WO2020208849A1

WO2020208849A1 - Analog meter reading device

Info

Publication number: WO2020208849A1
Application number: PCT/JP2019/042952
Authority: WO
Inventors: 黒澤　智明; 賀昭柴田
Original assignee: 株式会社オクタテック
Priority date: 2019-04-12
Filing date: 2019-10-31
Publication date: 2020-10-15
Also published as: JP2020173733A; JPWO2020208849A1; JP7423869B2; JP6667771B1

Abstract

This analog meter reading device is provided with a reading unit which includes: light emitting elements; an imaging device that detects, by means of a plurality of pixel sensors, the reflected light resulting from illumination of the pointer needle or scale plate of a pointer needle meter with light emitted from the light emitting elements; and a calculation processing unit that calculates the output signal of each of a plurality of virtual sensors using the output signal of one or more pixel sensors constituting each of the plurality of virtual sensors, wherein the plurality of virtual sensors are virtually arranged within the imaging device and outside or inside relative to the direction in which the numerals displayed on the scale plate of the pointer needle meter are aligned, and wherein index numbers are assigned to the plurality of virtual sensors sequentially in the arrangement direction. The analog meter reading device is further provided with an index number detection unit which detects the index number substantially corresponding to the current position of the pointer needle, on the basis of the output signal of at least one virtual sensor, from among the output signals of the plurality of virtual sensors as calculated by the calculation processing unit.

Description

Analog meter reader

The present invention relates to an analog meter reader.

Conventionally, there has been disclosed a technique capable of automatically reading a value indicated by an indicator needle of an analog meter (hereinafter referred to as "indicator value") by image processing and obtaining a digital signal suitable for computer management (Patent Document). 1. Patent Document 2).

The technology of Patent Document 1 uses a camera installed at a distance from the analog meter, and as images taken by the analog meter, a reference image having a known indicated value and a measured image in which the indicator needle is rotated. The two images of, and are obtained, and the indicated value indicated by the indicator needle is obtained by comparing the two images.

In the technique of Patent Document 2, as another method of reading the indicated value of the analog meter using a camera, a circle whose center coincides with the rotation center of the indicator needle is set for the captured image of the analog meter, and the circle is on the circle. The indicated value is output by obtaining the angle "θ" pointed by the indicator needle from the presence or absence of an arc in the portion where the pixel existing in the indicator overlaps with the indicator needle, or from the length comparison of a plurality of the arcs.

Further, there is disclosed a technique that can be retrofitted to an analog meter, can read the indicated value indicated by the indicator needle without impairing the calibration state of the analog meter, and can take out the indicated value to the outside without contact (Patent Document 3). ).

The technique of Patent Document 3 detects the relative position of the indicator needle with respect to the proximity sensor by attaching a conductive target to the indicator needle and attaching a measurement IC tag unit including a proximity sensor to the cover glass surface of the analog meter. By doing so, the indicated value is output.

Japanese Unexamined Patent Publication No. 2004-133560 Japanese Unexamined Patent Publication No. 2016-162412 JP-A-2017-203775

The present invention solves the above-mentioned problem of disclosure technology.

The analog meter reading device according to the present invention shows a scale plate on which numerical values assigned to a plurality of scales are displayed and an indicated value of the scale plate, and is on a plane at a predetermined distance from the display surface side of the scale plate. An analog meter reading device that reads the indicated value of an indicator needle meter having an indicator needle that rotates with respect to a predetermined center of rotation and a transparent member that covers the display surface side of the scale plate so as to include the indicator needle. A light emitting element, an image pickup element that detects light emitted from the light emitting element and reflected by the indicator needle of the indicator needle meter or the scale plate by a plurality of pixel sensors, and the indicator needle in the image pickup element. A plurality of virtual sensors are virtually arranged outside or inside the arrangement direction of the numerical values displayed on the scale plate of the meter, and index numbers are assigned to the plurality of virtual sensors in the order of the arrangement direction. A reading unit including an arithmetic processing unit that calculates an output signal of each of the plurality of virtual sensors using the output signals of one or more pixel sensors constituting each of the virtual sensors, and a reading unit including the arithmetic processing unit. An index number detection unit that detects a substantial index number corresponding to the position of the indicator needle based on the output signal of at least one virtual sensor among the output signals of the plurality of virtual sensors. It has.

The present invention can read the indicated value of the analog meter with high accuracy by simply attaching it to the analog meter.

FIG. 1A is a front view of the analog meter reading device of the first embodiment attached to the indicator needle meter, and FIG. 1B is an external side view thereof. FIG. 2 is a diagram showing a reading surface of the reading unit. FIG. 3 is a block diagram showing a functional configuration of the analog meter reader. FIG. 4 is a flowchart showing a rotation center identification routine of the indicator needle and a conversion table creation routine. FIG. 5 is a diagram showing an example of the scale area of the indicator needle meter. FIG. 6 is a diagram showing a scale region developed on a polar coordinate plane. FIG. 7 is a diagram showing a conversion table. FIG. 8 is a flowchart showing a reading processing routine. FIG. 9 is a schematic view when the indicator needle of the indicator needle meter is located in the center of one detection region of the photodiode, and is a diagram showing an output signal of each photodiode at that time. FIG. 10 is a schematic view when the indicator needle of the analog meter exists in the overlapping portion of the detection regions of the two photodiodes, and is a diagram showing the output signal of each photodiode at that time. FIG. 11 is a diagram showing a cross-sectional shape of the photodiode according to the second embodiment. FIG. 12 (A) is a schematic view showing a light-sensitive region of a photodiode when the photodiode does not have a lens, and FIG. 12 (B) is a schematic diagram showing a light-sensitive region when a lens is formed on the photodiode. Is. FIG. 13 is a diagram showing light sensitive regions having different diameters. FIG. 14A is a front view of the analog meter reading device of the third embodiment attached to the indicator needle meter, and FIG. 14B is an external side view thereof. FIG. 15A is a view showing a reading surface of the reading unit, and FIG. 15B is an external side view thereof. FIG. 16 is a side view of a main part of the camera module. FIG. 17 is a block diagram showing a functional configuration of the analog meter reader. FIG. 18 is a flowchart showing a rotation center identification routine. FIG. 19 is a diagram showing an example of the scale area of the indicator needle meter. FIG. 20 is a flowchart showing a virtual PD setting routine. FIG. 21 is a diagram showing a virtual PD set in the effective area of the meter image. FIG. 22 is a diagram showing an arrangement position of the virtual PD. FIG. 23 is a diagram showing an example of a virtual PD. FIG. 24 is a flowchart showing a conversion table creation routine. FIG. 25 is a diagram showing a scale region developed on a polar coordinate plane and a series of virtual PDs. FIG. 26 is a flowchart showing a reading processing routine. FIG. 27 is a flowchart showing a signal reading routine.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
[First Embodiment]
FIG. 1A is a front view of the analog meter reading device 1 of the first embodiment attached to the indicator needle meter 100, and FIG. 1B is an external side view thereof. The analog meter reading device 1 irradiates the indicator needle meter 100 with light to detect the position information of the indicator needle 101, and the reading unit 10 and the indicated value indicated by the indicator needle 101 based on the position information detected by the reading unit 10. It includes a control unit 30 for calculating.

The indicator needle meter 100 includes an indicator needle 101, a scale plate 102 that displays a physical quantity (numerical value) corresponding to a position indicated by the indicator needle 101, and a cover glass 103 for protecting the indicator needle 101. The indicator needle 101 rotates with respect to a predetermined rotation center on a plane at a predetermined distance from the display surface side of the scale plate 102. The cover glass 103 covers the display surface side of the scale plate 102 so as to include the indicator needle 101.

The scale plate 102 of the indicator needle meter 100 is white, and the indicator needle 101 is black. Therefore, the light emitted to the indicator needle meter 100 is reflected by the scale plate 102 and absorbed by the indicator needle 101. The reading unit 10 detects the position information of the indicator needle 101 of the indicator needle meter 100 by utilizing the light reflection characteristic of the indicator needle meter 100.

The reading unit 10 is an arc strip-shaped (doughnut-shaped) member having a substantially uniform thickness as a whole, a hole portion formed in the center, and a predetermined width in the radial direction. One of the two side surfaces of the reading unit 10 is directly mounted on the cover glass 103 surface of the indicator needle meter 100, and the side surface (reading surface) that optically reads the position information of the indicator needle 101 of the indicator needle meter 100. Is. The other side surface of the above two side surfaces is a side surface (marker surface) exposed to the outside and on which the marker 10M is printed.

The reading surface of the reading unit 10 is the cover glass 103 surface on the opposite side of the scale plate 102 of the indicator needle meter 100 so that the center of the reading unit 10 coincides with the rotation center of the indicator needle 101 of the indicator needle meter 100. Mounted directly on top. At this time, the marker surface of the reading unit 10 is on the front side. In the present embodiment, four rectangular markers 10M are printed on the marker surface of the reading unit 10. The four markers 10M are arranged at the positions of the vertices of the square, and the intersections of the diagonal lines of the square coincide with the center of the reading unit 10.

FIG. 2 is a diagram showing a reading surface of the reading unit 10. The reading unit 10 includes a disk-shaped storage case 12 in which a small hole portion 11 is formed, a plurality of photodiodes 13 arranged in an arc shape along the outer peripheral portion of the storage case 12, and a plurality of photodiodes and holes. A plurality of LED elements 14 arranged in an arc shape between the portions are provided.

The storage case 12 has a circular storage bottom 12a in which a small hole 11 is formed in the center and a plurality of photodiodes 13 and LED elements 14 are installed, and an outer peripheral wall 12b formed along the outer peripheral portion of the circular storage bottom 12a. And an inner peripheral wall 12c formed along the inner peripheral portion of the circular storage bottom 12a. The height of the outer peripheral wall 12b and the inner peripheral wall 12c is larger than the height from the arrangement surface of the photodiode 13 and the LED element 14. Therefore, when the reading unit 10 is attached to the cover glass 103, the ends of the outer peripheral wall 12b and the inner peripheral wall 12c are adhered to the cover glass 103. That is, the photodiode 13 and the LED element 14 do not have to come into contact with the cover glass 103, and damage due to contact is avoided.

Twenty-one photodiodes 13 from index number 0 to index number 20 are arranged in an arc shape on the reading unit 10. The index number is used not only to identify one photodiode 13 out of the 21 photodiodes 13, but also as position information of the corresponding photodiode 13. Therefore, the index number is used when calculating the indicated value of the indicator needle meter 100, which will be described in detail later.

Further, the relative positional relationship between each photodiode 13 in the reading unit 10 and each marker 10M printed on the marker surface of the reading unit 10 is predetermined. Therefore, when the reading unit 10 is attached to the indicator needle meter 100, if the position of each marker 10M with respect to the indicator needle meter 100 is determined, the position of each photodiode 13 with respect to the indicator needle meter 100 is also uniquely determined.

The plurality of LED elements 14 are arranged in an arc shape inside a concentric circle with respect to the plurality of photodiodes 13. In this embodiment, the number of photodiodes 13 is larger than the number of LED elements 14. However, the number of each of the photodiode 13 and the LED element 14 is not particularly limited.

Further, in the present embodiment, the photodiode 13 is arranged outside the LED element 14, but the photodiode 13 may be arranged inside the LED element 14.

FIG. 3 is a block diagram showing a functional configuration of the analog meter reading device 1. The reading unit 10 and the control unit 30 transmit and receive data to and from each other according to a predetermined protocol. The reading unit 10 detects the position information of the indicator needle 101 of the indicator needle meter 100, and supplies the detected position information to the control unit 30. The control unit 30 calculates the indicated value of the indicator needle meter 100 based on the position information supplied from the reading unit 10, and transmits the indicated value to the server device 200 by wireless communication. As a result, the server device 200 aggregates and manages the indicated values of the plurality of indicator needle meters 100 in real time.

The reading unit 10 controls switching between a plurality of photodiodes 13 that receive the reflected light from the indicator needle meter 100, a plurality of LED elements 14 that irradiate the indicator needle meter 100 with light, and an output signal of each photodiode 13. It is provided with a photodiode control unit 15 and an LED drive unit 16 for driving each LED element 14.

The control unit 30 analog-to-digital / outputs the output signals from the storage battery 31, the photodiode (PD) switching unit 32, and the PD switching unit 32, which store power for supplying power to each circuit in the control unit 30 and the reading unit 10. An A / D converter 33 that performs digital (A / D) conversion, a CPU 34, an LED control unit 35 that controls lighting / extinguishing of the LED element 14, a RAM 36 that is a data work area, and a ROM 37 that stores a program. A wireless communication unit 38 that wirelessly communicates with an external device such as a server device 200.

The analog meter reading device 1 configured as described above calculates the indicated value of the indicator needle meter 100 according to the following procedure.

The reading unit 10 detects the position information of the indicator needle 101 of the indicator needle meter 100 in a state of being mounted on the front center portion of the indicator needle meter 100. Ideally, the reading unit 10 is preferably located at a position not deviated in the radial direction and the rotational direction with respect to the rotation center of the indicator needle 101 of the indicator needle meter 100. However, in reality, the reading unit 10 is mounted at a position deviated to some extent in the radial direction and the rotational direction described above. Therefore, the following processes (1) and (2) are required.

(1) In order to accurately calculate the indicated value of the indicator needle meter 100, it is necessary to determine in advance whether or not the deviation in the radial direction and the rotational direction is within the allowable range.
(2) The position information of the indicator needle 101 detected by the reading unit 10 is merely the position information of the indicator needle 101 of the indicator needle meter 100 based on the index number of the photodiode 13, and the actual indicated value can be calculated. , It is necessary to create a conversion table for converting the position information to the indicated value.

Therefore, specifically, the following processing is performed.
First, the user attaches the reading unit 10 of the analog meter reading device 1 to the front center portion of the indicator needle meter 100.

Next, the user uses, for example, a mobile terminal to take a picture of the indicator needle meter 100 to which the reading unit 10 is mounted. The image of the indicator needle meter 100 generated by photographing (hereinafter, referred to as “meter image”) is transmitted to the server device 200 that manages each indicator needle meter 100.

The CPU 201 of the server device 200 rotates the indicator needle 101 for determining whether or not the above-mentioned radial and rotational deviations of the reading unit 10 are within an allowable range based on the meter image transmitted from the mobile terminal. Identify the center. Further, the CPU 201 of the server device 200 creates a conversion table for converting the position information of the indicator needle 101 of the indicator needle meter 100 into the indicated value based on the index number of the photodiode 13. Here, the conversion table refers to a table for converting the position information of the indicator needle 101 of the indicator needle meter 100 into the indicated value of the indicator needle meter 100.

FIG. 4 is a flowchart showing a rotation center identification routine and a conversion table creation routine of the indicator needle 101.

In step S1, the CPU 201 determines whether or not a meter image has been acquired from the mobile terminal that captured the indicator needle meter 100, and waits until the meter image is acquired. When the CPU 201 acquires the meter image, the CPU 201 proceeds to the next step S2.

In step S2, the CPU 201 corrects the distortion caused by the shooting direction and obtains a meter image equivalent to the meter image generated when the mobile terminal shoots the indicator needle meter 100 from the front. The process of correcting the distortion caused by the photographing direction corresponds to, for example, the technique disclosed in Japanese Patent Application Laid-Open No. 2006-12133.

In step S3, the CPU 201 performs a binarization process on the meter image. Specifically, the CPU 201 extracts a luminance component from the meter image, and for each pixel of the meter image, if the value of each luminance component is larger than a predetermined threshold value, it sets "1", otherwise it sets "0". Set.

In step S4, the CPU 201 extracts the scale area MR1 (see FIG. 5 described later) from the binarized meter image. Here, the scale area means an area on the scale plate 102 on which a scale showing the correspondence between the position indicated by the indicator needle 101 and the indicated value is drawn.

FIG. 5 is a diagram showing an example of the scale area MR1 of the indicator needle meter 100. In the present embodiment, the scale region MR1 is described using four parameter values: an outer radius a, an inner radius b, a central angle φ− to the left of the perpendicular line of the indicator needle meter 100, and a central angle φ + to the right. Modeled with an arc band template. From this, the center position of the scale region MR1 becomes the same as the rotation center of the indicator needle 101.

The template of the scale area MR1 is not limited to the shape of the arc band specified by the four parameters (a value, b value, φ- value and φ + value). That is, the template of the scale region MR1 may have a shape other than the arc band as long as it can identify the rotation center of the indicator needle 101.

In step S4, the CPU 201 further changes the center position of the template and the four parameter values to search for the optimum template position in the binarized meter image. Then, when the pixel value distribution in the template becomes uniform, the CPU 201 extracts the scale region MR1 according to the center position of the template and the four parameter values.

Specifically, when searching for a template, the CPU 201 divides the internal area of the template into several subregions, and the average pixel value in each subregion becomes substantially constant over the entire interior of the template. In addition, the scale region MR1 is extracted as the center position, a value, b value, φ− value, and φ + value at that time.

In step S5, the CPU 201 identifies the rotation center of the indicator needle 101 using the scale region MR1 (a value, b value, φ- value and φ + value) extracted in step S4. In the present embodiment, the center of rotation of the indicator needle 101 coincides with the center position of the arc band-shaped template that models the scale region MR1. Therefore, the CPU 201 identifies the rotation center of the indicator needle 101 by specifying the center position of the template of the scale region MR1 extracted in step S4.

In step S6, the CPU 201 detects the deviation of the mounting position of the reading unit 10 with respect to the rotation center of the indicator needle 101 based on the meter image. Here, the deviation of the mounting position of the reading unit 10 is defined by the translation deviation Δr from the rotation center of the indicator needle 101 and the rotation deviation Δθ of the reading unit 10 from the vertical line of the indicator needle meter 100.

The translation deviation Δr and the rotation deviation Δθ are calculated based on four rectangular markers 10M printed in advance at predetermined positions on the marker surface of the reading unit 10. Specifically, the CPU 201 calculates the position information of the center of the reading unit 10 from the position information of the four markers 10M in the meter image, and translates the translation deviation Δr as the distance from the position information to the rotation center of the indicator needle 101. Is calculated. Further, the CPU 201 calculates the rotation deviation Δθ as the slope of a straight line connecting the two markers 10M arranged in the vertical direction or the horizontal direction with reference to the above-mentioned perpendicular line.

In the present embodiment, as a marker for detecting the deviation of the mounting position of the reading unit 10, a rectangular marker 10M pre-printed at a predetermined position on the marker surface of the reading unit 10 is used, but the present invention is limited to this. It's not something. For example, when a reading unit 10 is attached to each of the plurality of indicating needle meters 100 and a unique identifier is assigned to each reading unit 10, a QR code (registered trademark) may be used as the marker. The QR code (registered trademark) is a two-dimensional bar code in which a unique identifier is encoded and is printed at a predetermined position on the marker surface of the reading unit 10. On the other hand, the CPU 201 may detect the deviation of the mounting position of the reading unit 10 by using the markers at the three corners of the QR code (registered trademark) in the meter image. In this case, the CPU 201 not only detects the deviation of the mounting position of the reading unit 10, but also can acquire the unique identifier of the reading unit 10.

In step S7, the CPU 201 determines whether or not the deviation of the mounting position of the reading unit 10 is within the allowable range. Specifically, the CPU 201 determines whether or not both the translation deviation Δr and the rotation deviation Δθ are equal to or less than a predetermined threshold value. Then, if both the translation deviation Δr and the rotation deviation Δθ are equal to or less than a predetermined threshold value, the process proceeds to step S9, and if at least one of the translation deviation Δr and the rotation deviation Δθ is not equal to or less than the predetermined threshold value, the process proceeds to step S8.

The respective threshold values of the translation deviation Δr and the rotation deviation Δθ are determined by the number of photodiodes 13 and LED elements 14 mounted on the reading surface of the reading unit 10, the arrangement location, and the indicator needle meter 100 required for the reading unit 10. It is determined by the reading accuracy and is not a particularly limited value.

In step S8, the CPU 201 performs a process for instructing the user to reattach the reading unit 10. The process for instructing reattachment corresponds to, for example, a process of sending an e-mail instructing reattachment to a pre-registered e-mail address, or a process of displaying an instruction for reattachment on a predetermined monitor screen. .. Then, this routine ends.

On the other hand, in step S9, the CPU 201 executes the polar coordinate conversion process of the position information of the scale region MR1 and the photodiode 13 shown in FIG.
Specifically, the CPU 201 expands the arc band-shaped scale region MR1 extracted from the meter image into a polar coordinate plane whose horizontal axis is the central angle with respect to the perpendicular line of the indicator needle meter 100 and whose vertical axis is the radius method. To do.

FIG. 6 is a diagram showing a rectangular scale region MR2 developed on a polar coordinate plane. The scale region MR2 is represented by a rectangular region having a width from φ− to φ + on the horizontal axis and a height from a to b on the vertical axis.

The plurality of photodiodes 13 arranged in the reading unit 10 do not appear in the meter image. However, the position information of each photodiode 13 is associated with the position information of the four markers 10M printed on the reading unit 10. That is, if the position information of the four markers 10M is detected, the position information of each photodiode 13 is also detected.

Therefore, the CPU 201 detects the position information of each of the four markers 10M using the meter image, detects the position information of each photodiode 13 based on the position information, and obtains the position information of these photodiodes 13. Expand to a polar plane.

For example, when 21 photodiodes 13 are arranged in the reading unit 10 at equal intervals in an arc shape, as shown in FIG. 6, a rectangular scale region MR2 is arranged on the polar coordinate plane described above, and the scale is concerned. Twenty-one photodiodes 13 (0th to 20th) are arranged at equal intervals along the longitudinal direction of the region MR2. At this time, the tenth position information (intermediate position) of the photodiode 13 corresponds to the position information of the rotation deviation Δθ = 0 ((φ +) = (φ−) = 0). This indicates a state in which the reading unit 10 is mounted without deviation with respect to the rotation center of the indicator needle 101 of the indicator needle meter 100. That is, the deviation of the mounting position of the reading unit 10 is within the allowable range and can be ignored in practice (Δr, Δθ = 0).

In step S10, the CPU 201 creates a conversion table for converting the position information (substantial index number) of the indicator needle 101 of the indicator needle meter 100 based on the index number of the photodiode 13 into the indicated value. Specifically, the CPU 201 recognizes the scale numbers in the vicinity of the scale region MR2 as numerical values, and executes a process of associating the recognized numerical values with the position information of the scale numbers on the polar coordinate plane.

In this embodiment, a known OCR (Optical Character Recognition) technology is used. Further, the position information of the scale numbers is converted into a polar coordinate value θ with reference to the perpendicular line of the indicator needle meter 100. Then, the position information (polar coordinate value θ) of the scale numbers is associated with the numerical values of the scale numbers. It should be noted that such association is performed for all the scale numbers in the vicinity of the scale region MR2.

In general, the scale number is often displayed only as a value corresponding to a typical rotation angle of the indicator needle 101. From this, the position information of the scale numbers is not limited to the value of the position information as it is detected from the recognition of the scale numbers, and is deviated by 5 °, 10 °, 15 °, 30 °, and 45 ° with respect to the above-mentioned perpendicular line. It may be replaced with a value (a typical divisor of 90 ° or a multiple thereof).

Finally, the CPU 201 creates a conversion table that converts the position information of the indicator needle 101 of the indicator needle meter 100 based on the index number of the photodiode 13 into an indicated value by using all the information developed on the polar coordinate plane.

FIG. 7 is a diagram showing a conversion table. The conversion table is composed of an index number of the photodiode 13 corresponding to the position information of the indicator needle 101 of the indicator needle meter 100, a rotation angle based on the perpendicular line of the indicator needle meter 100, and an instruction value corresponding to the rotation angle. ing.

Here, the index number of the photodiode 13 and the rotation angle with respect to the perpendicular line described above are values predetermined by mounting on the reading unit 10.

On the other hand, a plurality of typical rotation angles and indicated values (recognized numerical values) corresponding to them are acquired by the above-mentioned processing. Further, other rotation angles and the indicated values corresponding to them are obtained by linear interpolation using the acquired rotation angles and the indicated values corresponding to them.

Therefore, the CPU 201 creates the conversion table shown in FIG. 7 by associating the index number of the photodiode 13 thus obtained, the rotation angle with respect to the perpendicular line described above, and the indicated value corresponding to the rotation angle.

Note that, in FIG. 7, for convenience of explanation, the rotation angle of the indicator needle meter 100 with respect to the perpendicular line is shown, but the rotation angle may be omitted from the actual conversion table. That is, if the relationship between the index number of each photodiode 13 and the indicated value is known, there is no problem even if the rotation angle is omitted from the conversion table.

By the way, the conversion table creation processing routine shown in FIG. 4 is applied when the indicator needle meter 100 to which the reading unit 10 is mounted is unknown. Therefore, when the indicator needle meter 100 is known and the conversion table for the known indicator needle meter 100 has been created by the conversion table creation processing routine shown in FIG. 4, the scale area MR1 of the indicator needle meter 100 is created. For an indicator needle meter having the same scale area as the above, only the mounting position deviation of the reading unit 10 needs to be evaluated.

Further, if the conversion table has already been created and the required reading accuracy of the indicated value is not high, and as a result, the deviation at the time of mounting can be sufficiently tolerated, the rotation center identification routine of the indicator needle shown in FIG. And the conversion table creation processing routine is unnecessary.

The server device 200 transmits the conversion table created as described above to the analog meter reading device 1 by wireless communication. As a result, the analog meter reading device 1 calculates the indicated value of the indicator needle meter 100 from the index number of the photodiode 13 using the conversion table transmitted from the server device 200. Specifically, the analog meter reading device 1 calculates the reading value of the indicator needle meter 100 by executing the reading processing routine shown below.

FIG. 8 is a flowchart showing a reading processing routine.
In step S11, the CPU 34 of the control unit 30 determines whether or not the predetermined measurement time has come, and waits until the measurement time comes. The measurement time may be a time repeated at predetermined time intervals or a preset time. When the CPU 34 determines that the measurement time has come, the process proceeds to step S12.

In step S12, the CPU 34 turns on the switch of the power supply line from the storage battery 31 to the reading unit 10 to start supplying power to the reading unit 10.

In step S13, the CPU 34 controls the LED drive unit 16 of the reading unit 10 to light all the LEDs 14. As a result, the indicator needle meter 100 is irradiated with light.

In step S14, the CPU 34 controls the photodiode control unit 15 of the reading unit 10 to read a signal from each photodiode 13 according to the amount of received light. The signal read from each photodiode 13 is supplied to the PD switching unit 32 of the control unit 30.

Further, the CPU 34 controls the PD switching unit 32 so as to sequentially switch and output the signals of the photodiodes 13 having index numbers 0 to 20 from the signals supplied from each photodiode 13. The signal output from the PD switching unit 32 is analog-to-digital converted by the A / D converter 33 and supplied to the CPU 34. As a result, the CPU 34 receives the signals of the photodiodes 13 having index numbers 0 to 20.

FIG. 9A is a schematic diagram when the indicator needle 101 of the indicator needle meter 100 is located in the center of one detection region of the photodiode 13, and FIG. 9B is an output signal of each photodiode 13 at that time. It is a figure which shows.

In the figure, the small white circle indicates the detection area of the photodiode 13, and the square in the center of the small white circle indicates the photodiode 13. The large black circle indicates the light irradiation region by the LED element 14, and the square in the center of the large black circle indicates the LED element 14. The same applies to FIG. 10, which will be described later.

When the indicator needle 101 of the indicator needle meter 100 is located in the center of one detection area of the photodiode 13, the output signal of the photodiode 13 becomes low level. The output signals of the other photodiodes 13 are at a high level. As a result, it can be seen that the indicator needle 101 of the indicator needle meter 100 is located at a position where the output signal corresponds to the low-level photodiode 13.

FIG. 10A is a schematic diagram when the indicator needle 101 of the indicator needle meter 100 exists in the overlapping portion of the detection regions of the two photodiodes 13, and FIG. 10B is an output signal of each photodiode 13 at that time. It is a figure which shows.

When the indicator needle 101 of the indicator needle meter 100 exists in the overlapping portion of the detection areas of the two photodiodes 13, the output signals of the two photodiodes 13 are at the middle level. The output signals of the other photodiodes 13 are at a high level. As a result, it can be seen that the indicator needle 101 of the indicator needle meter 100 is located between the positions where the output signals correspond to the two photodiodes 13 at the middle level.

As described above, the indicator needle 101 of the indicator needle meter 100 may be detected in one detection region of the photodiode 13 or may be detected in each of the two detection regions. Therefore, in consideration of these situations, the CPU 34 of the control unit 30 detects the position information of the indicator needle 101 from the output signal of each photodiode 13 via the index number of the photodiode 13.

In step S15, the CPU 34 calculates an instruction value corresponding to the position information of the instruction needle 101 of the instruction needle meter 100 based on the output signals of the photodiodes 13 having index numbers 0 to 20.

Here, first, the CPU 34 uses the output signal of each photodiode 13 and the index numbers 0 to 20 of each corresponding photodiode 13, and the position information (substantial index) of the indicator needle 101 of the indicator needle meter 100. Number) is detected. Unlike the index number of the photodiode 13, this position information is not limited to an integer and may be a value including a decimal number. Next, the CPU 34 calculates an instruction value corresponding to the position information of the instruction needle 101 with reference to the conversion table shown in FIG. 7.

The method for detecting the position information of the indicator needle 101 is not particularly limited, but there are, for example, the following methods when the detection regions of the adjacent photodiodes 13 overlap.

(Detection method of indicator needle position information 1)
The CPU 34 can detect the position information of the indicator needle 101 by using the weighted average as follows. For example, the CPU 34 first selects, among the output signals of each photodiode 13, an output signal affected by the indicator needle 101, for example, an output signal whose level is equal to or less than a predetermined threshold value and the corresponding photodiode 13.

Next, the CPU 34 weights the index number of the selected photodiode 13. Here, the lower the level of the output signal, the larger the weight of the index number, and the higher the level of the output signal, the smaller the weight of the index number.

Finally, the CPU 34 calculates a weighted average of each weighted index number. The calculated weighted average value corresponds to the position information of the indicator needle 101.

(Detection method 2 of indicator needle position information)
When the detection regions of more adjacent photodiodes 13 overlap, the CPU 34 can also detect the position information of the indicator needle 101 using the maximum likelihood method as follows.

First, with reference to the photodiode 13 corresponding to the lowest level output signal among the output signals of each photodiode 13, five continuous photodiodes 13 are selected so that the reference photodiode 13 is at the center. .. Then, each of i = -2 relative index number of five photodiodes 13 selected, the -1,0,1,2 each _Y -2 the level of the output signal of the photodiode 13, Y _- Let it be ₁ , Y ₀ , Y ₁ , and Y ₂ . The index number of the reference photodiode 13 is i = 0, and the output signal level thereof is Y ₀ .

Next, with the index number i as an independent variable, the maximum likelihood model function for the levels of the five output signals is defined by a parabola (quadratic function) such as the following equation.
y = a · i ² + b · i + c

Here, the maximum likelihood model function refers to a model function that can most accurately represent the levels of all output signals of the five selected photodiodes 13 based on the least squares method. Here, the substantial index number of the photodiode 13 that minimizes the output signal level is i (= i _min ) when the parabolic value is minimized.

That is, from y'= 2a · i _min + b = 0, i _min = −b / 2a
= -0.7 × (-2Y _-2 -Y _-1 + Y ₁ + 2Y ₂ )
/ (2Y- _2- Y _-1 -2Y _0- Y ₁ + 2Y ₂ )
Will be.

For example, the output signal level Y ₀ of the photodiode 13 with index number 7 is the minimum (0.9 μA), and Y _-2 , Y _-1 , Y ₀ , Y ₁ , and Y ₂ are 52.9 μA and 16.9 μA, respectively. , 0.9 μA, 4.9 μA, and 28.9 μA, i _min = 0.3.

_imin corresponds to the amount of deviation as an index number from the reference photodiode 13. From this, the substantial index number of the photodiode 13 at which the level of the output signal becomes the minimum value, that is, the position information of the indicator needle 101 is the index number 7 + _imin = 7.3.

When the row of photodiodes 13 overlaps not only the portion in the tip direction of the indicator needle 101 but also the portion in the root direction, there are two places where the output signal level is minimized. In this case, the portion of the indicator needle 101 in the tip direction is narrower than the portion in the root direction. This corresponds to the degree of spread of the parabola in the maximum likelihood model function described above.

Therefore, in the maximum likelihood model functions obtained in the portion in the tip direction and the portion in the root direction of the indicator needle 101, a is obtained as shown in the following equation.
a = 1/14 × (2Y- _2- Y _-1 -2Y _0- Y ₁ + 2Y ₂ )

Then, the actual index number of the photodiode 13 may be obtained from the maximum likelihood model function in which the value of a becomes larger, in other words, the spread of the corresponding parabola is smaller.

In the present embodiment, the position information of the indicator needle 101 is obtained by using the output signals of the five photodiodes 13, but the number of the photodiodes 13 is not particularly limited.

Further, the indicator needle meter 100 is based on the premise that the black indicator needle 101 is on the white scale plate 102, but the present invention is not limited to such a configuration. For example, it is also possible to use an indicator needle meter having a black scale plate and a white indicator needle. In this case, the level of the output signal of the photodiode at or near the position of the indicator needle becomes high. Therefore, by detecting the position information of the photodiode 13 that maximizes the output signal level, the position information of the indicator needle is detected.

In this embodiment, a parabola is assumed as the maximum likelihood model function, but it is not limited to that. For example, the maximum likelihood model function may be any function having a minimum value (or maximum value) in the vicinity of the photodiode with respect to the relative index number (i = 0).

Further, in the present embodiment, when the indicator needle 101 does not exist in the vicinity, it is assumed that the light receiving amount of each photodiode 13 is the same and the level of each output signal becomes a predetermined default value. However, in reality, even if the amount of received light is the same, the level of the output signal may vary due to individual differences of each photodiode. Further, when a scale number or a manufacturer's logo mark is printed in the light-sensitive area of the photodiode on the scale plate, the light reflection is hindered by the scale number or the logo mark. In this case, even though there is no indicator needle 101, the level of the output signal as the default value becomes smaller than that of the periphery.

Therefore, instead of applying the above calculation directly to the output signal strength of the photodiode, the above calculation is obtained by subtracting the output signal strength of the photodiode as a default value from the detected output signal strength of the optical sensor. Such an operation may be applied.

The CPU 34 calculates the indicated value by detecting the position information of the indicator needle 101 as described above and then converting the position information of the indicator needle 101 into the indicated value with reference to the conversion table. This conversion table is created by the conversion table creation routine shown in FIG.

In step S16, the CPU 34 determines whether or not the indicated value is abnormal (outside the preset range). If it is determined that the indicated value is abnormal, the process returns to step S14, the signal is read again from each photodiode 13, and the indicated value is calculated again. If the indicated value is not abnormal, the process proceeds to step S17. If the indicated value becomes abnormal a predetermined number of times in succession, the process proceeds to step S18 without returning to step S14.

In step S17, the CPU 34 transmits the calculated instruction value to the server device 200 via the wireless communication unit 38. As a result, the server device 200 can constantly manage the indicated value of each indicator needle meter 100.

In step S18, the CPU 34 controls the LED drive unit 16 of the reading unit 10 to turn off all the LEDs 14.

In step S19, the CPU 34 turns off the switch of the power supply line from the storage battery 31 to the reading unit 10 to stop the power supply to the reading unit 10. Then, the process returns to step S11. After that, the process of step S11 and the like is repeated again.

As described above, in the analog meter reading device 1 according to the first embodiment, the position information of the indicating needle 101 of the indicating needle meter 100 is provided by the reading unit 10 having the photodiode 13 and the LED element 14 without using the image sensor. Is detected. As a result, the analog meter reading device 1 can significantly reduce the power consumption as compared with the device using the image sensor.

In particular, since the analog meter reading device 1 requires less time for the light detection itself than the device using the image sensor, the measurement frequency of the indicator needle meter 100 is not high (for example, in a few hours). In the case of one time), the power consumption can be suppressed to one-fifth to one-tenth.

In the present embodiment, the analog meter reading device 1 transmits the indicated value to the server device 200, but instead of the indicated value, the analog meter reading device 1 transmits the position information of the indicating needle 101 represented by the index number of the photodiode 13. May be good. In this case, since the server device 200 may convert the position information of the indicator needle 101 into the indicated value by referring to the conversion table created by itself, the server device 200 transmits the conversion table created by itself to the analog meter reading device 1. You can save time and effort.

[Second Embodiment]
Next, a second embodiment of the present invention will be described. The same parts as those in the first embodiment are designated by the same reference numerals, and redundant description will be omitted.

FIG. 11 is a diagram showing a cross-sectional shape of the photodiode 13 according to the second embodiment. In the second embodiment, the lens 13a is formed on the light receiving surface of the photodiode 13. The light-sensitive region of the photodiode 13 changes depending on the presence or absence of the lens 13a. Here, the light-sensitive region refers to a spatial region in which the presence of a point light source can be detected by the photodiode 13. The range in the depth direction and the range in the width direction of the light sensitive region are limited according to the optical characteristics of the lens 13a.

FIG. 12 (A) is a schematic view showing a light sensitive region A1 of the photodiode 13 when the photodiode 13 does not have a lens, and FIG. 12 (B) is a schematic diagram showing the light sensitivity when the lens 13a is formed on the photodiode 13. It is a schematic diagram which shows the region A2.

In the case of FIG. 12A, the light sensitive region A1 of the photodiode 13 includes not only the indicator needle 101 but also the scale plate 102. Therefore, when the scale number or the manufacturer's logo mark is displayed on the scale plate 102, the photodiode 13 obtains an output signal influenced by those displayed objects. Specifically, the level of the output signal of the photodiode 13 is lowered as if the indicator needle 101 actually exists even though the indicator needle 101 does not exist.

In the case of FIG. 12 (B), the light-sensitive region A2 of the photodiode 13 is limited in the depth direction as compared with the light-sensitive region A1 of FIG. 12 (A) due to the influence of the optical characteristics of the lens 13a. That is, the light-sensitive region A2 includes the indicator needle 101 or its vicinity in the depth direction, but does not include the scale plate 102. Therefore, even if the scale plate 102 has a display such as a scale number or a manufacturer's logo mark or does not have such a display, the output signal of the photodiode 13 is not affected.

Therefore, when the photodiode 13 detects the indicator needle 101, the photodiode 13 obtains a low-level output signal (a signal corresponding to black) as in the first embodiment. Further, when the photodiode 13 does not detect the indicator needle 101, it is not affected by the presence or absence of a display object such as a scale number on the scale plate 102 or a manufacturer's logo mark, and corresponds to a high-level output signal (corresponding to white). Signal). That is, as in the first embodiment, it can be seen that the indicator needle 101 is at a position where the output signal corresponds to the low-level photodiode 13.

Therefore, as shown in FIG. 12B, when the depth direction of the light-sensitive region is limited by the lens 13a, the photodiode 13 has a display object such as a scale number or a manufacturer's logo mark on the scale plate 102. Even in some cases, it is possible to obtain an output signal of an appropriate level depending only on the presence of the indicator needle 101.

FIG. 13 is a diagram showing light sensitive regions A11 to A15 and A21 to A25 having different diameters. When the photodiode 13 does not have the lens 13a (FIG. 12 (A)), the adjacent light-sensitive regions A11 to A15 partially overlap each other as shown in FIG. Therefore, when the indicator needle 101 is present at the position shown in FIG. 13, it is detected in the two light sensitive regions A12 and A13.

When the photodiode 13 has a lens 13a (FIG. 12B), the diameter of each of the light-sensitive regions A21 to A25 is smaller than the diameter of each of the light-sensitive regions A11 to A15. Therefore, the adjacent light-sensitive regions A21 to A25 do not overlap each other and are in contact with each other. Therefore, when the indicator needle 101 is present at the position shown in FIG. 13, it is detected only in one light sensitive region A22. That is, by limiting the diameter (width direction) of the light-sensitive regions, it is possible to suppress the overlap of adjacent light-sensitive regions, and the detection accuracy of the position information of the indicator needle 101 is improved.

As a result, the analog meter reading device 1 according to the second embodiment can detect the depth direction of the light-sensitive region even when the scale number or the manufacturer's logo mark is displayed on the scale plate 102 of the indicator needle meter 100. By limiting, the position information of the indicator needle 101 of the indicator needle meter 100 can be detected with high accuracy without being affected by those displayed objects. Further, the analog meter reading device 1 can detect the position information of the indicator needle 101 with higher accuracy by limiting the width direction of the light sensitive region.

In the second embodiment, as shown in FIG. 13, the adjacent light sensitive regions A21 to A25 are in contact with each other without overlapping, thereby improving the detection accuracy of the position information of the indicator needle 101. Such an example. It is not limited to. For example, as in the second embodiment, the diameter of the light-sensitive regions may be reduced, and as in the first embodiment, adjacent light-sensitive regions may overlap each other. At this time, the position information of the indicator needle 101 is detected by the "indicator needle position information detection method 1 (weighted average)" or "indicator needle position information detection method 2 (maximum likelihood method)" of the first embodiment. It is possible.

As described above, by reducing the diameter (width direction) of the light-sensitive regions and allowing the adjacent light-sensitive regions to overlap each other, the number of photodiodes 13 that can be mounted per unit length is dramatically increased. Can be increased to. As a result, the detection accuracy of the position information of the indicator needle 101 and the calculation accuracy of the indicator value can be further improved.

[Third Embodiment]
Next, a third embodiment of the present invention will be described. The same parts as those in the above-described embodiment are designated by the same reference numerals, and duplicate description will be omitted.

FIG. 14A is a front view of the analog meter reading device 1a of the third embodiment attached to the indicator needle meter 100a, and FIG. 14B is an external side view thereof.
The analog meter reading device 1a has a reading unit 10a that irradiates the indicator needle meter 100a with light to detect the position information of the indicator needle 101, and an instruction value indicated by the indicator needle 101 based on the position information detected by the reading unit 10a. It includes a control unit 30 for calculating.

The reading unit 10a is attached to the center of rotation of the indicator needle meter 100a or its vicinity. Then, the reading unit 10a detects the position information of the indicating needle 101 of the indicating needle meter 100a, and supplies the detected position information to the control unit 30. The control unit 30 calculates the indicated value of the indicator needle meter 100a based on the position information supplied from the reading unit 10a, and transmits the indicated value to the server device 200 (see FIG. 17) described later by wireless communication.

The indicator needle meter 100a includes an indicator needle 101, a scale plate 102 that displays a physical quantity (numerical value) corresponding to the position indicated by the indicator needle 101, a cover glass 103 for protecting the indicator needle 101, and the back of the cover glass 103. It is on the side (scale plate 102 side) and is provided with a cover member 104 that covers the vicinity of the rotation center of the indicator needle for aesthetic purposes.

The numerical values displayed on the tip side of the indicator needle 101 and the scale plate 102 can be visually recognized from the outside through the cover glass 103. The center of rotation of the indicator needle 101 and its peripheral portion are hidden by the cover member 104 between the cover glass 103 and the indicator needle 101, and cannot be seen from the outside.

One side surface of the two side surfaces of the reading unit 10a is directly mounted on the cover glass 103 surface of the indicator needle meter 100a, and the side surface (reading surface) that optically reads the position information of the indicator needle 101 of the indicator needle meter 100a. Is. The other side surface of the above two sides is exposed to the outside, and the side surface (marker surface) on which a two-dimensional code (for example, QR code (registered trademark)) for identifying the analog meter reader 1a is printed is printed. ).

FIG. 15A is a view showing a reading surface of the reading unit 10a, and FIG. 15B is a side view of the main part thereof.
The reading unit 10a was fixed to the disk-shaped case lid 21, the outer peripheral portion 22 formed along the outer peripheral portion of the case lid 21, the substrate 23 attached to one side of the case lid 21, and the substrate 23. It includes an LED element 24 and a camera module 25 fixed to a substrate 23.

The camera module 25 is fixed to the substrate 23 and is arranged at the center of the case lid 21 by a fixing member 22a extending from the outer peripheral portion 22. The LED element 24 is arranged on the substrate 23 and around the camera module 25.

FIG. 16 is a side view of a main part of the camera module 25.
The camera module 25 is a camera for fixing the arrangement position of the wide-angle lens 26, the image sensor 27 that generates a signal based on the incident light from the wide-angle lens 26, and the wide-angle lens 26, and protecting the image sensor 27 from the outside. It has a case 28 and. Reflected light from the indicator needle 101 and the scale plate 102 of the indicator needle meter 100a is incident on the image sensor 27 via the wide-angle lens 26.

FIG. 17 is a block diagram showing a functional configuration of the analog meter reading device 1a. The reading unit 10a and the control unit 30 transmit and receive data to and from each other according to a predetermined protocol, as in the first embodiment. That is, in the analog meter reading device 1a of the present embodiment, the reading unit 10 of the analog meter reading device 1 shown in FIG. 3 is replaced with the reading unit 10a shown in FIG.

The reading unit 10a is predetermined for a plurality of LED elements 24 that irradiate the indicator needle meter 100a with light, an image sensor 27, a drive control unit 111 that drives the image sensor 27, and an image read from the image sensor 27. It includes an image processing unit 112 that performs processing, a storage unit 113 that stores data, and an LED driving unit 114 that drives each LED element 24. The control unit 30 is configured in the same manner as in FIG.

The analog meter reading device 1a configured as described above executes the following processes (1) to (4) in order to calculate the indicated value of the indicator needle meter 100a.

(1) Identification of the center of rotation of the indicator needle 101 using a meter image (2) Setting of a virtual photodiode (virtual PD) based on the center of rotation of the indicator needle 101 (3) Of the indicator needle 101 based on the index number of the virtual PD Creation of a conversion table for converting the position information into the indicated value of the indicator needle 101 (4) Detection of the position information of the indicator needle 101 based on the index number of the virtual PD and calculation of the indicated value corresponding to the position information. , (1) to (3) are executed only once as the initial setting. (4) is executed every time the indicated value is read from the indicator needle meter 100a.

FIG. 18 is a flowchart showing a rotation center identification routine for identifying the rotation center of the indicator needle 101.

In step S21, the user directly mounts the reading unit 10a near the center of rotation of the indicator needle 101 of the indicator needle meter 100a. The mounting position of the reading unit 10a does not have to be the exact center of rotation of the indicator needle 101, and may be a position near the center of rotation estimated from the position of the indicator needle 101 or the shape of the scale region by the scale plate 102. That is, normally, the reading unit 10a is mounted in a state of being displaced from the indicator needle meter 100a.

Here, the center position of the reading unit 10a has a translational deviation of Δr from the rotation center of the indicator needle 101. Further, the perpendicular line of the reading unit 10a has a rotational deviation of Δθ with respect to the perpendicular line of the indicator needle meter 100a. However, Δr and Δθ can be ignored by the processing described later.

In step S22, the image processing unit 112 acquires a distorted meter image generated by the image sensor 27 based on the incident light from the wide-angle lens 26. The image sensor 27 is incorporated in the reading unit 10a directly mounted on the indicator needle meter 100a. Therefore, the distortion generated in the meter image is not caused by the photographing direction described in the first embodiment, but is caused by the wide-angle lens 26.

On the other hand, due to the action of the wide-angle lens 26, as shown in FIG. 14, the reading unit 10a is attached to the vicinity of the rotation center of the indicator needle meter 100a in which the rotation center of the indicator needle 101 is hidden. The numerical values displayed on the tip side of the indicator needle 101 and the scale plate 102 can be read in the lens field of view LV.

In step S23, the image processing unit 112 converts the distorted meter image acquired from the image sensor 27 into a flat image. The image conversion method for converting the distorted image into a flat image is not particularly limited. For example, when the wide-angle lens 26 is an equidistant projection type fisheye lens, the technique described in JP-A-11-18007 can be applied to the image conversion method, for example.

The meter image converted into a flat image includes an area where the indicator needle 101 and the scale plate 102 are displayed (effective area) and an area where the cover member 104 behind the cover glass 103 is displayed and the indicator needle 101 is hidden. (Invalid area) and is included. The center of rotation of the indicator needle 101 exists in the invalid region, but is identified based on the effective region of the meter image, as will be described in detail later.

In step S24, the image processing unit 112 performs binarization processing on the meter image. Specifically, the image processing unit 112 extracts the luminance component from the meter image, and for each pixel in the effective region of the meter image, if the value of each luminance component is larger than a predetermined threshold value, it is set to '1', otherwise. If so, set '0'. By setting '0' for all the pixels in the invalid area of the meter image, for example, in the template search described later, when all the pixels in the template are '0', the template is the invalid area. Make it easy to detect what is in.

In step S25, the image processing unit 112 extracts the scale region MR3 (see FIG. 19 described later) from the binarized meter image. Here, the scale area means an area on the scale plate 102 on which a scale showing the correspondence between the position indicated by the indicator needle 101 and the indicated value is drawn.

FIG. 19 is a diagram showing an example of the scale region MR3 of the indicator needle meter 100a. In the present embodiment, the scale region MR3 exists in the effective region of the meter image, and has an outer radius a, an inner radius b, a central angle φ− to the left of the perpendicular line of the indicator needle meter 100a, and a central angle φ + to the right. It is modeled by the arc band template described using the four parameter values. From this, the center position of the scale region MR3 becomes the same as the rotation center of the indicator needle 101.

The template of the scale area MR3 is not limited to the shape of the arc band specified by the four parameters (a value, b value, φ- value and φ + value). That is, the template of the scale region MR3 may have a shape other than the arc band as long as it can identify the rotation center of the indicator needle 101.

In step S25, the image processing unit 112 further changes the center position of the template and the four parameter values to search for the optimum template position within the effective region of the binarized meter image. Then, when the pixel value distribution in the template becomes uniform, the image processing unit 112 extracts the scale region MR3 according to the center position of the template and the four parameter values.

Specifically, at the time of template search, the image processing unit 112 divides the internal area of the template into several subregions, and the average pixel value in each subregion is substantially constant over the entire interior of the template. In that case, the scale region MR3 is extracted as the center position, a value, b value, φ− value, and φ + value at that time.

In step S26, the image processing unit 112 identifies the rotation center of the indicator needle 101 using the scale region MR3 (a value, b value, φ− value and φ + value) extracted in step S25. In the present embodiment, the center of rotation of the indicator needle 101 coincides with the center position of the arc band-shaped template that models the scale region MR3. Therefore, the image processing unit 112 identifies the rotation center of the indicator needle 101 by specifying the center position of the template of the scale region MR3 extracted in step S25.

If the scale region MR3 is not in the shape of an arc band, the image processing unit 112 may prepare in advance a template that models the relationship between the shape of the scale region MR3 and the rotation center of the indicator needle 101. Then, the image processing unit 112 performs matching processing between the scale region MR and the template, and identifies the rotation center of the indicator needle 101 based on the template after the matching processing.

FIG. 20 is a flowchart showing a virtual PD setting routine. FIG. 21 is a diagram showing a virtual PD set in the effective area of the meter image.

In step S31, the image processing unit 112 sets an arc (arrangement arc) in which the virtual PD is arranged by using the rotation center of the indicator needle 101 identified in the above process. The placement arc is an arc that is included in the effective area of the meter image and passes through the centers of all virtual PDs that are placed on the concentric circles of the rotation center of the indicator needle 101. The radius of the placement arc is determined according to the detection accuracy required for the analog meter reading device 1a, specifically, the number and size of the virtual PDs to be placed.

The analog meter reading device 1a of the present embodiment targets the indicator needle meter 100a shown in FIG. 14, but the indicator needle meter 100 shown in FIG. 1 (a type in which the rotation center of the indicator needle 101 and its periphery are not hidden). But it is possible. In this case, the arrangement arc is set as an arc whose center is at the same position as the rotation center of the indicator needle 101, the central angle to the left of the perpendicular line of the indicator needle meter 100 is φ−, and the central angle to the right is φ +. ..

As shown in FIG. 21, the central angle of the placement arc is equal to the angle formed by the respective half straight lines from the center of the placement arc identified in the above process to the minimum scale value and the maximum scale value of the scale area MR3. The upper limit of the radius of the placement arc is not particularly limited as long as the entire placement arc is included in the effective area of the meter image and the detection accuracy required for the analog meter reading device 1a is satisfied. On the other hand, the lower limit of the radius of the placement arc is equal to the length of the line segment connecting the point where each half line intersects the boundary line between the effective region and the invalid region and the center of the placement arc.

In step S32, the image processing unit 112 determines the center (placement position center) of each placement position of the series of virtual PDs defined on the set placement arc. The center of each placement position is from the center of the placement arc to the scale area MR3 for each rotation angle obtained by dividing the center angle of the above-mentioned placement arc by the number obtained by subtracting 1 from the number of virtual PDs to be placed. It is given as the intersection of the pointed half straight line and the arrangement arc.

For example, in the case of FIG. 21, the central angle of the arrangement arc is π / 2 (= 90 °). When 21 virtual PDs are set in this central angle, the scale is scaled from the center of the arrangement arc for each rotation angle of (π / 2) ÷ (21-1) = π / 40 (= 4.5 °). The intersection of the half straight line toward the region MR3 and the arrangement arc is the center of the arrangement position of each virtual PD.

As is clear from the above processing, the center of the arrangement arc of the virtual PD, in other words, the substantial center of the reading unit 10a, is dynamically determined by the rotation center itself of the identified indicator needle 101. Therefore, the translational deviation (Δr) from the center of rotation of the indicator needle 101 to the center of the mounting position of the reading unit 10a does not affect the detection accuracy of the indicated value and can be ignored in the present embodiment.

In step S33, the image processing unit 112 assigns an index number to each virtual PD (center of the placement position). For example, in the case of FIG. 21,

index numbers

1, 2, 3, ... Are assigned in order from the left of each virtual PD. In this case, the index number of the virtual PD at the right end is 21.

In step S34, the image processing unit 112 calculates the rotation deviation (Δθ) of the reading unit 10a from the perpendicular line of the indicator needle meter 100a. For example, in the case of FIG. 21, the rotation deviation (Δθ) is determined by the boundary line between the effective region and the invalid region extracted from the meter image and the pixel sensor array which is the horizontal arrangement of the pixel sensors of the image sensor 27. Corresponds to the angle of formation.

In step S35, the image processing unit 112 creates a virtual PD configuration table that summarizes the index number and the center of the placement position of each virtual PD.

Here, when the rotation deviation (Δθ) of the reading unit 10a from the perpendicular line of the indicator needle meter 100a described above is 0, the center of the arrangement arc is the origin, and the boundary line between the effective region and the invalid region is a straight line y = d. , The center of the placement position of the virtual PD shown in FIG. 21 is as follows.

FIG. 22 shows the arrangement position of the virtual PD shown in FIG. 21 when the rotation deviation (Δθ) is 0, the center of the arrangement arc is the origin, and the boundary line between the effective area and the invalid area is a straight line y = d. It is a figure which shows.

Comparing FIG. 21 and FIG. 22 here, the actual placement position of the virtual PD has a rotation deviation of Δθ in the positive direction. From this, the virtual PD composition table showing the relationship between the index number of the virtual PD and the placement position center of the virtual PD (xy coordinates of the placement position center) is as shown in Table 1.

In step S36, the image processing unit 112 sets the virtual PD for the placement position center of each virtual PD. Specifically, the image processing unit 112 sets the virtual PD with an aggregate of pixel sensors within a predetermined range based on the center of the arrangement position of the virtual PD.

FIG. 23 is a diagram showing a virtual PD composed of an aggregate of a total of 49 pixel sensors having 7 pixels in the vertical direction and 7 pixels in the horizontal direction. As shown in the figure, the center of the placement position of the virtual PD is determined by the intersection of the placement arc and the half straight line from the center of the placement arc to the scale area. The virtual PD is composed of a total of 49 pixel sensors, 7 pixels in the vertical direction and 7 pixels in the horizontal direction, with reference to the center of the arrangement position.

FIG. 24 is a flowchart showing a conversion table creation routine for creating a conversion table for converting the index number of the virtual PD to the indicated value. The conversion table creation routine is executed as the final processing of the initial setting.

In step S41, the image processing unit 112 identifies the center of rotation of the indicator needle 101. Here, the image processing unit 112 reuses the result obtained by the processing of the rotation center identification routine shown in FIG.

In step S42, the image processing unit 112 executes the polar coordinate conversion process centered on the arrangement position of the scale area MR3 and the virtual PD shown in FIG.
Specifically, the image processing unit 112 has the arc band-shaped scale region MR3 extracted from the meter image with the rotation angle with respect to the x-axis shown in FIG. 22 as the horizontal axis and the radius method as the vertical axis. Expand to the polar coordinate plane to generate the polar coordinate-converted scale region MR4. Then, the image processing unit 112 evenly arranges 21 virtual PDs with respect to the scale area MR4.

Here, the image processing unit 112 executes the same processing as in step S9 shown in FIG. However, in the image processing unit 112, the scale area MR3 shown in FIG. 19 or 21 and the half straight line from the center of the arrangement arc used when creating the virtual PD configuration table to the arrangement position center of each virtual PD are formed. The rotation angle of the eggplant can be used.

FIG. 25 is a diagram showing a rectangular scale region MR4 developed on a polar coordinate plane based on FIG. 22 and a series of virtual PDs arranged corresponding to the scale region MR4. The scale region MR4 is represented by a rectangular region having a width from (3π / 4 + Δθ) to (π / 4 + Δθ) on the horizontal axis and a height from a to b on the vertical axis in the polar coordinate plane.

The center of the placement position of each virtual PD is dynamically determined on the placement arc of the virtual PD obtained based on the extracted scale area MR3 so as to correspond to the main scale of the scale area MR3. From this, the influence of the rotation deviation (Δθ) of the reading unit 10a from the perpendicular line of the indicator needle meter 100a can be ignored here.

In other words, the rotation deviation (Δθ) affects the calculation of the placement position center of each virtual PD as shown in FIG. 22, but the correspondence between the main scale of the scale area MR3 and the placement position of the virtual PD. Does not affect.

In step S43, the image processing unit 112 creates a conversion table for converting the position information (substantial index number) of the indicator needle 101 of the indicator needle meter 100a based on the index number of the virtual PD into the indicated value.

Here, the image processing unit 112 recognizes the scale number in the vicinity of the scale region MR4 as a numerical value by using the OCR technique as in step S10 of FIG. 4, and is recognized by the position information of the scale number on the polar coordinate plane. Executes the process of associating the numbers. As a result, the conversion table shown in Table 2 below is created.

In the present embodiment, the virtual PD configuration table in Table 1 and the conversion table in Table 2 are generated by independent processing. On the other hand, both the virtual PD configuration table and the conversion table are created by using the scale area MR3 whose heading column is the index number (or the rotation angle corresponding to the index number 1: 1) and is extracted from the meter image. It is a thing. Therefore, the virtual PD configuration table and the conversion table may be created as one table.

However, the virtual PD configuration table is used by the reading unit 10a in FIG. 17, while the conversion table is used by the control unit 30 in FIG. Therefore, the table in which the virtual PD configuration table and the conversion table are integrated is suitable when the reading unit 10a and the control unit 30 are integrated.

After the above-mentioned initial setting is completed, the control unit 30 calculates the indicated value of the indicator needle meter 100a by executing the following reading processing routine using the signal output from the reading unit 10a.

FIG. 26 is a flowchart showing a reading processing routine. The data transmitted / received between the reading unit 10a and the control unit 30 in the present embodiment is the same as the data transmitted / received between the reading unit 10 and the control unit 30 in the first embodiment.

Therefore, in the reading processing routine shown in FIG. 26, the photodiode has replaced the virtual PD (note that the index numbers are 0 to 20 for the former and 1 to 21 for the latter), and It is the same as the reading processing routine shown in FIG. 8 except that step S14 shown in FIG. 8 has replaced step S14a. Therefore, in the following step S14a, the points different from the step S14 shown in FIG. 8 will be mainly described.

In step S14a, the CPU 34 controls the image processing unit 112 of the reading unit 10a to read the output signals of each virtual PD (a plurality of pixel sensors corresponding to the virtual PD). The output signal of each virtual PD read out is supplied to the PD switching unit 32 of the control unit 30.

Further, the CPU 34 controls the PD switching unit 32 so as to sequentially switch and read the output signals of the virtual PDs with index numbers 1 to 21 from the output signals of each virtual PD. The signal output from the PD switching unit 32 is analog-to-digital converted by the A / D converter 33 and supplied to the CPU 34. As a result, the CPU 34 receives the output signals of the virtual PDs with index numbers 1 to 21.

On the other hand, in the reading unit 10a, when the control unit 30 executes the reading processing routine, the LED 24 lights up according to the control signal (step S13) from the control unit 30. As a result, the indicator needle meter 100a is irradiated with light. Then, when the control unit 30 executes step S15a, the reading unit 10a executes the next signal reading routine to supply a signal to the control unit 30.

FIG. 27 is a flowchart showing a signal reading routine.
In step S51, the image processing unit 112 of the reading unit 10a acquires a distorted meter image generated by the image sensor 27 based on the incident light from the wide-angle lens 26 after the LED 24 is turned on.
In step S52, the image processing unit 112 converts the distorted meter image acquired from the image sensor 27 into a flat image.

In step S53, the image processing unit 112 reads the virtual PD configuration table (Table 1) from the storage unit 113, and obtains the placement position centers of all the virtual PDs based on the read virtual PD configuration table. Then, the image processing unit 112 sums the output signals of a total of 49 pixel sensors of the pixel sensor (center pixel sensor) existing at the center of each arrangement position and the pixel sensors existing around the pixel sensor, and each virtual PD Calculate the output signal of.

For example, in FIG. 23, the output signal strength of each pixel sensor constituting the virtual PD is v (p, q). However, v (0,0) is the output signal strength of the central pixel sensor of the virtual PD. p and q are relative addresses in the horizontal direction and the vertical direction of the other pixel sensors when the central pixel sensor is used as a reference.

V represents the output signal strength of the virtual PD composed of a total of 49 pixel sensors of 7 pixels in the vertical direction and 7 pixels in the horizontal direction shown in FIG. 23 by the following equation 1.

Equation 1

The image processing unit 112 calculates Equation 1 for all virtual PDs, and stores the calculation result in the storage unit 113 as an output signal of each virtual PD. The saved virtual PD output signal is read out in step S55, which will be described later, and then supplied to the control unit 30.

In step S54, the image processing unit 112 sets i = 1 for the index number i of the virtual PD.
In step S55, the image processing unit 112 reads the output signal of the virtual PD (i) from the storage unit 113 and supplies it to the control unit 30. At first, the output signal of the virtual PD (1) is read and supplied to the control unit 30.

In step S56, the image processing unit 112 confirms the presence / absence of virtual PD switching control. Specifically, the image processing unit 112 confirms whether or not the PD switching unit 32 of the control unit 30 has received a control signal to switch to the virtual PD having the next index number. When the control signal is received, the process proceeds to step S57.

In step S57, the image processing unit 112 increments the index number (i = i + 1) and returns to step S55. That is, steps S55 to S57 are repeatedly executed until the output signals of all the virtual PDs are read from the storage unit 113.

On the other hand, in step S56, when the above-mentioned control signal is not received (timed out), the output signals of all virtual PDs have been read, and this routine is terminated. The output signal of the virtual PD supplied from the reading unit 10a to the control unit 30 by this routine is used in the calculation of the indicated value in step S15 described above.

As described above, the analog meter reading device 1a according to the third embodiment uses the reading unit 10a for photographing the indicating needle meter 100a, so that the reading unit 10a is translated and rotated with respect to the indicating needle meter 100a. Even if there is, the indicated value of the indicator needle meter 100a can be calculated with high accuracy without being affected by the translation deviation and the rotation deviation.

Further, the analog meter reading device 1a can obtain the indicated value of the indicator needle meter 100a by identifying the rotation center based on the meter image even if the indicator needle meter 100a has the rotation center of the indicator needle 101 hidden. Can be calculated.

Further, since the analog meter reading device 1a can set a series of virtual PDs for detecting the position information of the indicator needle 101 to an arbitrary size and interval, the analog meter reader 1a indicates with the detection accuracy and reliability desired by the user. The value can be calculated.

The present invention is not limited to the above-described embodiment, but can also be applied to those whose design has been changed within the scope of the matters described in the claims.
For example, in the third embodiment, the routines of FIGS. 18, 20, and 24 were executed by the reading unit 10a, but after being executed by the control unit 30 and the server device 200, the results are read by the reading unit 10a. May be supplied to. Further, the virtual PD is not limited to the case where it is set inside the scale area MR3 as shown in FIG. 21, and may be set outside the scale area MR3.

1,1a

Analog meter reader

10,10a Reading unit 13 Photodiode 14 LED element 25 Camera module 30

Control unit

100, 100a Indicator needle Meter 101 Indicator needle 102 Scale plate 103 Cover glass

Claims

An instruction to indicate a scale plate on which numerical values assigned to a plurality of scales are displayed and an indicated value of the scale plate, and to rotate with reference to a predetermined rotation center on a plane at a predetermined distance from the display surface side of the scale plate. An analog meter reading device that reads the indicated value of an indicator needle meter having a needle and a transparent member that covers the display surface side of the scale plate so as to include the indicator needle.
A light emitting element, an image pickup element that detects light emitted from the light emitting element and reflected by the indicator needle or the scale plate of the indicator needle meter by a plurality of pixel sensors, and an image pickup element of the indicator needle meter in the image pickup element. A plurality of virtual sensors are virtually arranged outside or inside the arrangement direction of the numerical values displayed on the scale plate, index numbers are assigned to the plurality of virtual sensors in the order of the arrangement direction, and the plurality of virtual sensors are assigned. A reading unit including an arithmetic processing unit that calculates each output signal of the plurality of virtual sensors using the output signals of one or more pixel sensors constituting each of the above.
Of the output signals of each of the plurality of virtual sensors calculated by the arithmetic processing unit, a substantial index number corresponding to the presence position of the indicator needle is detected based on the output signal of at least one virtual sensor. Index number detector and
An analog meter reader equipped with.
The arithmetic processing unit
The brightness component of the scale region of the indicator needle meter is extracted and extracted from the captured image of the indicator needle meter generated by the image pickup device with the reading unit attached to the transparent member of the indicator needle meter. A rotation center identification unit that identifies the rotation center of the indicator needle of the indicator needle meter based on the brightness component of the scale region.
An arc is set outside or inside in the arrangement direction of the numerical values displayed on the scale plate of the indicator needle meter on a concentric circle centered on the rotation center identified by the rotation center identification unit. A virtual sensor setting unit that sets the plurality of virtual sensors by arranging the plurality of virtual sensors on an arc and assigning index numbers to the plurality of virtual sensors in the order of the arrangement direction.
An output signal calculation unit that calculates the output signal of each of the plurality of virtual sensors by using the output signals of one or more pixel sensors constituting each of the plurality of virtual sensors set by the virtual sensor setting unit.
The analog meter reading device according to claim 1.
The analog meter reader further comprises
A conversion table storage unit that stores a conversion table that describes the correspondence between the index numbers assigned to the plurality of virtual sensors and the numerical values displayed on the scale plate.
An instruction value calculation unit that calculates the instruction value based on the conversion table stored in the conversion table storage unit and the substantial index number detected by the index number detection unit.
The analog meter reading device according to claim 1 or 2.
The conversion table is based on the rotation center identified by the rotation center identification unit, the scale area including the numerical value displayed on the scale plate, and the index number assigned to the plurality of virtual sensors. It also has a conversion table creation unit to create
The analog meter reading device according to claim 3, wherein the storage unit stores the conversion table created by the conversion table creation unit.
The analog meter reading device according to claim 4, wherein the reading unit is separated from the indicated value calculation unit.