JP2012104032A - Analog meter reading device and reading method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To use a relatively inexpensive system to reduce wrong recognition of an indicated value due to a pointer shadow in automatic reading of an analog meter using a camera.SOLUTION: A reading device 100 for an analog meter 50 includes: an imaging part 110 for imaging a display part of the analog meter 50 to generate two-dimensional image data; a light projection part 140 for radiating linear visible light on a fixed surface of the display part; and a control part 120 for processing the image data to determine an indicated value indicated by a pointer of the analog meter 50. The control part 120 discriminates between the actual pointer and a pointer shadow on the basis of first one-dimensional data extracted along a virtual line on the display part from the image data and second one-dimensional data extracted along the visible light radiated from the light projection part 140 from the image data, thereby reducing wrong recognition of the indicated value.

Description

  The present invention relates to an analog meter reading apparatus and reading method, and more particularly to a technique for preventing erroneous recognition of a pointer in automatic reading of an analog meter.

  In order to perform maintenance management of electric power facilities and building facilities, it is necessary to collect values of measuring instruments attached to the facilities. However, many of these measuring instruments in existing facilities do not have a digital output terminal. For the value of a measuring instrument without such a digital output terminal, it is necessary to read the meter of the measuring instrument by an inspector during maintenance and inspection.

  In order to reduce the burden of meter reading and data management for maintenance inspectors, efforts are being made to replace these existing measuring instruments with meters equipped with digital output terminals. In order to deal with all of the above, a lot of costs are required for equipment purchase, replacement work, and field tests. In addition, in certain facilities such as electric power facilities, there are facilities that cannot be easily stopped due to the replacement work, and there are cases where time constraints are imposed.

  In order to deal with such a problem, for example, as shown in Japanese Patent Laid-Open No. 2009-75848 (Patent Document 1), an existing analog meter is imaged with a camera, and the captured data is analyzed by analyzing the imaged data. A technique for reading the image is being developed.

JP 2009-75848 A JP 2003-223893 A

  In the case of using the technique disclosed in Japanese Patent Application Laid-Open No. 2009-75848 (Patent Document 1), it is necessary to image the display unit of the analog meter with a camera. Light such as lighting or sunlight is irradiated.

  Since a gap is provided between the pointer of the meter, which is a movable part, and the scale plate, which is a fixed part, the shadow of the pointer is projected onto the scale plate depending on the direction of the irradiated light. As a result, the projected shadow may be misrecognized as a pointer, and the indicated value of the meter may be erroneously read.

  It is possible to properly identify true pointers and shadows by adopting a high-resolution camera and complex image analysis technology to deal with such problems. Equipment is required and the amount of data to be processed increases, so that transmission and processing take time. Therefore, it is often difficult to apply such expensive equipment to many existing equipment from the viewpoint of cost and the like.

  The present invention has been made to solve such a problem, and the object thereof is due to the shadow of the pointer by using relatively inexpensive equipment in automatic reading of an analog meter using a camera. This is to suppress erroneous recognition of the indicated value.

  An analog meter reading device according to the present invention includes a display unit that includes a light projecting unit, an imaging unit, and a control unit, and includes a fixed surface and a movable unit that is displaced along the fixed surface in accordance with a measurement amount. Read the indicated value of the analog meter. The imaging unit images the display unit and generates two-dimensional image data. The control unit is configured so that the light projecting unit overlaps the movable unit when viewed from the imaging unit even if the movable unit is displaced in the measurement target range from a direction having a predetermined angle with the direction in which the imaging unit is opposed to the display unit. The first candidate position of the movable part calculated based on the first one-dimensional data extracted from the image data along the irradiated linear light when linear light is irradiated on the fixed surface The instruction value instructed by the movable part is determined by comparing the second candidate position of the movable part calculated based on the image data.

  Preferably, the control unit determines a first pixel corresponding to the movable unit in the first one-dimensional data based on at least luminance signals of a plurality of pixels constituting the first one-dimensional data, and Based on at least the luminance signals of the plurality of pixels constituting the second one-dimensional data, the second pixel corresponding to the movable part is determined from the second one-dimensional data. And a control part determines the instruction | indication value which a movable part instruct | indicates by comparing a 1st and 2nd pixel.

  Preferably, the control unit calculates a threshold value for each of the first and second one-dimensional data based on a maximum value, a minimum value, and an average value of a plurality of luminance signals constituting the corresponding one-dimensional data. At the same time, the pixel corresponding to the movable part is determined based on the calculated threshold value.

  Preferably, the control unit recognizes a pixel corresponding to the second pixel among the determined first pixels as a pixel corresponding to the movable unit, and the movable unit receives an instruction value corresponding to the recognized pixel. It is determined that the instruction value is instructed.

  Preferably, the second candidate position is extracted from the image data along a virtual line set in advance on the fixed surface so as to overlap with the movable part as viewed from the imaging unit even if the movable part is displaced in the measurement target range. Calculation is performed based on the second one-dimensional data.

  Preferably, the analog meter is a pointer meter in which a movable portion that is a pointer rotates around a rotation center. The virtual line and the linear light are either an arc or a polygon centered on the rotation center.

  Preferably, the analog type meter is a pointer type meter in which a movable portion which is a pointer operates within a fan-shaped operating range around a rotation center. The virtual line and the linear light are either a straight line or a curve that crosses the fan-shaped operating range.

  Preferably, the reading device further includes a communication unit for transmitting the instruction value determined by the control unit to the data collection device.

  A reading method for an analog meter according to the present invention is a reading method for an analog meter including a display unit having a fixed surface and a movable unit that is displaced along the fixed surface in accordance with a measurement amount. Irradiate linear light on the fixed surface so that it overlaps with the movable part when viewed from the imaging part even if the movable part is displaced in the measurement target range from the direction that faces the display part and a predetermined angle. Extracting the first one-dimensional data from the image data along the irradiated linear light, and extracting the first one-dimensional data along the irradiated linear light. Calculating the first candidate position of the movable part based on the first one-dimensional data, calculating the second candidate position of the movable part based on the image data, and the first and second candidates Based on position Te, and a determining instruction value movable unit instructs.

  The step of calculating the second candidate position is based on image data along an imaginary line preset on the fixed surface so as to overlap the movable portion as viewed from the imaging portion even if the movable portion is displaced in the measurement target range. The second one-dimensional data is extracted, and the second candidate position is calculated based on the extracted second one-dimensional data.

  According to the present invention, in automatic reading of an analog meter using a camera, it is possible to suppress erroneous recognition of an instruction value caused by the shadow of a pointer using relatively inexpensive equipment.

1 is an overall block diagram of a reading system including an analog meter reading device according to a first embodiment. It is a figure which shows the 1st example of the image which imaged the pointer type meter. It is a figure which shows the 2nd example of the image which imaged the pointer type meter. It is a figure for demonstrating the influence of the irradiation light in a pointer type meter. It is a figure which shows the example of an image when the pointer type meter is imaged when light is irradiated. 6 is a diagram for explaining a pointer identification technique in the reading apparatus according to Embodiment 1. FIG. It is a figure which shows the example of the image which imaged the pointer type meter of FIG. 2 at the time of applying Embodiment 1. FIG. It is a figure which shows the example of the image which imaged the pointer type meter of FIG. 3 at the time of applying Embodiment 1. FIG. FIG. 6 is a diagram for describing a pointer identification process executed by the reading device in the first embodiment. 5 is a flowchart for illustrating an instruction value determination control process of an analog meter executed by a reading device in the first embodiment. 9 is a flowchart for illustrating an instruction value determination control process of an analog meter executed by a reading device in the second embodiment.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the drawings, the same or corresponding parts are denoted by the same reference numerals and description thereof will not be repeated.

[Embodiment 1]
FIG. 1 is an overall block diagram of a reading system 10 including analog meter reading devices 100, 100A, and 100B according to the first embodiment.

  With reference to FIG. 1, the reading system 10 includes reading devices 100, 100 </ b> A, and 100 </ b> B and a data collection device 200.

  The reading device is provided for each analog meter to be read, and the read data is transmitted to the data collecting device 200 by communication. In FIG. 1, a configuration in which the reading system 10 includes three reading devices 100, 100 </ b> A, and 100 </ b> B will be described as an example. However, the number of reading devices is not limited thereto, and is appropriately provided according to the number of meters to be read. It is done. In the following description, in order to facilitate understanding, details will be described by taking the reading device 100 for reading the indicated value of the analog meter 50 as an example. In FIG. 1, the detailed configuration of the reading devices 100A and 100B is not shown, but the main components are the same as those of the reading device 100, and therefore description thereof will not be repeated.

  The reading device 100 includes an imaging unit 110 for imaging an analog meter, a control unit 120 for processing captured two-dimensional image data, and the processed data to a data collection device 200 that is a central device. A communication unit 130 for transmission and a light projecting unit 140 configured to irradiate the analog meter 50 with predetermined light.

  The imaging unit 110 includes a camera main body 112 and an A / D (Analog-to-digital) conversion unit 114.

  The camera body 112 includes a CCD (Charge Coupled Device) image sensor or a COMS (Complementary Metal Oxide Semiconductor) image sensor, and is disposed at a position where the analog meter 50 to be read can be photographed. The camera body 112 images the analog meter at predetermined time intervals or at a timing according to an instruction from the data collection device 200. Then, the camera body 112 outputs the captured two-dimensional analog image data to the A / D converter 114.

  The A / D converter 114 converts the analog image data received from the camera body 112 into a digital format, and outputs the converted digital image data to the control unit 120.

  The control unit 120 includes a storage unit 122, a data generation unit 124, and a determination unit 126, and processes the image data output from the A / D conversion unit 114.

  The storage unit 122 stores the two-dimensional image data output from the A / D conversion unit 114 in an internal buffer or a storage device (both not shown).

  The data generation unit 124 generates one-dimensional data from the two-dimensional image data stored in the storage unit 122 in order to facilitate reading of the instruction value of the analog meter. In the first embodiment, as will be described later, at least two one-dimensional data along two paths are generated. Here, the one-dimensional data includes a luminance signal and a color difference signal of image data.

  The determination unit 126 receives a plurality of one-dimensional data along the two paths generated by the data generation unit 124. The determination unit 126 determines, for each received one-dimensional data, pixel position candidates corresponding to the pointer of the pointer-type meter based on the magnitude of the signal included therein. Then, the determination unit 126 determines a pixel position corresponding to a pointer estimated to be more likely from comparison of candidate pixel positions determined by a plurality of one-dimensional data. Thereafter, the determination unit 126 determines the instruction value of the pointer corresponding to the finally determined pixel position, and stores the determination result in the storage unit 122.

  The communication unit 130 acquires data related to the indicated value of the pointer stored in the storage unit 122 at a predetermined timing or a timing when a request is received from the data collection device 200, and collects data via the communication line. Transmit to device 200.

  Here, the communication line may be wired communication using electrical or optical wiring, or may be wireless communication. Further, it may be a communication line dedicated to this system, or a public communication network such as a telephone line or the Internet may be used.

  The data collection device 200 includes, for example, a personal computer and a server, and includes a communication unit 210, a control unit 220, a display device 230, an input / output device 240, and a storage device 250.

  The communication unit 210 transmits / receives data to / from a communication unit (for example, the communication unit 130) in each reading device via a communication line.

  The display device 230 is typically a display monitor, and displays the state of each reading device and various data stored in the storage device 250.

  The input / output device 240 includes a keyboard, a mouse, a joystick, a touch panel, and the like, and transmits an operation signal from the user to the control unit 220.

  The storage device 250 stores data related to the indicated value of the meter transmitted from each reading device, and data such as the state of the device, the presence / absence of an abnormality, and its history.

  The control unit 220 processes an operation signal from the user and received data, and controls peripheral devices such as the display device 230 and the storage device 250. The control unit 220 manages the entire reading system 10.

  FIG. 2 is a diagram schematically showing an image 300 obtained by imaging a pointer type meter. The analog meter in FIG. 2 is a type in which the pointer rotates in an arc around the rotation center. In FIG. 2, a thermometer meter is shown as an example.

  Referring to FIG. 2, display unit 310 of the pointer-type meter includes a dial plate 320 on which numerical values and scales are written, and an indicating hand 330 for indicating the measured temperature. In FIG. 2 and the subsequent drawings showing the pointer type meter, the case where there is one pointer will be described as an example for ease of explanation. For example, in the case of the thermometer of FIG. A plurality of other pointers may be included, such as a demand needle that indicates the maximum temperature experienced by the alarm and an alarm setting needle for setting the output level of the alarm signal to the alarm.

  In general, an analog meter displays a measurement amount by moving a movable portion along a fixed surface according to the measurement amount. In the case of the pointer type meter of FIG. 2, the pointer as a movable part rotates around the center of rotation along the dial as a fixed surface. The extraction of the one-dimensional data is performed in consideration of the movable range of the movable part (pointer) in the measurement target range. Specifically, first, an imaginary line that overlaps with the movable part as viewed from the camera main body 112 in FIG. 1 is set on the fixed surface regardless of the position of the movable part in the measurement target range. Next, the data generation unit 124 of FIG. 1 extracts one-dimensional data corresponding to a preset virtual line from the two-dimensional image data.

  In the case of FIG. 2, the virtual line 340 is set in an arc shape centering on the rotation center around which the indicator hand 330 rotates. Accordingly, since the virtual line 340 intersects the indicator needle 330 in the measurement target range, the instruction value of the indicator needle 330 can be read based on the one-dimensional data corresponding to the virtual line 340. Also, by setting the virtual line 340 in an arc shape, the correspondence between each point on the virtual line 340 and the indicated value indicated by the indicating needle 330 becomes a direct proportional relationship. Becomes easier.

  Further, the radius RAD of the arc is set so that the virtual line 340 does not overlap with the characters on the dial 320 so as not to be affected by the characters on the dial 320. Note that the virtual line 340 does not necessarily have to be on an arc, and may be changed to any shape as long as it does not overlap with the characters on the dial 320. For example, the virtual line 340 may be a rectangle. When the characters on the dial 320 and the indicator hand 330 are colored in different colors, the virtual line 340 can be set regardless of the position of the characters on the dial.

  Further, the image data 300 in FIG. 2 is obtained by imaging the display unit 310 with the lens of the camera body 112 in FIG. 1 facing the display unit 310 of the pointer meter 50. On the other hand, when the display unit 310 is imaged from an oblique direction, the virtual line 340 that was on the arc on the actual dial 320 is flattened in an elliptical shape on the captured image 300, and the long axis and the short axis are displayed. The direction of the axis is angularly displaced from the vertical and horizontal directions of the image 300. Therefore, it is necessary to correct the position and shape of the virtual line 340 in the captured image 300 according to the amount of deviation from the position facing the display unit 310. For example, the correction amount of the virtual line 340 can be determined by determining three specific points on the actual dial 320 in advance and examining the positions of the pixels corresponding to the three specific points on the image 300.

  FIG. 3 is a diagram showing an image 300A of another example of the pointer type meter. The pointer type meter of FIG. 3 includes a dial 320A and an indicating hand 330A, as in the pointer type meter of FIG. In the pointer type meter of FIG. 3, the indicator needle 330 operates in a fan shape with respect to the rotation center. The scale of the dial 320A is written on a straight line parallel to the string of the fan-shaped operating area of the indicator hand 330. In such a pointer-type meter, the virtual line 340A corresponding to the virtual line 340 in FIG. 2 is set to a straight line parallel to the scale. Or it is good also as a fan-shaped circular arc shape like the virtual line 340B.

  Next, problems in such a reading system will be described with reference to FIG. Referring to FIG. 4, in the reading system as described above, it is necessary to take an image of the analog meter 50 by the camera main body portion 112, so that generally, the display portion of the meter has sunlight or illumination light. Light is illuminated from the light source 400. In the analog meter 50, a slight gap is usually provided between the dial 52 on which a scale and the like are written and the indicating hand 54. Then, the shadow 56 of the pointing hand 54 is projected on the dial 52 by the light 400 that illuminates the display unit.

  Depending on the positional relationship between the light 400 and the camera body 112, the shadow 56 of the pointing hand 54 may be projected at a position shifted from the true pointing hand 54 on the dial 52. In particular, for example, when the dial 52 is white and the indicating hand 54 is colored black, it is difficult to distinguish the true indicating hand 54 and the shadow 56 in the image data.

  In such a case, an example of image data captured by the camera body 112 is shown in FIG. As shown in FIG. 5, a shadow 350 as if it is another indicator needle may appear on the image data at a position different from the indicator needle 330. Then, in the one-dimensional data generated along the virtual line 340, data that can be candidates for the indicator needle is generated at the true pixel position of the indicator needle 330 and the pixel position of the shadow 350, and the indication value is erroneously set. There is a risk of recognition.

  Therefore, in the reading apparatus according to the first embodiment, identification processing for identifying the true indicator needle and its shadow on the display unit of the meter is performed using a light source different from the light that illuminates the display unit of the meter.

  FIG. 6 is a diagram for explaining a configuration for identifying a true pointer and its shadow in the first embodiment. FIG. 6 is a schematic diagram of device arrangement when viewed from the side of the analog meter 50.

  Referring to FIG. 6, light projecting unit 140 included in reading apparatus 100 is, for example, laser, LED (Light Emitting Diode), or infrared light, and linear light that can be detected by camera body 112 is generated. It is a light source that can be irradiated. The light projecting unit 140 is installed so as to irradiate the dial 52 of the analog meter 50 with light from a direction in which the camera body 112 has a predetermined angle with the direction facing the analog meter 50 to be read. In FIG. 6, the light projecting unit 140 is disposed below the camera main body 112 and irradiates the laser light LGT in the direction of the elevation angle θ with respect to the direction facing the analog meter 50.

  In the analog meter as shown in the example of FIG. 2, for example, as shown in FIG. 7, the locus 360 of the laser light LGT is irradiated so as to be concentric with the virtual line 340.

  At this time, in FIG. 6, the indicator hand 54 is located closer to the camera body 112 than the dial 52, and therefore the locus R <b> 20 of the laser light LGT that appears on the indicator hand 54 is the laser light that appears on the dial 52. The LGT locus R10 is shifted.

  When this is seen in the image data taken by the camera body 112, the locus R10 of the laser beam LGT at the position of the shadow 350 is a circle of the locus 360 of the laser beam LGT on the dial 320 as shown in FIG. Although coincident with the circumference, the locus R20 of the laser beam LGT on the indicating hand 330 is at a position different from the circumference of the locus 360 of the laser beam LGT on the dial 320. That is, when viewed along the locus 360 of the laser beam LGT on the dial 320, the laser beam is not detected at the position of the true pointer 330, but is detected at the position of the shadow 350. Thereby, in the one-dimensional data along the arc to be the locus 360 of the laser light LGT on the dial 320, the luminance signal, the color difference signal, and / or both at the position of the true indicator hand 330 may be other parts. The value can be different from that of. Therefore, by comparing the one-dimensional data generated along the virtual line 340 with the one-dimensional data generated along the light emitted from the light projecting unit 140, the true indicator needle 330 and the shadow 350 thereof are compared. Can be easily identified.

  Also in the case of the analog meter as shown in FIG. 3, the laser light LGT from the light projecting unit 140 is parallel to the straight virtual line 340A (or fan-shaped virtual line 340B) as shown in FIG. The true indicator needle 330A and its shadow 350A can be identified by generating one-dimensional data along the locus 360A of the irradiated laser beam.

  In this way, by using a light source different from the light that illuminates the display unit of the meter and performing identification processing for identifying the true indicator needle and its shadow on the display unit of the meter, an expensive high-resolution camera or a complicated Without adopting an image analysis technique, it is possible to suppress erroneous recognition of the indicated value due to the shadow of the indicating hand at a relatively low cost.

  FIG. 9 is a diagram illustrating an example of the one-dimensional data generated in the first embodiment. The upper part of FIG. 9 shows the luminance signal of the one-dimensional data generated along the virtual line, and the lower part shows the luminance signal of the one-dimensional data generated along the laser beam from the light projecting unit 140. In addition, the horizontal axis of FIG. 9 indicates the pixel number along the one-dimensional data, and the position of this pixel corresponds to the indication value of the meter. For example, in the example of the thermometer meter in FIG. 2, the pixel number “150” indicates 50 ° C., and the pixel number “100” indicates 40 ° C.

  Referring to FIG. 9, in the one-dimensional data along the virtual line by the normal light in the upper stage, the luminance signal at the positions of pixel numbers N1 and N2 is larger than a predetermined threshold value TH1, and this pixel It is determined as a candidate position where there is a possibility that the pointing hand exists at the position of the number.

  On the other hand, in the one-dimensional data along the laser beam from the lower light projecting unit 140, the luminance signal of the laser beam at the pixel number N1 is below a predetermined threshold value TH2, and at this position the laser beam is not Not detected. On the other hand, in the pixel number N2, no decrease in luminance signal is observed, and laser light is detected. That is, as described in FIGS. 7 and 8, the pixel number N1 can be identified as the pixel position corresponding to the true pointer, and the pixel number N2 can be identified as the pixel position corresponding to the shadow of the pointer.

  The details of the above threshold value TH1, TH2 setting method are described in Japanese Patent Laid-Open No. 2009-75848 (Patent Document 1) and the like, but will not be repeated. For example, the generated one-dimensional data is configured. The maximum value MAX, the minimum value MIN, and the average value AVE may be calculated for the luminance signals of a plurality of pixels to be used, and the threshold value TH may be calculated using Equation (1) using these values.

TH = (MAX−MIN) · K1 + AVE (1)
Here, K1 is a coefficient that affects the noise resistance, and is set based on the color of the pointer, the dial, the condition of the irradiation light, and the like.

  FIG. 9 shows an example in which the determination is performed based on the luminance signal. For example, when the laser light emitted from the light projecting unit has a specific color such as a red laser, the luminance signal is used instead. Alternatively, by using the color difference signal in addition to the luminance signal, it may be possible to specify the pixel position of the pointer more accurately.

  FIG. 10 is a flowchart for explaining the details of the instruction value determination control process executed by the reading device in the first embodiment.

  Referring to FIGS. 1 and 10, in step (hereinafter, step is abbreviated as S) 100, laser light is emitted from light projecting unit 140 onto analog meter 50.

  Next, in S110, the camera main body 112 captures an image of the meter display. The captured two-dimensional image data is converted into digital data by the A / D conversion unit 114 (S120), and the converted data is stored in the storage unit 122 (S130).

  In S140, the data generation unit 124 generates one-dimensional data of a luminance signal along the virtual line based on the image data stored in the storage unit 122. Then, using the generated one-dimensional data, the determination unit 126 estimates a pixel position IMG1 (n) that is a candidate for the pointing needle.

  Further, in S150, the data generation unit 124 generates one-dimensional data of the luminance signal along the locus on the dial of the laser light emitted from the light projecting unit 140 based on the image data stored in the storage unit 122. Generated. Then, using the generated one-dimensional data, the determination unit 126 estimates a pixel position IMG2 (n) that is a candidate for the pointing needle.

  Next, at S160, the determination unit 126 compares the estimated pixel positions IMG1 (n) and IMG2 (n) to determine a more likely pixel position as the true indicator needle position. .

  Thereafter, in S170, the determination unit 126 converts the determined pixel position into a corresponding instruction value, and the converted instruction value is stored in the storage unit 122.

  In S180, at a predetermined timing, the communication unit 130 transmits data related to the instruction value stored in the storage unit 122 to the data collection device 200.

  By performing control according to the above processing, it is possible to identify the true pointer and its shadow in automatic reading of an analog meter using a camera. The method disclosed in this specification does not require a high-resolution camera and complicated image analysis technology, and therefore, using a relatively inexpensive facility, suppresses misrecognition of an instruction value caused by the shadow of the pointer. Can do.

  As described above, in the above-described embodiment, the case where there is one pointer has been described as an example. However, the technique of the first embodiment is also applied to an analog meter having a plurality of pointers having different colors. Thus, it is possible to eliminate the influence of the shadows of a plurality of pointers. In that case, by using the color difference signal in addition to the luminance signal, it becomes easy to specify the pixel position of each pointer in the obtained one-dimensional data.

[Embodiment 2]
In the first embodiment, the candidate position of the pointer calculated from the one-dimensional data extracted along the virtual line and the candidate of the pointer calculated from the one-dimensional data extracted along the light emitted from the light projecting unit A configuration has been described in which the true pointer and its shadow are identified by comparing the position. However, the other candidate position to be compared with the candidate position of the pointer calculated from the one-dimensional data extracted along the light emitted from the light projecting unit is not limited to the above-described method, and other methods are used. Even when the candidate position is calculated, the technique presented in this application can be applied.

  For example, as disclosed in Japanese Patent Application Laid-Open No. 2003-223893 (Patent Document 2), extracted data obtained by extracting an image of a pointer part from two-dimensional image data and reference image data in which the position of the pointer is specified The candidate position may be calculated by pattern matching. A specific method of pattern matching is disclosed in Japanese Patent Application Laid-Open No. 2003-223893 (Patent Document 2) and will not be repeated here. However, even in such pattern matching, binary imaged image data is also used. In the conversion process, the shadowed part of the pointer may be misrecognized. Therefore, it is effective to identify the true pointer and its shadow as compared with the candidate position using the light emitted from the light projecting unit as described above.

  In this case, the data generation unit 124 in FIG. 1 generates the extracted data and the reference image data. Then, the decision unit 126 calculates the candidate position of the pointer by pattern matching these data, and the calculated candidate position is extracted from the one-dimensional data extracted along the light emitted from the light projecting unit. It is compared with the calculated candidate position of the pointer, and the indication value based on the true pointer is determined based on the comparison result.

  FIG. 11 is a flowchart for explaining the instruction value determination control process of the analog meter executed by the reading device in the second embodiment.

  Referring to FIGS. 1 and 11, in S300, laser light is emitted from light projecting unit 140 to analog meter 50.

  Next, in S <b> 310, the display unit of the meter is imaged by the camera body 112. The captured two-dimensional image data is converted into digital data by the A / D conversion unit 114 (S320), and the converted data is stored in the storage unit 122 (S330).

  In S340, the data generation unit 124 extracts data indicating the pointer portion (extraction data) based on the image data stored in the storage unit 122, and generates reference image data for comparison with the extraction data. Is done. Then, using these data, the determination unit 126 estimates the candidate position POS1 (n) of the pointing hand.

  Further, in S350, the data generation unit 124 generates one-dimensional data of the luminance signal along the trajectory on the dial of the laser light emitted from the light projecting unit 140 based on the image data stored in the storage unit 122. Generated. Then, using the generated one-dimensional data, the determination unit 126 estimates the candidate position POS2 (n) of the pointing needle.

  Next, in S360, the determination unit 126 determines the position of the true pointer that seems to be more likely by comparing the estimated candidate positions POS1 (n) and POS2 (n).

  Thereafter, in S370, the determination unit 126 converts the determined candidate position into a corresponding instruction value, and the converted instruction value is stored in the storage unit 122.

  In S380, the communication unit 130 transmits data related to the instruction value stored in the storage unit 122 to the data collection device 200 at a predetermined timing.

  By performing control according to the above processing, it is possible to identify a true pointer and its shadow in automatic reading of an analog meter using pattern matching.

  The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

  10 Reading System, 50 Analog Meter, 52, 320, 320A Dial, 54 Indicator Hand, 56, 350, 350A Shadow, 100, 100A, 100B Reading Device, 110 Imaging Unit, 112 Camera Body, 114 A / D Conversion Unit, 120, 220 control unit, 122 storage unit, 124 data generation unit, 126 determination unit, 130, 210 communication unit, 140 projector unit, 200 data collection device, 230 display device, 240 input / output device, 250 storage device, 300, 300A image, 310, 310A display unit, 330, 330A indicator needle, 340, 340A, 340B virtual line, 360, 360A locus, 400 light source, LGT laser light.

Claims (11)

A reading apparatus for an analog meter including a display unit having a fixed surface and a movable unit that is displaced according to a measurement amount along the fixed surface,
A light projecting unit;
An imaging unit for imaging the display unit to generate two-dimensional image data;
Even if the movable part is displaced in a measurement target range from a direction having a predetermined angle with a direction in which the imaging unit is opposed to the display unit, the light projection is performed so as to overlap the movable unit when viewed from the imaging unit. The movable portion calculated based on the first one-dimensional data extracted from the image data along the irradiated linear light when the linear light is irradiated on the fixed surface from the portion A control unit for determining an instruction value instructed by the movable part by comparing the first candidate position of the movable part and the second candidate position of the movable part calculated based on the image data. An analog meter reader provided.   The second candidate position is along the imaginary line preset on the fixed surface so as to overlap the movable part when viewed from the imaging unit even if the movable part is displaced in the measurement target range. The analog meter reading device according to claim 1, wherein the analog meter reading device is calculated based on second one-dimensional data extracted from image data.   The control unit discriminates a first pixel corresponding to the movable unit from the first one-dimensional data based on at least luminance signals of a plurality of pixels constituting the first one-dimensional data. Determining a second pixel corresponding to the movable part in the second one-dimensional data based at least on luminance signals of a plurality of pixels constituting the second one-dimensional data; and The analog meter reading device according to claim 2, wherein an instruction value indicated by the movable part is determined by comparing second pixels.   The control unit calculates a threshold value for each of the first and second one-dimensional data based on a maximum value, a minimum value, and an average value of a plurality of luminance signals constituting the corresponding one-dimensional data. The analog meter reading device according to claim 3, wherein a pixel corresponding to the movable portion is determined based on the calculated threshold value.   The control unit recognizes a pixel corresponding to the first pixel among the determined second pixels as a pixel corresponding to the movable portion, and an instruction value corresponding to the recognized pixel is movable. The analog meter reading device according to claim 4, wherein the reading is determined to be an instruction value indicated by the unit. The analog meter is a pointer-type meter in which the movable part that is a pointer rotates around a rotation center,
The analog meter reading device according to claim 2, wherein the virtual line and the linear light are any one of an arc and a polygon centered on the rotation center. The analog meter is a pointer-type meter in which the movable part, which is a pointer, operates within a fan-shaped operating range around the center of rotation.
6. The analog meter reading device according to claim 2, wherein the virtual line and the linear light are any one of a straight line and a curve that traverse the fan-shaped operating range. 7.   The second candidate position is calculated by pattern matching between extracted data representing the movable part extracted from the image data and reference image data in which the position of the movable part is specified. Analog meter reader. The reader is
The analog meter reading device according to claim 1, further comprising a communication unit configured to transmit the instruction value determined by the control unit to a data collection device. A method for reading an analog meter including a display unit having a fixed surface and a movable unit that is displaced according to a measurement amount along the fixed surface,
The fixed surface such that the imaging unit overlaps the movable unit when viewed from the imaging unit even if the movable unit is displaced in a measurement target range from a direction having a predetermined angle with a direction facing the display unit. Irradiating linear light on the top;
Imaging the display unit with the imaging unit to generate two-dimensional image data;
The first one-dimensional data is extracted from the image data along the irradiated linear light, and the first candidate position of the movable part is calculated based on the extracted first one-dimensional data. And steps to
Calculating a second candidate position of the movable part based on the image data;
And a step of determining an instruction value indicated by the movable part based on the first and second candidate positions.   The step of calculating the second candidate position includes a virtual line preset on the fixed surface so as to overlap with the movable part when viewed from the imaging unit even if the movable part is displaced in the measurement target range. 11. The analog meter according to claim 10, wherein the second one-dimensional data is extracted from the image data along the line and the second candidate position is calculated based on the extracted second one-dimensional data. Reading method.

Images