CN103207987A - Indicating value identification method of dial instrument - Google Patents

Abstract

The invention discloses an indicating value identification method of a dial instrument. The indicating value identification method of the dial instrument belongs to the field of artificial intelligence. The technical scheme of the indicating value identification method of the dial instrument is that a computer automatically acquires a dial plate image of the dial instrument, divides out needed image feature regions, obtains the feature of a number to determine the numerical value of the number according to centre-of-gravity position, perpendicular line features and horizontal projection operations, and finally computes the actual evaluation of the dial instrument through an abscissa of the number and a former pointer and a scale mark. The image processing method for reading the indicating value of the dial instrument can accurately acquires the coordinate value of the pointer and coordinate value of an intersection of the pointer and scale marks, can avoid indicating value calculation deviation problems caused by camera lens distortion, uneven scales of dial plates, or the noncoincidence of a fixed point of the pointer and the centre of a circle which passes through the scale mark of the dial plate due to the fact that the angle method is utilized in the prior art, improves the efficiency and accuracy of reading the indicating value of the dial instrument, is applicable to a plurality of hardware platforms, and has high transportability and stability.

Description

Indicating number identification method of pointer instrument

Technical Field

The invention belongs to the field of artificial intelligence, and particularly relates to a method for identifying the number of a pointer instrument.

Background

Pointer instruments are still widely applied in industrial production and daily life, and the reading of the pointer instruments are generally finished manually, namely, the relative position of a pointer and a scale mark and the absolute position of the scale mark are identified by human eyes, and the indicating value of the instrument is estimated and read. The method has low efficiency, high labor intensity and easy error.

To solve such problems, pointer instrument interpretation systems based on computer vision technology have been developed. However, the existing method has the defects of large computation amount, high complexity and requirement on the real-time performance of a hardware platform.

Disclosure of Invention

The invention provides a method for identifying the number of a pointer instrument, aiming at the problems of low accuracy, low efficiency and the like of identifying the positions of a pointer and a scale mark of the instrument in the prior art.

A method for identifying the number of a pointer instrument is characterized by comprising the following steps:

step 1: collecting an instrument dial image by using a camera, and transmitting the instrument dial image to a computer for storage and display;

step 2: determining the rotation center coordinate of the pointer by using a difference method;

and step 3: segmenting an instrument dial image by using an area segmentation method, setting the tracking area condition of the dial as the longest circumference so as to obtain an image containing numbers, scale marks and a pointer part in the dial image, and binarizing the obtained image;

and 4, step 4: performing polar coordinate transformation on an image containing numbers, scale marks and a pointer part;

and 5: obtaining information of the scale mark by making a vertical straight line in an image including the number, the scale mark and the pointer portion;

step 6: obtaining an image only containing digital information by adopting a contour tracking method and taking the maximum perimeter and the minimum area as filling conditions of the area;

and 7: performing row-column linear scanning on an image only containing digital information, and determining a region of a number to be identified;

and 8: obtaining the feature of the number according to the gravity center position, the vertical line feature and the horizontal projection operation to determine the numerical value of the number;

and step 9: and finally, calculating the actual evaluation value of the pointer instrument according to the number and the abscissa of the previous pointer and the scale mark.

The specific step of determining the rotation center coordinate of the pointer by using the difference method comprises the following steps:

step 201: firstly, an original image I is collected₁(x, y), rotating the pointer by 30-45 degrees, and then collecting an image I₂(x,y)；

Step 202: carrying out subtraction operation on the two images, setting the pixels in different gray scale areas as 1 and setting the pixels with the same gray scale as 0;

step 203: the clipped image Δ I (x, y) with the background and other indicia on the dial removed is obtained, via step 202, to determine the center of rotation of the pointer.

The process of transforming by the coordinate system is as follows:

step 401: setting the image obtained in the step 3 as I (x, y);

step 402: translating the origin coordinates of the image in the step 401 to the rotation center (X, Y) of the pointer, and vertically overturning the translated image I' (X-X, Y-Y);

step 403: performing polar coordinate transformation on the overturned image I '' ((X-X), Y-Y) according to a polar coordinate transformation formula; the polar coordinate transformation formula coordinates are:

$\left\{\begin{array}{c}\mathrm{\θ}=\arctan \left(\frac{y-Y}{-(x-X)}\right)\\ \mathrm{\ρ}=\sqrt{{(x-X)}^2+{(y-Y)}^2}\end{array}\right.$

the process of obtaining the information of the scale mark by making a vertical straight line in the dial plate image of the pointer instrument is as follows:

step 501: drawing a vertical straight line from top to bottom in the image obtained in the step 3;

step 502: obtaining information of the scale marks through the data of the vertical lines and the coincidence of the scale marks in the image obtained in the step 3 and the number of the scale mark pixel points;

the column-line scanning is carried out on the image only containing the digital information, and the process of determining the area of the number to be identified comprises the following steps:

step 701: performing column-line scanning on the image only containing the numbers obtained in the step 6, and calculating the central abscissa cx of each number_j；

Step 702: comparing each cx in turn_jWith the centre of rotation X of the pointer_pAnd d = | cx_j-X_p|；

Step 703: satisfy min (d) and cx_j<X_pI.e. the area of the number to be identified.

The process of obtaining the feature of the number to determine the numerical value of the number according to the gravity center position, the vertical line feature and the horizontal projection operation is as follows:

step 801: normalizing the region of the number to be recognized, dividing the region of the number to be recognized into an upper part and a lower part by taking a central line as a reference, setting the size of the region to be recognized to be M multiplied by N, counting the number of pixel points of the upper part and the lower part of the region of the number to be recognized, and adopting the calculation formula as follows:

${\mathrm{<figure-callout\; id="1"\; label="S"\; filenames="HDA00002867206700012.png"\; state="\{\{state\}\}">S</figure-callout>}}_1=\underset{x=1}{\overset{M}{\mathrm{\&Sigma;}}}\underset{y=1}{\overset{\frac{N}{2}}{\mathrm{\&Sigma;}}}g(x,y)$ ${\mathrm{<figure-callout\; id="2"\; label="S"\; filenames="HDA00002867206700012.png"\; state="\{\{state\}\}">S</figure-callout>}}_2=\underset{x=1}{\overset{M}{\mathrm{\&Sigma;}}}\underset{y=\frac{N}{2}+1}{\overset{N}{\mathrm{\&Sigma;}}}g(x,y)$

wherein S is₁The number of the lower half part pixel points is; s₂The number of the pixels of the working part is the number of the pixels of the working part;

g (x, y) is the seating of a pixel point in the region of the number to be recognizedAnd (4) marking.

Step 802: according to S₁And S₂The gravity center is divided into upper, middle and lower 3 types according to the size relationship of (1); when S is₁>S₂The center of gravity is on the top; when S is₁=S₂If so, the center of gravity is centered; when S is₁<S₂The center of gravity is off;

step 803: all numbers are grouped into 3 groups according to step 802, the upper group including 5,7 and 9; the middle group includes 0,1,3, and 8; the lower group comprised 2,4 and 6;

step 804: judging whether the number to be identified belongs to the middle group, if so, executing the step 805 in sequence: otherwise, execution begins with step 807;

step 805: for the 4 numbers of the middle group, 0,1,3, 8; respectively arranging 3 vertical scanning lines at N/3, N/2 and 2N/3 of the width of the digital image, and calculating the number of intersection points of the scanning lines and the digital image;

${\mathrm{<figure-callout\; id="1"\; label="S"\; filenames="HDA00002867206700012.png"\; state="\{\{state\}\}">S</figure-callout>}}_{\frac{1}{3}}=\underset{y=1}{\overset{N}{\mathrm{\&Sigma;}}}\underset{x=\frac{M}{3}}{\mathrm{\&Sigma;}}g(x,y)$ ${\mathrm{<figure-callout\; id="1"\; label="S"\; filenames="HDA00002867206700012.png"\; state="\{\{state\}\}">S</figure-callout>}}_{\frac{1}{2}}=\underset{y=1}{\overset{N}{\mathrm{\&Sigma;}}}\underset{x=\frac{M}{2}}{\mathrm{\&Sigma;}}g(x,y)$ ${\mathrm{<figure-callout\; id="2"\; label="S"\; filenames="HDA00002867206700012.png"\; state="\{\{state\}\}">S</figure-callout>}}_{\frac{2}{3}}=\underset{y=1}{\overset{N}{\mathrm{\&Sigma;}}}\underset{x=\frac{2M}{3}}{\mathrm{\&Sigma;}}g(x,y)$

wherein,

the number of the intersection points of the scanning lines and the numbers is respectively; g (x, y) is the coordinates of the pixel points in the region of the number to be identified;

thus, for 1 feature vector per number:

$R\left(i\right)=[{\mathrm{<figure-callout\; id="1"\; label="S"\; filenames="HDA00002867206700012.png"\; state="\{\{state\}\}">S</figure-callout>}}_{\frac{1}{3}},{\mathrm{<figure-callout\; id="1"\; label="S"\; filenames="HDA00002867206700012.png"\; state="\{\{state\}\}">S</figure-callout>}}_{\frac{1}{2}},{\mathrm{<figure-callout\; id="2"\; label="S"\; filenames="HDA00002867206700012.png"\; state="\{\{state\}\}">S</figure-callout>}}_{\frac{2}{3}}]i=\mathrm{0,1,3,8}$

step 806: after extracting the characteristic vectors, matching the characteristic vectors corresponding to the numbers of the middle group, and jumping to 809; the feature vector table is shown in table 3:

TABLE 1 feature vector Table

Number of	Feature vector
		0	R(0)=[2,2,2]
1	R(1)=[1,1,N]N>1,
		3	R(3)=[2,3,3]
8	R(8)=[3,3,3]

Step 807: carrying out feature extraction and identification on the numbers of the upper group and the lower group according to horizontal direction projection, so that the number to be identified has N features C (i);

$C\left(i\right)=\underset{1\&le;x\&le;M}{\mathrm{\&Sigma;}}\underset{y=i}{\mathrm{\&Sigma;}}g(x,y)i=\mathrm{2,4,5,6,7,9}$

step 808: finally, the extracted digital features need to be distinguished based on a minimum distance method distinguishing method:

$J\left(i\right)=\underset{k=1}{\overset{N}{\mathrm{\&Sigma;}}}|Y\left(k\right)-X(i,k)|i=\mathrm{2,4,5,6,7,9}$

wherein J (i) is the distance between the sample to be detected and the type i template; y (k) is the ith feature of the sample to be tested, and X (i, k) is the ith feature of the kth class template.

Step 809: acquiring intersection point coordinates of the pointer and the scale marks, and calculating the numerical value of the instrument according to a numerical calculation formula and the numerical value closest to the left side of the pointer; the indication calculation formula is as follows:

$V=i+\frac{L_i}{L}$

wherein V is the indicating number value of the instrument, i is the specific value of the number nearest to the pointer on the left side of the pointer, and L_iThe distance between the coordinate of the intersection point of the pointer and the scale mark and the nearest number on the left side from the pointer, and L is the distance between two numbers on the dial plate.

The image processing method for reading the indicating value of the pointer instrument has the advantages that the coordinate values of the pointer and the intersection point of the pointer and the scale mark can be accurately obtained, the problem of indicating value calculation deviation caused by distortion of a camera lens, uneven dial scale or misalignment of the circle center of a circle passing through the dial scale mark and a fixed point of the pointer due to the adoption of an angle method in the prior art can be solved, and the efficiency and the accuracy for reading the indicating value of the pointer instrument are improved; and the method can be suitable for various hardware platforms and has strong portability and stability.

Drawings

FIG. 1 is a flow chart of a method for identifying the number of a pointer instrument provided by the invention;

FIG. 2 is a diagram illustrating the transformation result of the coordinate system provided by the present invention;

fig. 3 is a schematic diagram of vertical line scanning of numbers provided by the present invention.

Detailed Description

The preferred embodiments will be described in detail below with reference to the accompanying drawings. It should be emphasized that the following description is merely exemplary in nature and is not intended to limit the scope of the invention or its application.

The invention is described in further detail below with reference to the accompanying drawings:

the dial plate image of the pointer instrument is acquired by a camera, read into a computer system through a USB interface, read a frame of data into a computer memory in an array form by utilizing an automatic identification digital program in the computer, and simultaneously sent to a corresponding display memory, and finally, the specific indication value of the instrument is displayed on a screen.

The rotation center coordinates of the pointer are determined by a difference method. After the position of the measured table is determined, an original image is taken, and then a second image after the pointer deflects at a large angle is acquired. If the two image pointers are close, fitting of two straight lines is difficult to carry out respectively by using a least square method, and subtraction operation is carried out on corresponding pixel points of the two images with different pointer positions. The pixel position of the same gray scale is 0 and the pixel position of the different gray scale areas is 1. This subtracts the background, including the tick marks and numerical indicia in the dial, leaving only the portion of the image that has changed, the different location pointers. A silhouette image is obtained so that the presence of the pointer can be detected and the rotation center information of the dial pointer can be determined.

The acquired pointer type instrument panel image data are stored in a memory in a matrix form; therefore, "coordinate system transformation" is to convert the image from a rectangular coordinate system to a polar coordinate system. According to the characteristics of the dial plate image, firstly, carrying out coordinate translation on the dial plate image; taking the rotation center of the pointer as a new coordinate origin; the translated image is inverted upward with the abscissa as an axis, and then subjected to polar coordinate transformation. The scale marks are always vertical to the abscissa axis, so that accurate scale mark information can be obtained by performing column-to-column linear scanning on the converted dial plate image from top to bottom. Vertical lines (from top to bottom) are drawn on the binary rectangular dial plate refined image, the number of the scale line pixel points is combined with the data of the overlapped part of the vertical lines and the scale lines, and the scale lines in the changed dial plate image are characterized by being continuous straight lines, so that the types of the scale lines can be effectively judged.

And setting the conditions of region filling as maximum perimeter and minimum area by using a contour tracing method to obtain an image of the region with the pointer and the scale mark removed. And (4) solving the center of a number nearest to the abscissa of the rotation center of the pointer line or the coordinate of the upper left corner, and determining the digital area to be identified.

The feature extraction of the numbers is mainly performed according to the gravity center position, the vertical line feature and the horizontal projection.

After a digital area to be identified is extracted, dividing the image into an upper part and a lower part by taking a central line of the image as a reference; and accumulating the total number of the black pixels of each part in sequence. Comparing the two results, if the total number of the upper part is larger than that of the lower part, the number in the image belongs to the upper group; if equal, the two belong to the middle group; if the total number of the upper parts is less than that of the lower parts, the lower parts belong to the lower group. The numbers of the upper group are 5,7, 9; the numbers of the middle group are 0,1,3, 8; the lower group is 2,4, 6.

To distinguish the four numbers of the middle group, a vertical line feature must be used. The vertical line feature is that the vertical scanning lines are respectively arranged on 1/3,1/2 and 2/3 on the horizontal coordinate of the digital image, and the number of the intersection points of each vertical line and the digital image is respectively calculated. The 4 numbers of the middle group, the number of intersections on these vertical lines, are clearly different and therefore can be well distinguished.

For the differentiation of the numbers in the upper and middle groups, a horizontal projection operation must be used. And horizontal projection, wherein horizontal projection refers to a set of gray value accumulation calculated quantities along a section of the image in the horizontal direction, and can be simplified into a set of count values of black pixels in a binary image. Thus after horizontal projection, the digital image is divided longitudinally into several regions; the total number of black pixels in this area is then counted. We call this total number a feature of the digital image in this region. After the feature extraction, the minimum distance method can be adopted to match the sample feature to be detected with the template feature.

The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims (6)

1. A method for identifying the number of a pointer instrument is characterized by comprising the following steps:

step 1: collecting an instrument dial image by using a camera, and transmitting the instrument dial image to a computer for storage and display;

step 2: determining the rotation center coordinate of the pointer by using a difference method;

and step 3: segmenting an instrument dial image by using an area segmentation method, setting the tracking area condition of the dial as the longest circumference so as to obtain an image containing numbers, scale marks and a pointer part in the dial image, and binarizing the obtained image;

and 4, step 4: performing polar coordinate transformation on an image containing numbers, scale marks and a pointer part;

and 5: obtaining information of the scale mark by making a vertical straight line in an image including the number, the scale mark and the pointer portion;

step 6: obtaining an image only containing digital information by adopting a contour tracking method and taking the maximum perimeter and the minimum area as filling conditions of the area;

and 7: performing row-column linear scanning on an image only containing digital information, and determining a region of a number to be identified;

and 8: obtaining the feature of the number according to the gravity center position, the vertical line feature and the horizontal projection operation to determine the numerical value of the number;

and step 9: and finally, calculating the actual evaluation value of the pointer instrument according to the number and the abscissa of the previous pointer and the scale mark.

2. The method for identifying the number of the pointer instrument as claimed in claim 1, wherein the step of determining the rotation center coordinates of the pointer by using the difference method comprises the following steps:

step 201: firstly, an original image I is collected₁(x, y), rotating the pointer by 30-45 degrees, and then collecting an image I₂(x,y)；

Step 202: carrying out subtraction operation on the two images, setting the pixels in different gray scale areas as 1 and setting the pixels with the same gray scale as 0;

step 203: the silhouette image Δ I (x, y) with the background and other indicia on the dial removed is obtained, via step 202, to determine the center of rotation of the pointer.

3. The method for identifying the number of the pointer instrument as claimed in claim 1, wherein the process of transforming by the coordinate system is as follows:

step 401: setting the image obtained in the step 3 as I (x, y);

step 402: translating the origin coordinates of the image in the step 401 to the rotation center (X, Y) of the pointer, and vertically overturning the translated image I' (X-X, Y-Y);

step 403: performing polar coordinate transformation on the overturned image I '' ((X-X), Y-Y) according to a polar coordinate transformation formula; the polar coordinate transformation formula coordinates are:

$\left\{\begin{array}{c}\mathrm{\&theta;}=\arctan \left(\frac{y-Y}{-(x-X)}\right)\\ \mathrm{\&rho;}=\sqrt{{(x-X)}^2+{(y-Y)}^2}\end{array}\right..$

4. the method for identifying the number of the pointer instrument as claimed in claim 1, wherein the process of obtaining the information of the graduation mark by making a vertical straight line in the dial image of the pointer instrument is:

step 501: drawing a vertical straight line from top to bottom in the image obtained in the step 3;

step 502: and 3, obtaining the information of the scale marks according to the data of the vertical line and the coincidence of the scale marks in the image obtained in the step 3 and the number of the scale mark pixel points.

5. The method for identifying the number of the pointer instrument as claimed in claim 1, wherein the step of performing column-line scanning on the image only containing the digital information to determine the area of the number to be identified comprises:

step 701: performing column-line scanning on the image only containing the numbers obtained in the step 6, and calculating the central abscissa cx of each number_j；

Step 702: comparing each cx in turn_jWith the centre of rotation X of the pointer_pAnd d = | cx_j-X_p|；

Step 703: satisfy min (d) and cx_j<X_pI.e. the area of the number to be identified.

6. The method as claimed in claim 1, wherein the step of determining the numerical value of the number by deriving the feature of the number according to the gravity center position, the vertical line feature and the horizontal projection operation comprises:

step 801: normalizing the region of the number to be recognized, dividing the region of the number to be recognized into an upper part and a lower part by taking a central line as a reference, setting the size of the region to be recognized to be M multiplied by N, counting the number of pixel points of the upper part and the lower part of the region of the number to be recognized, and adopting the calculation formula as follows:

$S_1=\underset{x=1}{\overset{M}{\mathrm{\&Sigma;}}}\underset{y=1}{\overset{\frac{N}{2}}{\mathrm{\&Sigma;}}}g(x,y)$ $S_2=\underset{x=1}{\overset{M}{\mathrm{\&Sigma;}}}\underset{y=\frac{N}{2}+1}{\overset{N}{\mathrm{\&Sigma;}}}g(x,y)$

wherein S is₁The number of the lower half part pixel points is; s₂The number of the pixels of the working part is the number of the pixels of the working part;

g (x, y) is the coordinates of the pixel points in the region of the number to be recognized.

Step 802: according to S₁And S₂The gravity center is divided into upper, middle and lower 3 types according to the size relationship of (1); when S is₁>S₂The center of gravity is on the top; when S is₁=S₂If so, the center of gravity is centered; when S is₁<S₂The center of gravity is off;

step 803: all numbers are grouped into 3 groups according to step 802, the upper group including 5,7 and 9; the middle group includes 0,1,3, and 8; the lower group comprised 2,4 and 6;

step 804: judging whether the number to be identified belongs to the middle group, if so, executing the step 805 in sequence; otherwise, execution begins with step 807;

step 805: for the 4 numbers of the middle group, 0,1,3, 8; respectively arranging 3 vertical scanning lines at N/3, N/2 and 2N/3 of the width of the digital image, and calculating the number of intersection points of the scanning lines and the digital image;

$S_{\frac{1}{3}}=\underset{y=1}{\overset{N}{\mathrm{\&Sigma;}}}\underset{x=\frac{M}{3}}{\mathrm{\&Sigma;}}g(x,y)$ $S_{\frac{1}{2}}=\underset{y=1}{\overset{N}{\mathrm{\&Sigma;}}}\underset{x=\frac{M}{2}}{\mathrm{\&Sigma;}}g(x,y)$ $S_{\frac{2}{3}}=\underset{y=1}{\overset{N}{\mathrm{\&Sigma;}}}\underset{x=\frac{2M}{3}}{\mathrm{\&Sigma;}}g(x,y)$

wherein,

the number of the intersection points of the scanning lines and the numbers is respectively; g (x, y) is the coordinates of the pixel points in the region of the number to be identified;

thus, for 1 feature vector per number:

$R\left(i\right)=[S_{\frac{1}{3}},S_{\frac{1}{2}},S_{\frac{2}{3}}]i=\mathrm{0,1,3,8}$

step 806: after extracting the feature vectors, matching the feature vectors corresponding to the numbers in the middle group, and jumping to step 809;

step 807: carrying out feature extraction and identification on the numbers of the upper group and the lower group according to horizontal direction projection, so that the number to be identified has N features C (i);

$C\left(i\right)=\underset{1\&le;x\&le;M}{\mathrm{\&Sigma;}}\underset{y=i}{\mathrm{\&Sigma;}}g(x,y)i=\mathrm{2,4,5,6,7,9}$

step 808: finally, the extracted digital features need to be distinguished based on a minimum distance method distinguishing method:

$J\left(i\right)=\underset{k=1}{\overset{N}{\mathrm{\&Sigma;}}}|Y\left(k\right)-X(i,k)|i=\mathrm{2,4,5,6,7,9}$

wherein J (i) is the distance between the sample to be detected and the type i template; y (k) is the ith feature of the sample to be tested, and X (i, k) is the ith feature of the kth class template.

Step 809: acquiring intersection point coordinates of the pointer and the scale marks, and calculating the numerical value of the instrument according to a numerical calculation formula and the numerical value closest to the left side of the pointer; the indication calculation formula is as follows:

$V=i+\frac{L_i}{L}$

wherein V is the indicating number value of the instrument, i is the specific value of the number nearest to the pointer on the left side of the pointer, and L_iThe distance between the coordinate of the intersection point of the pointer and the scale mark and the nearest number on the left side from the pointer, and L is the distance between two numbers on the dial plate.

Images