RU2404440C1 - Method of automatic checking of pointer-type instruments and device for its implementation - Google Patents

Abstract

FIELD: instrument making.

SUBSTANCE: device contains transmitting video camera, checkable instrument, processing and control unit, calibration signal forming unit, program-controlled scanning unit, optical system. Preliminary there installed the signal between pointer turning angle and instrument readings, the image of instrument indication part is scanned at minimum quantity of bars for the specified inaccuracy value. The image corresponding to instrument zero reading is saved. Discrete increasing signal is specified and the second image corresponding to instrument new reading is saved. The comparison of element values of two received array variables with the threshold level is performed. Inclination angle of two straight lines passing through the points of corresponding array variables and turning angle, which determines the readings of pointer-type instrument and inaccuracy is defined.

EFFECT: enhancement.

4 cl, 2 dwg

Description

The invention relates to electrical engineering and can be used for automatic calibration of measuring instruments (SIP).

The purpose of the invention is to increase the accuracy and performance of verification.

At the first stage, the automatic verification method involves obtaining a scale equation or an analytical relationship between the rotation angle (φ) of the dial indicator and the readings on the scale (A). This connection can be generally represented by a polynomial of the second degree

where a ₀ , a ₁ , a ₂ are the coefficients of the polynomial. In the event that the coefficient values in expression (1) are known in advance and the approximation error is much less than the accuracy class of the instrument being verified, the first step is not performed. Let F (x, y) be a continuous image of the indicator part of the device, then a discrete image obtained as a result of scanning by an incomplete raster can be represented

where the second factor is a sampling function consisting of identical pulses P (x, y), which form a lattice with a step (Δx, Δy). Denote the discrete image corresponding to the zero reading of the device as F ⁰ (i, j). F ⁰ (i, j) is remembered. When a reference signal is received at the input of the device, the pointer is rejected and, after the end of the transition process, the corresponding image is converted in accordance with expression (2), which we denote by F '(i, j). Remember F '(i, j). After that, subtract

Moreover, in the process of subtraction, two arrays are formed, one of which contains an image of the pointer f ^- (i, j) corresponding to the zero reading, and the other to the current f ⁺ (i, j).

Formation occurs according to the following rule:

∇ - operator, which, given the sign of the difference, produces the formation of arrays;

R (i, j) is the difference modulus made by expression (3).

Thus, in the array f ^- (i, j) there will be an image of only the pointer in the zero position, while other visual information located on the scale of the device (labels, risks, etc.) disappears due to subtraction. Similarly, the f ⁺ (i, j) array stores only the image corresponding to the current (after deviation) position of the pointer. After that, each element f ^- (i, j) and f ⁺ (i, j) is compared with the threshold П and two arrays of coordinates [X ₁ , Y ₁ ] and [X ₂ , Y ₂ ] are formed that determine the position of the pointer in the system X0Y coordinates, i.e.

Where

ξ is the threshold function;

f * (i, j) - value of array elements:

P ₁ and P ₂ - thresholds.

The values of P ₁ , P _{2 are} determined by analyzing the histograms of the distribution of the numerical values of the elements of the arrays f ^- (i, j) and f ⁺ (i, j). This histogram will have a dual-mode character. The value of P will correspond to a “depression” between two “peaks”. The need to choose two thresholds is due to the fact that, in the general case, the numerical values of the arrays f ^- (i, j) and f ⁺ (i, j) are different.

Then the coordinates of the arrays [X ₁ , Y ₁ ] and [X ₂ , Y ₂ ] are processed using the least squares method, the angle of rotation of the pointer is determined as follows:

where X ₁ , Y ₁ - coordinate values that determine the initial position of the pointer;

X ₂ , Y ₂ - coordinate values that determine the current (as a result of moving) position of the pointer.

The angle of rotation of the pointer φ is calculated as

The value of this rotation angle is substituted into expression (1) and the value of the measured value is determined A. After that, knowing the value of the reference signal A ₀ , which is supplied to the device to be verified, the measurement error is calculated

Figure 1 shows a block diagram of a device for implementing the method; figure 2 is a functional diagram of a block of program-controlled scanning.

The device for implementing the method comprises a series-connected verified device 1, an optical system 2, a transmitting television camera 3, a program-controlled scanning unit 4, a control and processing unit 5 and a calibrated signal generation unit 6, the output of which is connected to the input of the device being verified 1. the second output of the control and processing unit 5 is connected to the controlled input of the program-controlled scanning unit 4.

Block 4 software-controlled scanning (figure 2) contains a series-connected amplifier 7, the voltage converter 8 is a code, a pairing unit 9, a control unit 10, a memory unit 11 and a read window driver 12, the output of which is connected to the second input of the voltage converter 8 code, the second output of the control unit 10 is connected to the input of the memory unit 11, the second input of which is the output of the unit 4, the input of the amplifier 7 is the input, and the output of the interface unit 9 is the second output of the program-controlled scanning unit 4.

The device operates as follows.

The image of the indicator part of the device under verification 1 is projected through the optical system 2 onto the photosensitive surface of the television camera 3. Using block 4 of program-controlled scanning, the image is scanned according to the minimum number of columns (rows) for a given accuracy, quantized in time and level, and recorded in block 5 processing and control, and the number of input columns (rows) and the step between them are determined before the start of verification, based on the required accuracy, and are set programmatically in block 4. After that from the output of the calibrated signal generating unit 6, which is controlled by the unit 5, a signal is supplied to the device under test 1. The image corresponding to the new readout of the device is again recorded in the memory of block 5, which performs element-by-element subtraction of two images. In the process of elementwise subtraction, two arrays are formed according to the following rule: if the sign of the difference between the corresponding elements of two images is less than zero, then the module of this difference is signed in the first array, with a positive sign of subtraction, the value of the difference module is entered in the second array. Two arrays are obtained that determine the zero and current positions of the pointer, after which each element of these arrays is compared with a threshold level. The threshold level is determined by analysis, a histogram of the distribution of the numerical values of the elements, i.e. difference modules.

If the result of comparing the elements of these arrays with the threshold level is positive, then the coordinates of these elements are stored in block 5 processing and control. In this case, two arrays of point coordinates are formed, which determine the initial and current (as a result of moving) position of the middle line of the pointer of the instrument under test 1, as two straight lines. This information is processed by block 5, while the angle of inclination of each straight line corresponding to the midline of the pointer is calculated by the method of least square deviations or in accordance with formula (6). The value of the angle of rotation is calculated by block 5 according to the formula (7). The value of this angle is presented in a previously defined scale equation that relates ∝ and the value of the measured quantity A, read by the pointer. Thus, block 5, using equations (3) - (8), determines the measurement error, which is stored in the memory of block 5 or displayed on a recording device (not shown in Fig. 3) in the form of a verification protocol.

Block 4 operates as follows. Before starting verification, values are determined in the memory block 11 from block 5, which determine the size and position of the input image fragment, as well as the number of input columns and the step between them. After submitting the appropriate command from block 5, the control unit 10 starts to work, which synchronizes the operation of the shaper 12 of the read window. In this case, the previously recorded values are selected from the memory block 11 and written to the counters of the read window generator 12, which determines the scan window and issues control pulses corresponding to this window to the converter 8. The converter 8 generates a digital brightness code for each point of the scanning raster and, using the block 9 pairing transmits them to the block 5 _e , where the processing occurs according to the method described. Amplifier 7 amplifies and normalizes the input signal by level.

The proposed device will allow, in comparison with the known ones [1, 2], to increase the verification performance by eliminating preparatory operations before checking each device, to increase accuracy and noise immunity due to processing several lines of a television raster and using statistical processing methods, to expand the scope, having the ability to verify any arrow devices without entering on the scale of auxiliary information.

Values are recorded that determine the size and position of the input image fragment, as well as the number of input columns and the step between them. After submitting the appropriate command from block 5, the control unit 10 starts to work, which synchronizes the operation of the driver 12 of the read window. In this case, the previously recorded values are selected from the memory unit 11 and written to the counters of the read window generator 12, which determines the scan window and issues control pulses corresponding to this window to the converter 8. The converter 8 generates a digital brightness code for each point of the scanning raster and, using the block 9 pairing transmits them to block 5, where the processing occurs according to the method described. Amplifier 7 amplifies and normalizes the input signal by level.

The proposed device will allow, in comparison with the known one, to increase the verification performance by eliminating preparatory operations before the verification of each device, to increase the accuracy and noise immunity due to processing several lines of a television raster and to apply statistical processing methods, to expand the scope, having the ability to verify any pointer devices without entering on the scale supporting information.

Claims (4)

1. The method of automatic calibration of dial gauges, based on image processing of the indicator part of the measuring instrument, which consists in determining the angle of rotation of the dial gauge of the measuring device and comparing it with the specified values of the angles at which the digitized marks are located, and determining the error when a signal is supplied to the calibrated instrument from a source of reference signals, characterized in that pre-establish a connection between the angle of rotation of the pointer and showing with the help of the instrument, the image of the indicator part of the instrument is scanned according to the minimum number of columns for a given error value, the image corresponding to the zero reading of the instrument is stored, a discretely increasing signal is set and the second image corresponding to the new reading of the instrument is stored, one is subtracted from one another, the numerical values of the elements of two are compared the resulting arrays with a threshold level, determine the angle of two straight lines passing through the points of the corresponding arrays, subtract them, find the angle of rotation, which determine the reading of the pointer device on a pre-established connection and subtract the error. 2. The method according to claim 1, characterized in that to establish a connection between the angle of rotation of the pointer and the readable indication, the angles of rotation of the pointer are determined, when the pointer is combined with digitized marks, they are stored and the scale equation is determined, for example, in the form of a polynomial of the second degree connecting the rotation angle and the reading. 3. A device for automatically checking dial gauges, comprising a transmitting television camera optically connected to a verified instrument, a processing and control unit, the output of which is connected to a calibrated signal generating unit connected to terminals for connecting a calibrated instrument, characterized in that a software-controlled scanning unit, the input of which is connected to the output of the television transmitting camera, and the output to the processing and control unit, and the control input of the grammno-controlled scanning is connected to the second output of the processing and control unit. 4. The device according to claim 3, characterized in that the program-controlled scanning unit comprises a series-connected amplifier, a voltage-code converter, an interface unit, a control unit, a memory unit and a read window driver, the output of which is connected to the second input of the voltage-code converter and the second input is with the second output of the control unit, while the amplifier input is the first input, the input of the memory unit is the control input, and the output of the interface unit is the output of the program-controlled scanning unit.

Images