CN114898347A - Machine vision identification method for pointer instrument - Google Patents
Machine vision identification method for pointer instrument Download PDFInfo
- Publication number
- CN114898347A CN114898347A CN202210416240.4A CN202210416240A CN114898347A CN 114898347 A CN114898347 A CN 114898347A CN 202210416240 A CN202210416240 A CN 202210416240A CN 114898347 A CN114898347 A CN 114898347A
- Authority
- CN
- China
- Prior art keywords
- image
- pointer
- instrument
- dial
- machine vision
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
-
- 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
- G06V10/20—Image preprocessing
- G06V10/32—Normalisation of the pattern dimensions
-
- 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
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/757—Matching configurations of points or features
-
- 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)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a machine vision identification method for a pointer instrument, which comprises the steps of collecting an instrument image, processing the instrument image, carrying out Hough circle detection, identifying and positioning a round dial plate image mask, carrying out adaptive threshold binarization, carrying out contour searching, extracting scale lines, fitting a circle center, positioning to obtain a dial plate center coordinate, identifying a straight line pointer and a vertex coordinate by using Hough line detection, matching local features of the instrument image, and identifying unique feature point coordinates on the instrument: and (4) sending out the meter reading by using an angle according to the obtained three coordinates of the dial center, the pointer vertex and the unique characteristic point. The invention has the beneficial effects that: the accuracy of the Hough circle and the recognition speed are improved, the influence of a complex background in an image on the recognition accuracy is eliminated, the problem that zero scales of an instrument caused by rotation in an image plane are not right above the image due to image distortion is solved, the influence on the recognition accuracy is solved, and the recognition speed is high by adopting a simple machine vision algorithm.
Description
Technical Field
The invention relates to the field of instruments, in particular to a pointer instrument machine vision identification method.
Background
With the continuous improvement of social productivity, the industrial production process is continuously leading to intellectualization, and production informatization and digitization are development trends. The instrument is an important component of industrial facilities such as transformer substations, oil fields, construction and construction, is an important safety guarantee for monitoring the production process, and is an important source for acquiring data in production informatization. The traditional pointer instrument is still widely applied to various industrial fields by virtue of excellent reliability, low cost and electromagnetic interference resistance. The method is different from a convenient data acquisition and recording mode of a digital instrument, most pointer instruments do not have data interfaces, autonomous data acquisition cannot be realized, manual meter reading methods are mainly used, and the method has higher professional requirements on meter reading personnel. This is not only time consuming and laborious, but also risks introducing human error. With the development of science and technology, especially the development of computer vision technology, the reading of the instrument readings by using a computer instead of a human is becoming a reliable means.
At present, most meter reading algorithms are based on instrument pictures obtained by a fixed instrument and a fixed camera, or an instrument panel is not inclined at a large angle. In many cases, the algorithm can only obtain good reading accuracy through complicated manual processing, which is not practical for many industrial fields. Moreover, in order to ensure the accuracy of the reading, it is often required that the background of the dashboard is not too cluttered, and the instrument must be correctly placed in the photograph without a large rotation. In addition, some instruments do not need real-time monitoring, and only an inspector needs to record instrument data regularly. If the fixed camera is installed, the cost is wasted. Images acquired by other means are often affected by the angle and position of the shot if the shot is not taken with a fixed camera. The resulting image rotation puts high demands on the reading algorithm. Especially the rotation of the dial, which in some places is in a special position or initially rotates when the dial is mounted, the meter in the captured image will also have a rotation angle. This will greatly affect the automatic reading of the pointer instrument.
In addition, some researchers use deep learning methods for pointer meter reading. However, deep learning requires extensive training of the data set to ensure the robustness of the model. If the instrument changes, retraining is likely to be required to achieve satisfactory results, and data has to be collected again, which is a tedious task.
Disclosure of Invention
In view of the problems in the prior art, the invention discloses a pointer instrument machine vision identification method, which comprises the following steps:
step 1: collecting an image containing a pointer instrument, and preprocessing the image;
step 2: carrying out Hough circle detection on the preprocessed instrument image, identifying and positioning a dial plate, and establishing a mask to extract a circular dial plate image;
and step 3: after certain processing is carried out on the extracted dial plate image, adaptive threshold binarization is carried out, contour searching is carried out on the binarized image, scale marks are extracted, and circle center positioning is carried out according to the scale marks in a fitting mode to obtain the center coordinates of the dial plate;
and 4, step 4: performing one-step processing on the binary image, removing scale lines and miscellaneous points, identifying a linear pointer by using Hough line detection, and further obtaining a pointer vertex coordinate;
and 5: identifying unique feature point coordinates on the instrument by local feature matching on the instrument image:
step 6: and (4) sending out the meter reading by using an angle according to the obtained three coordinates of the dial center, the pointer vertex and the unique characteristic point.
As a preferred embodiment of the present invention, the step 1 specifically comprises,
step 1-1: optimizing the image size, and uniformly setting the widths of the input pictures to be the same size;
step 1-2: performing mean filtering on the image to improve the overall smoothness of the image, and realizing the overall smoothness by using an OpenCV function cv2. pyrMeanShiftFilterning;
step 1-3: carrying out gray level and binarization processing to reduce the storage space occupied by the image and using the image as an input picture for Hough circle detection;
in a preferred embodiment of the present invention, the step 2 specifically comprises,
step 2-1: inputting a gray level image after mean value filtering, detecting possible circles by probability Hough circles, and automatically selecting one of the circles with the highest probability as a dial circle;
step 2-2: and (3) establishing a mask, setting the pixel outside the circle as 1, setting the pixel inside the circle as 0, covering the image, removing the background from the image, and only extracting the information in the instrument dial.
As a preferred scheme of the present invention, hough circle detection is implemented using OpenCV function cv2.houghcircles, and actual test parameters are as follows: dp is 1, param1 is 100, param2 is 20, minRadius is 80, maxRadius is 0.
In a preferred embodiment of the present invention, the step 3 specifically includes,
step 3-1: graying the obtained image by adopting Gaussian filtering and taking the grayed image as binarization input;
step 3-2: self-adaptive threshold binarization is realized by using an OpenCV function cv2.adaptive threshold, morphological framework characteristics of instruments in an image are extracted, a local threshold is determined by adopting a Gaussian filtering method, and binarization is finally performed according to the comparison relation between each pixel point and the local threshold. The binarization formula is expressed as follows:
setting: comparison ═ ThreshValue-C
Step 3-3: carrying out contour search on the generated binary image, identifying and positioning scale marks, and realizing by using an OpenCV function cv2.findContours, so that the contours of all white pixel sets can be obtained, wherein the contour is obtained according to the following characteristics of the scale marks: distance, length-width ratio of the central point of the scale mark near the range of the radius r, and area, and carrying out area statistics on the selected contour; wherein the scale mark is a slender area, and the length-width ratio is 1: and 4 or more rectangles, and secondary screening is carried out near the statistical median.
Step 3-4: thinning the obtained scale lines, and calculating an intersection point set obtained by pairwise intersection of extension lines of the scale lines, wherein a calculation formula of straight line intersection points is as follows:
and calculating the coordinates of the central point of the dial plate through the intersection point set.
In a preferred embodiment of the present invention, the step 4 specifically includes,
step 4-1: strengthening the generated binary image, identifying a pointer and a dial according to the contour extracted in the steps 3-1 to 3-4 by using the unique characteristic points of the dial with 50 scale readings, and obtaining a picture only containing the pointer and the disc;
step 4-2: using Hough line detection to identify the starting point and the ending point of the pointer according to an OpenCV function cv2. HoughLinesP;
step 4-3: and through distance judgment, points far away from the circle center are needed pointer vertex coordinates.
In a preferred embodiment of the present invention, the step 5 specifically includes,
step 5-1: the method comprises the steps of using a K-D tree local feature matching method, obtaining all matching points by a local feature matching algorithm when the adopted unique feature points of a dial plate are 50 scale readings, and using the Flann (Fast _ Library _ for _ Approximate _ Nearest _ Neighbors) of an OpenCV function cv2. FlanBasedMatcher and realizing the function of a Fast Nearest neighbor search Library;
step 5-2: and (4) calculating the average coordinate of all the characteristic point coordinates obtained by matching to serve as the unique characteristic point coordinate of the dial. The coordinate calculation formula is as follows:
in a preferred embodiment of the present invention, the step 6 specifically includes,
according to the center O of the dial c Pointer vertex P i And a characteristic point Z 50 Calculating the absolute coordinate value in the image to obtain a vector O c Z 50 And O c P i In the presence of O c The polar coordinate value is established clockwise for the origin X axis, and the conversion formula of the absolute coordinate and the polar coordinate is as follows:
wherein atan2 is an arctangent function that has taken quadrants into consideration;
converting the polar coordinate value into an included angle theta between the pointer and the zero scale;
the final instrument index calculation expression is as follows:
the invention has the beneficial effects that: instrument images shot by different devices are optimized into images with a uniform format, the accuracy of Hough circle detection can be improved by 200%, and the processing time is reduced to 1% of the original processing time; a mask extraction dial is established, so that the influence of a complex background in an image on the identification precision is eliminated; the center point of the dial is positioned through scale mark identification, so that the problem that the center of the dial in an actual photo is not positioned on the center of the Hough circle detection circle due to image distortion is solved; the unique characteristic points of the dial are identified by adopting a K-D tree local characteristic matching algorithm, so that the influence on the identification precision caused by the fact that the zero scale of the instrument is not right above the image due to rotation in the image plane is solved; the recognition speed is high by adopting a simple machine vision algorithm, and the problem of low recognition speed of a deep learning algorithm is solved.
Drawings
Fig. 1 is a technical route adopted by the present invention.
FIG. 2 is a diagram illustrating the image preprocessing process and the result thereof according to the present invention.
Fig. 3 is a schematic diagram of the dial plate recognition and extraction process and results of the present invention.
FIG. 4 is a diagram illustrating the adaptive threshold binarization process and result of the image according to the present invention.
Fig. 5 is a schematic diagram of the process and results of scale mark identification and dial plate center location according to the present invention.
FIG. 6 is a diagram illustrating the process and results of pointer identification and vertex positioning according to the present invention.
FIG. 7 is a diagram illustrating a local feature point matching process and result according to the present invention.
FIG. 8 is a schematic view of the angle meter reading process of the present invention.
Detailed Description
Example 1
The invention discloses a pointer instrument machine vision identification method, which comprises the following steps:
step 1: collecting an image containing a pointer instrument, and preprocessing the image;
step 2: carrying out Hough circle detection on the preprocessed instrument image, identifying and positioning a dial plate, and establishing a mask to extract a circular dial plate image;
and step 3: after certain processing is carried out on the extracted dial plate image, adaptive threshold binarization is carried out, contour searching is carried out on the binarized image, scale marks are extracted, and circle center positioning is carried out according to the scale marks in a fitting mode to obtain the center coordinates of the dial plate;
and 4, step 4: performing one-step processing on the binary image, removing scale lines and miscellaneous points, identifying a linear pointer by using Hough linear detection, and further obtaining a pointer vertex coordinate;
and 5: identifying unique feature point coordinates on the instrument by local feature matching on the instrument image:
step 6: and (4) sending out the meter reading by using an angle according to the obtained three coordinates of the dial center, the pointer vertex and the unique characteristic point.
As a preferred embodiment of the present invention, the step 1 specifically comprises,
step 1-1: optimizing the image size, and uniformly setting the widths of the input pictures to be the same size;
step 1-2: performing mean filtering on the image to improve the overall smoothness of the image, and realizing the overall smoothness by using an OpenCV function cv2. pyrMeanShiftFilterning;
step 1-3: carrying out gray level and binarization processing to reduce the storage space occupied by the image, and using the image as an input picture for Hough circle detection;
instrument images shot by different devices are optimized into images with a uniform format, the accuracy of Hough circle detection can be improved by 200%, and the processing time is reduced to 1% of the original processing time.
FIG. 2 is a graph showing the results after the treatment in steps 1-1 to 1-3.
In a preferred embodiment of the present invention, the step 2 specifically comprises,
step 2-1: inputting a gray level image after mean value filtering, detecting possible circles by probability Hough circles, and automatically selecting one of the circles with the highest probability as a dial circle;
step 2-2: and (3) establishing a mask, setting the pixel outside the circle as 1, setting the pixel inside the circle as 0, covering the image, removing the background from the image, and only extracting the information in the instrument dial.
FIG. 3 is a graph showing the results after the treatment in steps 2-1 to 2-2.
As a preferred scheme of the present invention, hough circle detection is implemented using OpenCV function cv2.houghcircles, and actual test parameters are as follows: dp is 1, param1 is 100, param2 is 20, minRadius is 80, maxRadius is 0.
In a preferred embodiment of the present invention, the step 3 specifically includes,
step 3-1: graying the obtained image by adopting Gaussian filtering and taking the grayed image as binarization input;
step 3-2: self-adaptive threshold binarization is realized by using an OpenCV function cv2.adaptive threshold, morphological framework characteristics of instruments in an image are extracted, a local threshold is determined by adopting a Gaussian filtering method, and binarization is finally performed according to the comparison relation between each pixel point and the local threshold. The binarization formula is expressed as follows:
setting: comparison ═ ThreshValue-C
Step 3-3: carrying out contour search on the generated binary image, identifying and positioning scale marks, and realizing by using an OpenCV function cv2.findContours, so that the contours of all white pixel sets can be obtained, wherein the contour is obtained according to the following characteristics of the scale marks: distance, length-width ratio and area of the central point of the scale mark near the radius r range, and carrying out area statistics on the selected contour; wherein the scale mark is a slender area, and the length-width ratio is 1: and 4. selecting the rectangle with the value above the statistical median value for secondary screening.
Step 3-4: thinning the obtained scale lines, and calculating an intersection point set obtained by intersecting extension lines of the scale lines pairwise, wherein a calculation formula of the straight line intersection point is as follows:
and calculating the coordinates of the central point of the dial plate through the intersection point set.
FIG. 4 is a graph showing the results after the treatment in steps 3-1 to 3-2;
FIG. 5 is a graph showing the results after the treatment in steps 3-3 to 3-4.
In a preferred embodiment of the present invention, the step 4 specifically comprises,
step 4-1: strengthening the generated binary image, identifying a pointer and a dial according to the contour extracted in the steps 3-1 to 3-4 by using the unique characteristic points of the dial with 50 scale readings, and obtaining a picture only containing the pointer and the disc;
step 4-2: using Hough line detection to identify the starting point and the ending point of the pointer according to an OpenCV function cv2. HoughLinesP;
step 4-3: and through distance judgment, points far away from the circle center are required pointer vertex coordinates. Through scale mark identification, the central point of the dial plate is positioned, and the problem that the center of the dial plate in an actual photo is not positioned on the detection circle center of the Hough circle due to image distortion is solved.
FIG. 6 is a graph showing the results after the treatment in steps 4-1 to 4-3.
In a preferred embodiment of the present invention, the step 5 specifically includes,
step 5-1: the method comprises the steps of using a K-D tree local feature matching method, obtaining all matching points by a local feature matching algorithm when the adopted unique feature points of a dial plate are 50 scale readings, and using the Flann (Fast _ Library _ for _ Approximate _ Nearest _ Neighbors) of an OpenCV function cv2. FlanBasedMatcher and realizing the function of a Fast Nearest neighbor search Library;
step 5-2: and (4) calculating the average coordinate of all the characteristic point coordinates obtained by matching to serve as the unique characteristic point coordinate of the dial. The coordinate calculation formula is as follows:
the unique characteristic points of the dial are identified by adopting a K-D tree local characteristic matching algorithm, and the influence on identification precision caused by the fact that the zero scale of the instrument is not right above the image due to rotation in the image plane is solved.
In a preferred embodiment of the present invention, the step 6 specifically comprises,
according to the center O of the dial c Pointer vertex P i And a characteristic point Z 50 Calculating the absolute coordinate value in the image to obtain a vector O c Z 50 And O c P i At the presence of O c The polar coordinate value is established clockwise for the origin X axial direction, and the conversion formula of the absolute coordinate and the polar coordinate is as follows:
wherein atan2 is an arctangent function that has taken quadrants into consideration;
converting the polar coordinate value into an included angle theta between the pointer and the zero scale;
the final instrument index calculation expression is as follows:
the recognition speed is high by adopting a simple machine vision algorithm, and the problem of low recognition speed of a deep learning algorithm is solved.
Electrical connections or structures not described in detail herein are prior art.
In the description of the present invention, it is to be understood that the terms "center", "longitudinal", "lateral", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate orientations or positional relationships based on those shown in the drawings, and are used only for convenience in describing and simplifying the description, but do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore are not to be construed as limiting the invention, and further, the terms "first", "second", etc., are used only for descriptive purposes and are not intended to indicate or imply relative importance or to implicitly indicate the number of technical features indicated, whereby the features defined as "first", "second", etc., may explicitly or implicitly include one or more of such features, in the description of the present invention, "a plurality" means two or more unless otherwise specified.
In the description of the present invention, it should be noted that, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, e.g., as meaning either a fixed connection, a removable connection, or an integral connection; can be mechanically or electrically connected; the two components can be directly connected or indirectly connected through an intermediate medium, and the two components can be communicated with each other, so that the specific meaning of the above terms in the present invention can be understood by those skilled in the art through specific situations
Although the present invention has been described in detail with reference to the specific embodiments thereof, the present invention is not limited to the above embodiments, and various changes and modifications can be made without departing from the spirit of the present invention within the knowledge of those skilled in the art without departing from the scope of the present invention.
Claims (8)
1. A pointer instrument machine vision identification method is characterized by comprising the following steps:
step 1: collecting an image containing a pointer instrument, and preprocessing the image;
step 2: carrying out Hough circle detection on the preprocessed instrument image, identifying and positioning a dial plate, and establishing a mask to extract a circular dial plate image;
and step 3: after certain processing is carried out on the extracted dial plate image, adaptive threshold binarization is carried out, contour searching is carried out on the binarized image, scale marks are extracted, and circle center positioning is carried out according to the scale marks in a fitting mode to obtain the center coordinates of the dial plate;
and 4, step 4: performing one-step processing on the binary image, removing scale lines and miscellaneous points, identifying a linear pointer by using Hough line detection, and further obtaining a pointer vertex coordinate;
and 5: identifying unique feature point coordinates on the instrument by local feature matching on the instrument image:
step 6: and (4) sending out the meter reading by using an angle according to the obtained three coordinates of the dial center, the pointer vertex and the unique characteristic point.
2. The machine vision identification method of the pointer instrument as claimed in claim 1, wherein: the step 1 specifically comprises the following steps of,
step 1-1: optimizing the image size, and uniformly setting the widths of the input pictures to be the same size;
step 1-2: performing mean filtering on the image to improve the overall smoothness of the image, and realizing the overall smoothness by using an OpenCV function cv2. pyrMeanShiftFilterning;
step 1-3: carrying out gray level and binarization processing to reduce the storage space occupied by the image, and using the image as an input picture for Hough circle detection;
3. the machine vision identification method of the pointer instrument as claimed in claim 1, wherein: the step 2 specifically includes the steps of,
step 2-1: inputting a gray level image after mean value filtering, detecting possible circles by probability Hough circles, and automatically selecting one of the circles with the highest probability as a dial circle;
step 2-2: and (3) establishing a mask, setting the pixel outside the circle as 1, setting the pixel inside the circle as 0, covering the image, removing the background from the image, and only extracting the information in the instrument dial.
4. The machine vision identification method of the pointer instrument as claimed in claim 3, wherein: hough circle detection was performed using OpenCV functions cv2. houghcirles, with the actual experimental parameters as follows: dp is 1, param1 is 100, param2 is 20, minRadius is 80, maxRadius is 0.
5. The machine vision identification method of pointer meters as claimed in claim 1, characterized in that the step 3 specifically includes,
step 3-1: graying the obtained image by adopting Gaussian filtering and taking the grayed image as binarization input;
step 3-2: self-adaptive threshold binarization is realized by using an OpenCV function cv2.adaptive threshold, morphological framework characteristics of instruments in an image are extracted, a local threshold is determined by adopting a Gaussian filtering method, and binarization is finally performed according to the comparison relation between each pixel point and the local threshold. The binarization formula is expressed as follows:
setting: comparison ═ ThreshValue-C
Step 3-3: carrying out contour search on the generated binary image, identifying and positioning scale marks, and realizing by using an OpenCV function cv2.findContours, so that the contours of all white pixel sets can be obtained, wherein the contour is obtained according to the following characteristics of the scale marks: distance, length-width ratio and area of the central point of the scale mark near the radius r range, and carrying out area statistics on the selected contour; wherein the scale mark is a slender area, and the length-width ratio is 1: and 4 or more rectangles, and secondary screening is carried out near the statistical median.
Step 3-4: thinning the obtained scale lines, and calculating an intersection point set obtained by pairwise intersection of extension lines of the scale lines, wherein a calculation formula of straight line intersection points is as follows:
and calculating the coordinates of the central point of the dial plate through the intersection point set.
6. The machine vision identification method of the pointer instrument as claimed in claims 1 and 5, wherein: the step 4 specifically includes the steps of,
step 4-1: reinforcing the generated binary image, identifying a pointer and a dial according to the contour extracted in the steps 3-1 to 3-4 by using the unique characteristic points of the dial with 50 scale readings, and obtaining a picture only containing the pointer and the dial;
step 4-2: using Hough line detection to identify the starting point and the ending point of the pointer according to an OpenCV function cv2. HoughLinesP;
step 4-3: and through distance judgment, points far away from the circle center are required pointer vertex coordinates.
7. The machine vision identification method of the pointer instrument as claimed in claim 1, wherein: the step 5 specifically includes the steps of,
step 5-1: the method comprises the steps of using a K-D tree local feature matching method, obtaining all matching points by a local feature matching algorithm when the adopted unique feature points of a dial plate are 50 scale readings, and using the Flann (Fast _ Library _ for _ Approximate _ Nearest _ Neighbors) of an OpenCV function cv2. FlanBasedMatcher and realizing the function of a Fast Nearest neighbor search Library;
step 5-2: and (4) calculating the average coordinate of all the characteristic point coordinates obtained by matching to serve as the unique characteristic point coordinate of the dial. The coordinate calculation formula is as follows:
8. the machine vision identification method of the pointer instrument as claimed in claim 1, wherein: the step 6 specifically includes the steps of,
according to the center O of the dial c Pointer vertex P i And a characteristic point Z 50 Calculating the absolute coordinate value in the image to obtain a vector O c Z 50 And O c P i In the presence of O c The polar coordinate value is established clockwise for the origin X axis, and the conversion formula of the absolute coordinate and the polar coordinate is as follows:
wherein atan2 is an arctangent function that has taken quadrants into consideration;
converting the polar coordinate value into an included angle theta between the pointer and the zero scale;
the final instrument index calculation expression is as follows:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210416240.4A CN114898347A (en) | 2022-04-20 | 2022-04-20 | Machine vision identification method for pointer instrument |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210416240.4A CN114898347A (en) | 2022-04-20 | 2022-04-20 | Machine vision identification method for pointer instrument |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114898347A true CN114898347A (en) | 2022-08-12 |
Family
ID=82718467
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210416240.4A Pending CN114898347A (en) | 2022-04-20 | 2022-04-20 | Machine vision identification method for pointer instrument |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114898347A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116704510A (en) * | 2023-08-08 | 2023-09-05 | 深圳中集智能科技有限公司 | Intelligent identification method for machine vision industrial instrument data |
-
2022
- 2022-04-20 CN CN202210416240.4A patent/CN114898347A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116704510A (en) * | 2023-08-08 | 2023-09-05 | 深圳中集智能科技有限公司 | Intelligent identification method for machine vision industrial instrument data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111626190B (en) | Water level monitoring method for scale recognition based on clustering partition | |
| CN112906694B (en) | Reading correction system and method for transformer substation inclined pointer instrument image | |
| CN112699876B (en) | Automatic reading method for various meters of gas collecting station | |
| WO2022057103A1 (en) | Deep-learning-based method for automatically reading pointer instrument | |
| CN112818988A (en) | Automatic reading identification method and system for pointer instrument | |
| CN102704215B (en) | Automatic cutting method of embroidery cloth based on combination of DST file parsing and machine vision | |
| CN106529559A (en) | Pointer-type circular multi-dashboard real-time reading identification method | |
| CN108764234B (en) | Liquid level meter reading identification method based on inspection robot | |
| CN112257676A (en) | Pointer instrument reading method and system and inspection robot | |
| CN111368906B (en) | Pointer type oil level meter reading identification method based on deep learning | |
| CN111539330B (en) | Transformer substation digital display instrument identification method based on double-SVM multi-classifier | |
| CN111814919A (en) | Instrument positioning and identifying system based on deep learning | |
| CN114549835A (en) | Pointer instrument correction identification method and device based on deep learning | |
| CN110659637A (en) | Electric energy meter number and label automatic identification method combining deep neural network and SIFT features | |
| CN115841669A (en) | Pointer instrument detection and reading identification method based on deep learning technology | |
| CN108960236B (en) | Pointer instrument identification method based on two-dimensional code matching and image identification | |
| CN115497109A (en) | Character and image preprocessing method based on intelligent translation | |
| CN113705564B (en) | Pointer type instrument identification reading method | |
| CN115546795A (en) | Automatic reading method of circular pointer instrument based on deep learning | |
| CN115457556A (en) | Reading method for disc pointer type instrument of nuclear power plant | |
| CN114898347A (en) | Machine vision identification method for pointer instrument | |
| CN112036391B (en) | Electronic inspection method and system based on shale gas field production station | |
| CN109784257B (en) | Transformer thermometer detection and identification method | |
| CN113191351B (en) | Reading identification method and device of digital electric meter and model training method and device | |
| CN109858474B (en) | Detection and identification method for transformer oil surface temperature controller |
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 |