CN111160337B - Automatic identification method, system, medium and equipment for reading of pointer instrument - Google Patents

Abstract

The invention relates to an automatic identification method, a system, a medium and equipment for reading of a pointer instrument, wherein the method comprises the following steps: preprocessing the acquired image of the target area containing the dial plate; extracting a dial plate image from the preprocessed image of the target area; determining a pointer position, a minimum scale line position and a maximum scale line position from the dial image; and calculating the reading of the pointer instrument according to the pointer position, the minimum scale line position and the maximum scale line position. According to the invention, the reading identification of the pointer instrument is automatically completed by carrying out image processing on the acquired image of the target area containing the dial plate, so that the manual participation in the automatic identification process of the instrument can be reduced, the inspection robot is helped to smoothly complete the inspection task, and the method is suitable for the identification of various pointer instruments and has good practicability and expansibility.

Description

A kind of automatic identification method, system, medium and equipment of pointer meter reading

technical field

The invention relates to the field of image pattern recognition, in particular to an automatic recognition method, system, medium and device for reading a pointer type meter.

Background technique

Due to its simple structure, low price, reliable performance, convenient maintenance, and strong anti-magnetic and anti-interference ability, pointer-type instruments are widely used in smart substations. In the intelligent substation, the inspection robot is used to inspect the pointer instrument. The main tasks performed by the inspection robot mainly include: walking according to the path planned based on the substation environment, reaching the inspection point, and the pan/tilt drives the camera to align the object to be detected. The existing inspection robots cannot automatically identify the readings of the meters, and the image of the meters needs to be transmitted to the operation and maintenance personnel, so that the operation and maintenance personnel can manually read the meters through the image of the meters, and the degree of automation is low.

SUMMARY OF THE INVENTION

In view of the above technical problems, the present invention provides an automatic identification method, system, medium and device for reading a pointer meter.

The technical solution of the present invention to solve the above-mentioned technical problems is as follows: an automatic identification method for the reading of a pointer meter, comprising:

Preprocess the acquired image of the target area containing the dial;

Extract the dial image from the preprocessed image of the target area;

determine the position of the pointer, the position of the minimum tick mark and the position of the maximum tick mark from the dial image;

The reading of the pointer meter is calculated based on the pointer position, the minimum tick mark position and the maximum tick mark position.

The beneficial effects of the present invention are: by performing image processing on the acquired image of the target area including the dial, the reading recognition of the pointer meter can be automatically completed, the manual participation in the automatic recognition process of the meter can be reduced, and the inspection robot can be smoothly Complete the inspection task, suitable for the identification of various pointer instruments, with good practicability and expansibility.

On the basis of the above technical solutions, the present invention can also be improved as follows.

Further, extracting the dial image from the preprocessed image of the target area specifically includes:

The SURF algorithm is used to register the image of the preprocessed target area and the front face image of the dial, and the perspective transformation matrix is solved, and the image of the preprocessed target area is subjected to affine transformation according to the perspective transformation matrix;

The dial image is extracted from the affine-transformed image of the target area using the Hough circle transform.

The beneficial effect of adopting the above-mentioned further solution is that the stretched image due to the observation angle can be corrected into a frontal image through affine transformation, thereby improving the reading accuracy of the pointer meter.

Further, determining the position of the pointer from the dial image specifically includes:

Searching for each annular area in the dial image along the radial direction, calculating the sum of non-zero pixels of each annular area, and using the annular area corresponding to the non-zero pixel and the minimum value as the target annular area;

Eight neighborhood contours are used to track each non-zero pixel point in the target annular region to determine the position of the pointer.

Further, determine the position of the minimum tick mark and the position of the maximum tick mark from the dial image, specifically including:

The scale line area in the dial image is extracted according to the HSV feature of the dial image, and the minimum scale line position and the maximum scale line position are determined according to the scale line area.

Further, the reading of the pointer meter is calculated according to the position of the pointer, the position of the minimum tick mark and the position of the maximum tick mark, specifically including:

Multiply the ratio of the angle between the pointer and the minimum scale line to the angle between the minimum scale line and the maximum scale line by the maximum scale value of the pointer meter to obtain the Pointer meter readings.

In order to achieve the above purpose of the invention, the present invention also provides an automatic identification system for the reading of a pointer meter, including:

The image preprocessing module is used to preprocess the acquired image of the target area including the dial;

The image extraction module is used to extract the dial image from the image of the preprocessed target area;

a position determination module for determining the position of the pointer, the position of the minimum tick mark and the position of the maximum tick mark from the dial image;

The reading calculation module is used for calculating the reading of the pointer type instrument according to the position of the pointer, the position of the minimum tick mark and the position of the maximum tick mark.

Further, the image extraction module specifically includes:

The image correction unit is used for registering the image of the pre-processed target area and the front face image of the dial by using the SURF algorithm, solving the perspective transformation matrix, and performing the pre-processing on the image of the pre-processed target area according to the perspective transformation matrix. Affine transformation;

The dial extraction unit is used for extracting the dial image from the image of the target area which has undergone affine transformation by adopting the Hough circle transformation.

Further, the reading calculation module is specifically used for:

Multiply the ratio of the angle between the pointer and the minimum scale line to the angle between the minimum scale line and the maximum scale line by the maximum scale value of the pointer meter to obtain the Pointer meter readings.

The present invention also provides a computer-readable storage medium comprising instructions that, when executed on a computer, cause the computer to perform the above method.

The present invention also provides a computer device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, and the processor implements the above method when executing the program.

Description of drawings

1 is a flowchart of an automatic identification method for a pointer meter reading provided by an embodiment of the present invention;

FIG. 2 is an effect diagram of an image processing process of an analog meter according to an embodiment of the present invention.

Detailed ways

The principles and features of the present invention will be described below with reference to the accompanying drawings. The examples are only used to explain the present invention, but not to limit the scope of the present invention.

1 is a flowchart of an automatic identification method for a pointer meter reading provided by an embodiment of the present invention, and each step in the method can be based on Visual Studio 2015 under the Windows 10 operating system, and in the open source OpenCV library of version 3.0.4 implementation of each library function.

As shown in Figure 1, the method includes:

S1, preprocessing the acquired image of the target area containing the dial;

Specifically, the dial shot by the inspection robot is stored as an image, and then the image is read in and preprocessed. The preprocessing includes denoising, grayscale, and smoothing filtering with a 5x5 Gaussian convolution kernel.

S2, extract the dial image from the image of the preprocessed target area;

Specifically, in this step, based on the circular feature of the dial, the Hough circle transformation method can be used to extract the dial image in the target area image, so as to determine the circle center and the corresponding radius of the circular surface where the dial image is located.

S3, determine the position of the pointer, the position of the minimum tick mark and the position of the maximum tick mark from the dial image;

Specifically, in this step, an image recognition algorithm is used to determine the position of the pointer, the position of the minimum tick mark and the position of the maximum tick mark in the dial image, wherein, based on the prior knowledge that the ring where the pointer is located has the least interference factor, the information obtained in step S2 is obtained. frame difference processing of the dial image of The non-zero pixel sum of the area, the ring area corresponding to the non-zero pixel and the minimum value is obtained, which is the ring with the smallest interference factor. In the ring area where the pointer is located, calculate the eight neighbors of each non-zero pixel. Domain feature, obtain the continuous pointer outline, output the coordinates corresponding to the pointer center, and the coordinates connecting the center of the circle and the position of the pointer center are the detected pointer positions.

In addition, the scale area of the pointer meter is generally marked with red and green. Therefore, according to the HSV characteristics of different colors, the red area and the green area of the scale can be extracted respectively, and the pixel with the largest y value in the two areas can be found. , the connection between the two pixels and the center of the circle is the corresponding minimum and maximum tick marks.

S4. Calculate the reading of the pointer meter according to the pointer position, the minimum scale line position and the maximum scale line position.

Specifically, since a pointer-type meter generally adopts a uniform scale, in this step, the angle between the pointer and the minimum scale line can be compared with the angle between the minimum scale line and the maximum scale line. Multiplying the ratio of the degree to the maximum scale value of the pointer meter, the reading of the pointer meter can be obtained.

The embodiment of the present invention provides an automatic identification method for the reading of a pointer type meter, which automatically completes the reading recognition of the pointer type meter by performing image processing on the acquired image of the target area including the dial, which can reduce the process of automatic meter identification. Manual participation helps the inspection robot to complete the inspection task smoothly and quickly. It can be applied to the identification of various pointer-type instruments such as barometers and leakage current meters, and has good practicability and expansibility.

Optionally, in this embodiment, step S2 specifically includes:

S2.1. Use the SURF (Speeded Up Robust Features) algorithm to register the image of the preprocessed target area and the front face image of the dial, and solve the perspective transformation matrix, according to the perspective transformation matrix to the described The image of the preprocessed target area is subjected to affine transformation;

Specifically, in the actual operation process, when the inspection robot does not reach the designated inspection point, or when the height of the inspection robot's pan/tilt and the height of the instrument are greatly different, the observation angle will look up or squint, such as As shown in part a in Figure 2, the captured instrument image will be stretched to a certain extent, which will affect the accuracy of the reading. To solve this problem, the SURF registration algorithm is used to correct the image in step S2.1. The process is:

First, the standard template instrument image is acquired, the SURF feature matrix of the detected instrument image and the standard template image is extracted, the four pairs of points with better matching effect are saved, and finally the perspective transformation matrix is solved.

The principle of perspective transformation is as follows:

The essence of perspective transformation is to project the image to a new viewing plane, and its general formula is:

x=x'/w', y=y'/w'

In formula (1) and formula (2), [x'y'w'] is the coordinate after transformation, [uvw] is the coordinate before transformation, a matrix is the perspective transformation matrix, (u, v) is the original image pixel The coordinates, (x, y) are the pixel coordinates after transformation. Since the processing is a two-dimensional image, w in the coordinates before transformation is always 1, that is, a ₃₃ in the perspective transformation matrix of formula (1) is always 1.

Perform affine transformation on the image of the preprocessed target area according to the above perspective transformation matrix, and then the corrected dial image can be obtained, as shown in part b in Fig. 2 .

S2.2, using the Hough circle transformation to extract the dial image from the image of the target area that has undergone affine transformation.

Specifically, the specific process of step S2.2 is: first, use the Canny algorithm to extract the edge of the image, and then perform binarization processing on the image to obtain the edge of the image. Calculate the derivative of the non-zero pixels in the image edge in the x and y directions to obtain the gradient of the image. The formula for calculating the image gradient is as follows:

G _x =(-1)*f(x-1,y-1)+0*f(x,y-1)+1*f(x+1,y-1)+(-2)*f( x1,y)+0*f(x,y)+2*f(x-1,y)+(-1)*f(x-1,y+1)+0*f(x,y+1 )+1*f(x+1,y+1) (3)

G _y =1*f(x-1,y-1)+2*f(x,y-1)+1*f(x+1,y-1)+0*f(x-1,y) +0*f(x,y)+0*f(x+1,y)+(-1)*f(x-1,y+1)+(-2)*f(x,y+1) +(-1)*f(x+1,y+1) (4)

In equations (3) and (4), G _x and G _y are the gradients of the image in the x and y directions at the (x, y) point, respectively, f() is the grayscale transformation function, f(a, b) represents the gray value of the image at point (a, b).

Then use formula (5) and formula (6) to calculate the gradient and gradient direction at each pixel position:

ang(x,y)=arctan(G _y /G _x ) (6)

In the formula, G(x, y) represents the gradient value of the pixel at (x, y), and ang(x, y) represents the gradient direction of the pixel at (x, y).

Then, from the direction of the gradient and its opposite direction, for each pixel within the range of the minimum and maximum values of the radius parameter, superposition calculation is performed in the accumulator, and the position information of these pixels is recorded, and finally selected in the accumulator Candidate centers, all the candidate centers are arranged in descending order of the sum size, and the candidate center with the largest sum is selected as the best center, and then find the largest supported radius as the center and radius of the circle where the dial is located. In addition, for the pointer-type instrument with a hollow part in the center, it is also necessary to use a mask to remove the hollow part in the center of the dial, so as to extract the ring area of the dial, as shown in part c in Figure 2.

In this embodiment, the stretched image due to the observation angle is corrected into a frontal image through affine transformation, and subsequent reading recognition is performed to improve the reading accuracy of the pointer meter.

Optionally, in this embodiment, in step S3, the position of the pointer is determined from the dial image, which specifically includes:

S3.1. Search each ring area in the dial image along the radial direction, calculate the sum of non-zero pixels of each ring area, and use the ring area corresponding to the non-zero pixel and the minimum value as the target ring area ;

Specifically, in this step, the ring can be searched from the outside to the inside in the radial direction or the ring can be searched from the inside to the outside. For a pointer-type instrument with a hollow center, since the pointer cannot be identified in the hollow part, it needs to be radially from the outside to the inside. The specific search process is as follows:

S3.1.1. Use the mask to extract the center and radius r of the entire disk, then the area of the initial disk is:

S=π*r^2 (7)

S3.1.2, the search progress size dp=0.01, the distance accuracy in pixels, the searched ring area is:

S1=(π*r^2)-(π*(r-dp)^2) (8)

S3.1.3. Calculate the total number sum of non-zero pixels in the S1 area, and output the ring area containing the pointer corresponding to the minimum sum. For a pointer-type instrument with a hollow center, the search can be stopped when the radius of the hollow part is found. For example , when the search reaches the circle radius r1<0.3r to end the search.

S3.2. Use eight neighborhood contours to track each non-zero pixel point in the target annular region, and determine the position of the pointer.

Specifically, in the corresponding circular area found in step S3.1, use the Canny algorithm to obtain the edge, detect the point with a pixel value of 255, and use the eight-area contour tracking to find the position of the pointer. The coordinates corresponding to the center of the pointer are output, and the coordinates connecting the center of the circle and the position of the center of the pointer are the detected pointers, as shown in part d in Figure 2.

Optionally, in this embodiment, in step S3, the position of the minimum tick mark and the position of the maximum tick mark are determined from the dial image, which specifically includes:

The scale line area in the dial image is extracted according to the HSV feature of the dial image, and the minimum scale line position and the maximum scale line position are determined according to the scale line area.

Specifically, HSV represents Hue hue, Saturation saturation, and Value lightness, respectively. For the red tick area, the h value extraction range is 320-360, the S value is 90-255, and the V value is 90-255. Regarding the green tick line The h value extraction value of the region is 113-189, the S value is 90-255, and the V value is 90-255. The point with the largest pixel coordinate y value in the red area is the position of the minimum scale line, and the point with the largest pixel coordinate y value in the green area is the position of the largest scale line, as shown in parts e and f in Figure 2.

An embodiment of the present invention provides an automatic identification system for pointer meter readings, including:

The image preprocessing module is used to preprocess the acquired image of the target area including the dial;

The image extraction module is used to extract the dial image from the image of the preprocessed target area;

a position determination module for determining the position of the pointer, the position of the minimum tick mark and the position of the maximum tick mark from the dial image;

The reading calculation module is used for calculating the reading of the pointer type instrument according to the position of the pointer, the position of the minimum tick mark and the position of the maximum tick mark.

Optionally, in this embodiment, the image extraction module specifically includes:

The image correction unit is used for registering the image of the pre-processed target area and the front face image of the dial by using the SURF algorithm, solving the perspective transformation matrix, and performing the pre-processing on the image of the pre-processed target area according to the perspective transformation matrix. Affine transformation;

The dial extraction unit is used for extracting the dial image from the image of the target area which has undergone affine transformation by adopting the Hough circle transformation.

Optionally, in this embodiment, the reading calculation module is specifically used for:

Multiply the ratio of the angle between the pointer and the minimum scale line to the angle between the minimum scale line and the maximum scale line by the maximum scale value of the pointer meter to obtain the Pointer meter readings.

Embodiments of the present invention further provide a computer-readable storage medium, which includes instructions, when the instructions are run on a computer, causing the computer to execute the method steps in the foregoing method embodiments; or store various software of the foregoing system embodiments The corresponding command of the module.

An embodiment of the present invention further provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, and the processor implements the above method when executing the program method steps in the example.

The reader should understand that in the description of this specification, reference to the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples", etc., is intended to incorporate the embodiment or example. A particular feature, structure, material, or characteristic described is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms are not necessarily directed 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, those skilled in the art may combine and combine the different embodiments or examples described in this specification, as well as the features of the different embodiments or examples, without conflicting each other.

Those skilled in the art can clearly understand that, for the convenience and brevity of description, for the specific working process of the modules and units in the above-described system embodiments, reference may be made to the corresponding processes in the foregoing method embodiments, which will not be repeated here.

In the several embodiments provided in this application, it should be understood that the disclosed apparatus and method may be implemented in other manners. For example, the device embodiments described above are only illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated. to another system, or some features can be ignored, or not implemented.

Units described as separate components may or may not be physically separated, and components shown as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solutions of the embodiments of the present invention.

In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit. The above-mentioned integrated units may be implemented in the form of hardware, or may be implemented in the form of software functional units.

The integrated unit, if implemented as a software functional unit and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on such understanding, the technical solution of the present invention is essentially or a part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , including several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, Read-Only Memory (ROM, Read-Only Memory), Random Access Memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program codes .

The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of various equivalent modifications or changes within the technical scope disclosed by the present invention. Replacement, these modifications or replacements should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims (9)

1. an automatic identification method of pointer type meter reading, is characterized in that, comprises: Preprocess the acquired image of the target area containing the dial; Extract the dial image from the preprocessed image of the target area; determine the position of the pointer, the position of the minimum tick mark and the position of the maximum tick mark from the dial image; Determine the reading of the pointer meter according to the position of the pointer, the position of the minimum tick mark and the position of the maximum tick mark; Determining the position of the pointer in the dial image specifically includes: Searching for each annular area in the dial image along the radial direction, calculating the sum of non-zero pixels of each annular area, and using the annular area corresponding to the non-zero pixel and the minimum value as the target annular area; Eight neighborhood contours are used to track each non-zero pixel point in the target annular region to determine the position of the pointer. 2. The automatic identification method of a pointer-type meter reading according to claim 1, wherein the extraction of the dial image from the image of the preprocessed target area specifically comprises: The SURF algorithm is used to register the image of the pre-processed target area and the front face image of the dial, and the perspective transformation matrix is solved, and the image of the pre-processed target area is subjected to affine transformation according to the perspective transformation matrix; The dial image is extracted from the affine-transformed image of the target area using the Hough circle transform. 3. the automatic identification method of a kind of pointer meter reading according to claim 1, it is characterized in that, from described dial image, determine the minimum tick mark position and the maximum tick mark position, specifically comprises: The tick line area in the dial image is extracted according to the HSV feature of the dial image, and the minimum tick line position and the maximum tick line position are determined according to the tick line area. 4. The automatic identification method of a pointer meter reading according to any one of claims 1 to 3, wherein the reading of the pointer meter is determined according to the pointer position, the minimum tick mark position and the maximum tick mark position , including: Multiply the ratio of the angle between the pointer and the minimum scale line to the angle between the minimum scale line and the maximum scale line by the maximum scale value of the pointer meter to obtain the Pointer meter readings. 5. An automatic identification system for pointer meter readings, characterized in that, comprising: The image preprocessing module is used to preprocess the acquired image of the target area including the dial; The image extraction module is used to extract the dial image from the image of the preprocessed target area; a position determination module for determining the position of the pointer, the position of the minimum tick mark and the position of the maximum tick mark from the dial image; a reading calculation module for determining the reading of the pointer meter according to the position of the pointer, the position of the minimum tick mark and the position of the maximum tick mark; The position determination module is specifically used for: Searching for each annular area in the dial image along the radial direction, calculating the sum of non-zero pixels of each annular area, and using the annular area corresponding to the non-zero pixel and the minimum value as the target annular area; Eight neighborhood contours are used to track each non-zero pixel point in the target annular region to determine the position of the pointer. 6. The automatic identification system of a pointer meter reading according to claim 5, wherein the image extraction module specifically comprises: The image correction unit is used for registering the image of the preprocessed target area and the front face image of the dial by using the SURF algorithm, solving the perspective transformation matrix, and performing the preprocessing on the image of the preprocessed target area according to the perspective transformation matrix. Affine transformation; The dial extraction unit is used for extracting the dial image from the image of the target area which has undergone affine transformation by adopting the Hough circle transformation. 7. The automatic identification system of a pointer meter reading according to claim 5 or 6, wherein the reading calculation module is specifically used for: Multiply the ratio of the angle between the pointer and the minimum scale line to the angle between the minimum scale line and the maximum scale line by the maximum scale value of the pointer meter to obtain the Pointer meter readings. 8. A computer-readable storage medium comprising instructions, characterized in that, when executed on a computer, the instructions cause the computer to perform the method according to any one of claims 1 to 4. 9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the program as claimed in the claims when executing the program The method of any one of 1 to 4.

Images