the conversion of the pointer and the scale area into the rectangular area refers to that the outer circumference of the pointer and the scale area is taken as the length of the converted rectangular area, the outer radius is taken as the height of the converted rectangular area, the inner radius subtracted from the outer radius is taken as the width of the converted rectangular area, the outermost circle of the pointer and the scale area corresponds to the first row in the converted rectangular area, and the arc width of the pointer and the scale area corresponds to the number of rows of the converted rectangular area. Therefore, a polar coordinate system is formed by taking the circle center as the center, and the real coordinates of the corresponding points in the converted rectangular area can be determined through coordinate conversion calculation according to the representation of the pointer and any point on the scale area through the polar coordinate. It should be understood that constructing a polar coordinate system and determining the real coordinates of the corresponding points in the transformed rectangular region through coordinate transformation calculation belongs to the mature prior art, and details thereof are not repeated. Illustratively, as shown in fig. 9, the image is expanded into a rectangle through the annular scale region.
(3) And identifying the positions of the nearest adjacent scale marks on the left and right of the pointer and the position of the pointer according to the number of the pixels.
In practical application, according to image characteristics, the maximum pixel point position of the rectangular area is the pointer position, and the left and right extreme point positions of the pointer position are the positions of the closest scales on the left and right of the pointer.
Through the embodiment, the mask image is generated according to the corresponding radius of each concentric circle region in the dial plate region, the pointer and the scale region are divided through the mask image, the position of the nearest scale mark on the left and right of the pointer and the position of the pointer are identified through image conversion, and a data basis is provided for automatic reading of the pointer instrument in the follow-up power inspection.
In an optional embodiment, in step S106, the process of performing reading identification based on the distance relationship between the pointer and the scale area specifically includes:
identifying the scale value through a character identification algorithm, and calculating the indicating value of the pointer through the following formula:
wherein R represents the index value of the pointer, v l And v r Respectively representing the values corresponding to the nearest neighbor scale marks at the left and right sides of the pointer line; d l And d r Representing the distance of the nearest neighboring graduation lines at the left and right of the pointer from the pointer.
Illustratively, as shown in FIG. 9, is an image after being expanded into a rectangle through an annular scale region, an exemplary show v l And v r The location to which it belongs.
In the above embodiment, the reading recognition is realized by the indication value of the pointer, the values corresponding to the nearest neighbor scale marks on the left and right of the pointer line, and the distance between the nearest neighbor scale marks on the left and right of the pointer and the pointer.
Fig. 10 is a schematic structural diagram of an electric quantity threshold dynamic adjustment device of a device according to an embodiment of the present invention, and includes: the device comprises an image acquisition module 11, an image correction module 12, an image segmentation module 13, a concentric circle detection module 14, a pointer and scale mark identification module 15 and a meter reading module 16.
The image acquisition module 11 is configured to acquire a meter image of a meter to be detected, where the meter image includes a plurality of dial areas with different radii. For a specific process, reference may be made to the related description about step S101 in the above embodiment, which is not described herein again.
And the image correction module 12 is configured to extract the feature points of the instrument image to be detected, and perform matching error correction on the feature points and the acquired template instrument image to obtain a corrected instrument image. For a specific process, reference may be made to the related description of step S102 in the foregoing embodiment, and details are not described herein.
And the image segmentation module 13 is configured to perform dial area segmentation on the correction instrument image to obtain a segmented instrument image. For a specific process, reference may be made to the related description about step S103 in the above embodiment, which is not described herein again.
And a concentric circle detection module 14 configured to determine parameter information of each dial area based on the split meter image, and determine dial areas belonging to concentric circles in each dial area based on the parameter information. For a specific process, reference may be made to the related description of step S104 in the foregoing embodiment, which is not described herein again.
And a pointer and scale mark identifying module 15 configured to generate a mask image based on a dial area belonging to the concentric circle in the dial area, and identify a pointer and a scale area from the dial area belonging to the concentric circle based on the mask image. For a specific process, reference may be made to the related description of step S105 in the above embodiments, and details are not repeated here.
And the meter reading module 16 is configured to perform reading identification based on the distance relationship between the pointer and the scale area, and obtain a reading result. For a specific process, reference may be made to the related description about step S106 in the above embodiment, which is not described herein again.
In an optional embodiment of the invention, through the combined action of the image acquisition module, the image correction module, the image segmentation module, the concentric circle detection module, the pointer and scale mark identification module and the instrument reading module, in the process of realizing automatic reading of a pointer instrument in electric power inspection by using the gas pressure instrument reading device, specifically, through carrying out dial area segmentation on an image of the correction instrument, parameter information of a dial area is determined, and a pointer and scale area in the dial area is identified based on the parameter information of the dial area, so that a reading result is obtained through the distance relationship between the pointer and the scale area, and the automatic reading of the pointer instrument in the electric power inspection is realized. In the process of automatic reading of the pointer instrument in the power inspection, the processes of heavy data set marking, model training and the like are avoided, so that the inspection efficiency is improved; and due to the low implementation cost, the algorithm deployment and application are more convenient, and the universality of the method is improved.
In one or more embodiments of the present invention, there is also provided an electronic device, as shown in fig. 11, fig. 11 is a schematic structural diagram of the electronic device in the embodiment of the present invention, and includes a processor 71 and a memory 72, where the memory stores computer readable instructions, and the computer readable instructions, when executed by the processor, cause the processor to execute the steps of the gas pressure meter reading method.
The processor 71 may be a Central Processing Unit (CPU), other general-purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware component, or the combination of the above chips. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory 72 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application program required by at least one function; the storage data area may store data created from use of the trace reconnecting means, and the like. Further, the memory may include high speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, the memory optionally includes memory remotely located from the processor, and these remote memories may be connected to the trace reconnecting device via a network.
In one or more embodiments of the invention, there is also provided a computer-readable storage medium having stored thereon computer instructions for causing the computer to perform the steps of the gas pressure meter reading method described above. The storage medium may be a magnetic disk, an optical disk, a Read-only memory (ROM), a Random Access Memory (RAM), a flash memory (FlashMemory), a hard disk (hard disk drive, abbreviated as HDD) or a Solid State Drive (SSD), etc.; the storage medium may also comprise a combination of memories of the kind described above.
It can be known that the computer device or the storage medium of the computer readable instructions provided by the present invention are all used for executing the corresponding method provided by the above, and therefore, the beneficial effects achieved by the present invention can refer to the beneficial effects in the corresponding method, which are not described herein again.
It is noted that, in this document, relational terms such as "first" and "second," and the like, are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrases "comprising a component of' 8230; \8230;" does not exclude the presence of additional identical elements in the process, method, article, or apparatus that comprises the element.
In the description of the present specification, reference to the description of the terms "this embodiment," "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
It should be understood that the above examples are only for clarity of illustration and are not intended to limit the embodiments. Other variations and modifications will be apparent to persons skilled in the art in light of the above description. And are neither required nor exhaustive of all embodiments. And obvious variations or modifications therefrom are within the scope of the invention.
Claims (10)
1. A method of gas pressure gauge reading, comprising:
acquiring a meter image of a meter to be detected, wherein the meter image comprises a plurality of dial areas with different radiuses;
extracting characteristic points of the instrument image to be detected, and performing matching error correction on the characteristic points and the acquired template instrument image to obtain a corrected instrument image;
carrying out dial area segmentation on the correction instrument image to obtain a segmented instrument image;
determining parameter information of each dial area based on the segmented instrument image, and judging dial areas belonging to concentric circles in each dial area based on the parameter information;
generating a mask image based on a dial plate area belonging to a concentric circle in the dial plate area, and identifying a pointer and a scale area from the dial plate area belonging to the concentric circle based on the mask image;
and reading identification is carried out based on the distance relation between the pointer and the scale area, and a reading result is obtained.
2. The gas pressure meter reading method of claim 1, wherein the area segmenting the prover image comprises:
and detecting the dial circle radius of the correction instrument image, generating a mask image of a dial area according to the dial circle radius, and performing AND operation on the mask image and the correction instrument image to obtain the segmentation instrument image.
3. The gas pressure meter reading method according to claim 1, wherein the determining parameter information for each of the dial areas based on the segmented meter image comprises:
fuzzifying the character information in each dial plate area to obtain a plurality of first dial plate area images;
carrying out binarization processing on the images of the first dial areas to obtain image edge points of the first dial areas;
and determining parameter information of each dial plate area based on the image edge point of each first dial plate area.
4. The gas pressure meter reading method according to claim 3, wherein the binarizing processing the image of each first dial plate area to obtain the image edge point of each first dial plate area comprises:
carrying out binarization processing on the images of the first dial areas through a self-adaptive threshold segmentation algorithm;
traversing all connected regions of each first dial plate region image subjected to binarization processing by using a contour searching method, and performing area screening to remove image isolated points to obtain a plurality of second dial plate region images;
and detecting the image of each second dial area through a Canny edge detection algorithm to obtain the image edge point of each first dial area.
5. The gas pressure meter reading method according to claim 3 or 4, wherein the parameter information includes a minimum circle radius, a maximum circle radius, a minimum distance to the center of the circle, and a center accumulation threshold, and the judging, based on the parameter information, each of the dial areas belonging to concentric circles includes:
based on the minimum circle radius and the maximum circle radius, respectively increasing progressively according to corresponding preset step lengths and detecting corresponding concentric circles and circle characteristic information; the circle feature information includes: circle center coordinates and circle radius information;
and screening based on the image edge points, the coordinates of the circle center and the circle radius information, and eliminating circles with the circle radius smaller than a preset threshold value to obtain the dial area belonging to the concentric circles.
6. The gas pressure meter reading method according to claim 5, wherein said identifying a pointer and a scale area from the dial area belonging to concentric circles based on the mask image comprises:
performing AND operation on the mask image and the first dial area to divide a pointer and a scale area;
converting the pointer and the scale area into a rectangular area, and counting the number of pixels in the rectangular area;
and identifying the positions of the nearest adjacent scale marks on the left and right of the pointer and the position of the pointer according to the number of the pixels.
7. The method of reading a gas pressure meter according to claim 6, wherein the identification of the reading based on the distance relationship between the pointer and the scale area comprises:
identifying the scale value through a character identification algorithm, and calculating the indicating value of the pointer through the following formula:
r represents the index value of the pointer, v l And v r Respectively representing the values corresponding to the nearest neighbor scale marks at the left and right sides of the pointer line; d is a radical of l And d r Representing the distance of the nearest neighboring graduation lines at the left and right of the pointer from the pointer.
8. A gas pressure gauge reading device, comprising:
the device comprises an image acquisition module, a display module and a control module, wherein the image acquisition module is configured to acquire a meter image of a meter to be detected, and the meter image comprises a plurality of dial areas with different radiuses;
the image correction module is configured to extract the characteristic points of the instrument image to be detected and perform matching error correction with the acquired template instrument image to obtain a corrected instrument image;
the image segmentation module is configured to perform dial area segmentation on the correction instrument image to obtain a segmentation instrument image;
a concentric circle detection module configured to determine parameter information of each of the dial plate areas based on the split meter image, and determine dial plate areas belonging to concentric circles among the dial plate areas based on the parameter information;
the pointer and scale mark identification module is configured to generate a mask image based on a dial plate area belonging to a concentric circle in the dial plate area, and identify a pointer and a scale area from the dial plate area belonging to the concentric circle based on the mask image;
and the meter reading module is configured to perform reading identification based on the distance relation between the pointer and the scale area to obtain a reading result.
9. An electronic device, comprising: a memory and a processor, the memory and the processor being communicatively coupled to each other, the memory having stored therein computer instructions, the processor executing the computer instructions to perform the gas pressure meter reading method of any of claims 1-7.
10. A computer-readable storage medium storing computer instructions for causing a computer to perform the gas pressure gauge reading method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211034647.7A CN115471650A (en) | 2022-08-26 | 2022-08-26 | Gas pressure instrument reading method, device, equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211034647.7A CN115471650A (en) | 2022-08-26 | 2022-08-26 | Gas pressure instrument reading method, device, equipment and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115471650A true CN115471650A (en) | 2022-12-13 |
Family
ID=84369057
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211034647.7A Pending CN115471650A (en) | 2022-08-26 | 2022-08-26 | Gas pressure instrument reading method, device, equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115471650A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115984282A (en) * | 2023-03-21 | 2023-04-18 | 菲特（天津）检测技术有限公司 | Spandex product detection method, device, equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104573702A (en) * | 2014-12-01 | 2015-04-29 | 长沙众治电气技术有限公司 | Method for automatically identifying sulfur hexafluoride pressure instrument image |
| CN109993154A (en) * | 2019-04-19 | 2019-07-09 | 山东大学 | The lithium sulfur type instrument intelligent identification Method of substation's simple pointer formula |
| CN111507186A (en) * | 2020-03-13 | 2020-08-07 | 郑州航空工业管理学院 | Substation pointer instrument reading identification method |
| CN114863409A (en) * | 2022-04-08 | 2022-08-05 | 昆明能讯科技有限责任公司 | Sulfur hexafluoride instrument automatic reading method based on substation inspection image |
| CN114898198A (en) * | 2022-05-19 | 2022-08-12 | 中国计量大学 | Image processing method for automatic reading of pointer type pressure gauge |
-
2022
- 2022-08-26 CN CN202211034647.7A patent/CN115471650A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104573702A (en) * | 2014-12-01 | 2015-04-29 | 长沙众治电气技术有限公司 | Method for automatically identifying sulfur hexafluoride pressure instrument image |
| CN109993154A (en) * | 2019-04-19 | 2019-07-09 | 山东大学 | The lithium sulfur type instrument intelligent identification Method of substation's simple pointer formula |
| CN111507186A (en) * | 2020-03-13 | 2020-08-07 | 郑州航空工业管理学院 | Substation pointer instrument reading identification method |
| CN114863409A (en) * | 2022-04-08 | 2022-08-05 | 昆明能讯科技有限责任公司 | Sulfur hexafluoride instrument automatic reading method based on substation inspection image |
| CN114898198A (en) * | 2022-05-19 | 2022-08-12 | 中国计量大学 | Image processing method for automatic reading of pointer type pressure gauge |
Non-Patent Citations (1)
| Title |
|---|
| 尚秋峰;刘薇;毛训;: "基于图像处理和光电传感技术的SF6气体在线监测系统设计", 电测与仪表, no. 04, 25 February 2017 (2017-02-25) * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115984282A (en) * | 2023-03-21 | 2023-04-18 | 菲特（天津）检测技术有限公司 | Spandex product detection method, device, equipment and storage medium |
| CN115984282B (en) * | 2023-03-21 | 2023-06-16 | 菲特（天津）检测技术有限公司 | Spandex product detection method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112949564B (en) | Pointer type instrument automatic reading method based on deep learning | |
| CN110807355B (en) | Pointer instrument detection and reading identification method based on mobile robot | |
| CN103759758B (en) | A kind of method for detecting position of the automobile meter pointer based on mechanical angle and scale identification | |
| CN105046252B (en) | A kind of RMB prefix code recognition methods | |
| CN114549981A (en) | Intelligent inspection pointer type instrument recognition and reading method based on deep learning | |
| CN111539330B (en) | Transformer substation digital display instrument identification method based on double-SVM multi-classifier | |
| CN114926839B (en) | Image identification method based on RPA and AI and electronic equipment | |
| CN111354047B (en) | Computer vision-based camera module positioning method and system | |
| CN109993154A (en) | The lithium sulfur type instrument intelligent identification Method of substation's simple pointer formula | |
| CN110415296B (en) | Method for positioning rectangular electric device under shadow illumination | |
| CN114022439B (en) | Flexible circuit board defect detection method based on morphological image processing | |
| CN111950559A (en) | Pointer instrument automatic reading method based on radial gray scale | |
| CN114005108A (en) | Pointer instrument degree identification method based on coordinate transformation | |
| CN108154496B (en) | Electric equipment appearance change identification method suitable for electric power robot | |
| CN110348307B (en) | Path edge identification method and system for crane metal structure climbing robot | |
| CN115797813A (en) | Water environment pollution detection method based on aerial image | |
| CN115471650A (en) | Gas pressure instrument reading method, device, equipment and medium | |
| CN113408519B (en) | Method and system for pointer instrument reading based on template rotation matching | |
| CN113610041A (en) | Reading identification method and device for pointer instrument | |
| CN112418226B (en) | Method and device for identifying opening and closing states of fisheyes | |
| CN112418210B (en) | Intelligent classification method for tower inspection information | |
| CN117392201A (en) | Target paper bullet hole identification and target reporting method based on visual detection | |
| CN114862761B (en) | Power transformer liquid level detection method, device, equipment and storage medium | |
| CN114926635B (en) | Target segmentation method in multi-focus image combined with deep learning method | |
| CN113159027B (en) | Seven-segment digital display instrument identification method based on minimum external rectangular variant |
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 |