JP2021124905A - Meter reading device, meter reading system, meter reading method, and computer program - Google Patents

Abstract

To provide a meter reading device, a meter reading system, a meter reading method, and a computer program that can read a value indicated by a meter with high accuracy even to an intermediate value of numbers on a drum based on an image obtained by imaging the meter equipped with a mechanical counter.SOLUTION: A meter reading device reads a number indicated by a mechanical counter from a captured image obtained by imaging a meter including the mechanical counter using a rotational drum. The meter reading device cuts out a drum image of each digit of the number from the captured image in a range based on a vertical position serving as a reference in the image, and calculates an intermediate value by a position where cut-out multiple drum images are most likely to match in a template image. The meter reading device outputs a number having the number of digits larger than the number of digits of the mechanical counter based on the calculated intermediate value.SELECTED DRAWING: Figure 16

Description

The present invention relates to a meter reading device, a meter reading system, a meter reading method and a computer program for reading a value indicated by a meter provided with a mechanical counter.

As an instrument used in an IoT (Internet of Things) system, a type of digital sensor that outputs a measurement result by a digital signal such as a digital thermometer, a hygrometer, or a vibration meter is often used. This is because it is easy for the communication device to transmit a numerical value indicating the measurement result. However, it does not replace all conventionally used meters. There is a demand for a method of reading measured values by using an existing single-needle rotary analog meter or a meter equipped with a mechanical counter as it is.

Patent Document 1 discloses a method of reading a value indicated by a mechanical counter by an OCR (Optical Character Recognition) method. In the method disclosed in Patent Document 1, a meter reader patrols a meter provided with a mechanical counter and images an image with a digital camera. Patent Document 1 discloses that the obtained image data is transmitted to a computer provided in a meter reader's management office, image processed by the computer, and character recognition is performed by a compound similarity method, a matching method, or the like. ing.

As disclosed in Patent Document 1, when the OCR method of recognizing one character at a time by matching is used, it is difficult to recognize numbers from a drum in the middle of rotation. Patent Document 2 discloses a method of accurately reading the value indicated by the mechanical counter even while the drum is rotating. In Patent Document 2, the numbers are accurately read by pattern matching using a pattern image in the middle of rotation for each number.

Japanese Unexamined Patent Publication No. 2008-147844 Japanese Unexamined Patent Publication No. 2008-234160

Considering the use of the meter equipped with the mechanical counter, the reading accuracy of the value indicated by the meter equipped with the mechanical counter was sufficient as the reading accuracy identifiable from the numbers of each drum. However, the state during the rotation of the drum corresponds to the intermediate value, and even with the existing meter, it is possible to measure a numerical value finer than the numerical value represented only by the numerical value of the drum. If even this intermediate value can be automatically read, an IoT system can be realized by collecting detailed measured values from an existing meter.

The present invention is a meter that reads a value indicated by a meter equipped with a mechanical counter, which can read a value indicated by the meter even to an intermediate value of a drum number with high accuracy based on an image obtained by capturing an image of the meter equipped with a mechanical counter. It is an object of the present invention to provide a reading device, a meter reading system, a meter reading method, and a computer program.

The meter reading device according to the embodiment of the present disclosure is a meter reading device that reads a numerical value indicated by the mechanical counter from an image captured by an image of a meter including a mechanical counter using a rotary drum. From the image processing unit that cuts out the drum image of each digit of the number in the range based on the reference vertical position in the image, and the drum image corresponding to at least the last digit of the cut out plurality of drum images. A calculation unit that calculates the intermediate value between the above numbers with the set number of digits of 2 or more, the value of the number shown in the drum image of each digit cut out, and the intermediate value. Based on the above, the calculation unit includes an output unit that outputs a number having a number of digits larger than the number of digits of the mechanical counter, and the calculation unit refers to a template image that is a developed image of the rotary drum stored in advance. The intermediate value is calculated based on the position in the rotation direction of the rotary drum in the range where the drum image cut out by the image processing unit has the highest possibility of matching in the template image.

The meter reading system of one embodiment of the present disclosure receives a camera device that captures a meter including a mechanical counter using a rotary drum, and image data of an image captured from the camera device, and receives the image data of the image captured by the camera device, and the mechanical counter. A meter reading system including a meter reading device that reads a number indicated by the meter and a client device that receives the number read by the meter reading device, wherein the camera device is a distance to the mechanical counter and the height of the meter. The meter reading device includes an image pickup unit attached via an attachment that is positionably adjustable in a direction, and the meter reading device uses a drum image of each digit of the number as a reference in the image from the image of the image data. From the image processing unit that cuts out in the range based on the position and the drum image corresponding to at least the last digit of the cut out multiple drum images, the intermediate value between the numbers on the drum is set to 2 or more. Outputs a number with a number of digits larger than the number of digits of the mechanical counter based on the calculation unit that calculates with the number of digits of, the value of the number shown in the drum image of each digit cut out, and the intermediate value. The setting receiving unit includes an output unit for receiving data and a setting receiving unit that accepts a setting of a cutout range of a drum image from the image in association with information identifying the meter or a camera device that images the meter. The display unit and the operation unit included in the client device accept the setting of the cutout range.

The meter reading method of the embodiment of the present disclosure is a meter reading method for reading a number indicated by the mechanical counter from an image captured by an image of a meter including a mechanical counter using a rotary drum. Then, the drum image of each digit of the number is cut out in a range based on the reference vertical position in the image, and among the plurality of cut out drum images, the drum image corresponds to at least the last digit. Determine the position of the drum in the rotation direction in the template image of the range most likely to match in the template image which is the unfolded image of the drum stored in advance, and the numbers and numbers on the drum based on the determined position. The intermediate value between and is calculated by the set number of digits of 2 or more, and based on the numerical value shown in the drum image of each digit cut out and the intermediate value, the mechanical counter Includes processing to output a number with a number of digits larger than the number of digits.

The computer program of the embodiment of the present disclosure is a computer program that causes a computer to perform a process of reading a number indicated by the mechanical counter from an image captured by an image of a meter including a mechanical counter using a rotary drum. Then, the computer cuts out a drum image of each digit of the number from the captured image in a range based on a reference vertical position in the image, and corresponds to at least the last digit among the plurality of cut out drum images. Regarding the drum image, the position in the rotation direction of the drum in the template image in the range in which the drum image is most likely to match in the template image which is the developed image of the drum stored in advance is determined, and the determined position. The intermediate value between the numbers on the drum is calculated based on the set number of digits of 2 or more, and the value of the number shown in the drum image of each digit cut out and the intermediate value Based on the value, a process of outputting a number having a number of digits larger than the number of digits of the mechanical counter is executed.

In the present disclosure, a range in which the numbers of each digit are copied from an image captured by a meter including a mechanical counter is cut out based on a reference of a vertical position in the image, and the numbers are obtained for each of the cut out images. Is identified. The identification of the numbers is realized based on the position of the range in which the cut-out image matches in the template image which is the developed image of the drum stored in advance. As a result, for at least the last digit of the drum image, the intermediate value (including after the decimal point) between the numbers can be calculated from the rotation angle of the drum, and the number of digits larger than the number of digits of the mechanical counter (including the number of digits after the decimal point). It is possible to output a numerical value with (significant figures).

The mechanical counter has a reference in the rotation direction of the drum in order to read the numbers uniformly for all digits or divided into upper digits and lower digits. The state in the middle of rotation of the drum represents an intermediate value between the numbers of the drum, and when reading this in the same way as a human reads, it is necessary to cut out according to the standard. Therefore, in the above-mentioned cropping, if the captured image is distorted or rotated, it cannot be accurately cropped. Therefore, the meter reader corrects the distortion, rotation, and color tone of the captured image with respect to the captured image of the meter. You may.

In the meter reading device of the embodiment of the present disclosure, the calculation unit determines the likelihood that the drum image cut out by the image processing unit is likely to match the image cut out from the template image. The cutout range of the image from the template image is calculated while shifting in the rotation direction of the drum, and from the likelihood distribution obtained by calculating each likelihood cutout position, the cutout position from the template image having the highest likelihood is determined. A numerical value corresponding to the position of the drum in the rotation direction is calculated as the intermediate value.

In the meter reading device of the present disclosure, which position in the template image the cut-out drum image is likely to match is determined by slightly shifting the range to be compared in the template image. The rotation of the drum during the rotation of the mechanical counter correctly represents the intermediate value between the numbers on the drum, and the intermediate value indicated by the rotation can be calculated by image processing. ..

In the meter reading device of the embodiment of the present disclosure, the calculation unit has a pixel of interest and a periphery for each of the plurality of pixels included in the drum image cut out by the image processing unit and the image cut out from the template image. The amount of change in the pixel value between the pixels is calculated, the direction of the change and the vector having the amount of change are obtained, and the vector of the pixels at the same position between the drum image and the image cut out from the template image. The inner product is calculated, and the calculated inner product is summed over the entire drum image to calculate the likelihood.

In the meter reading device of the present disclosure, a process of determining which position in the template image the cut-out drum image is likely to match is calculated based on the feature amount of the pixel in the image. Specifically, the direction of change in the pixel value with the peripheral pixels calculated for each pixel and the inner product of the vectors having the amount of change are calculated, and the total of the inner product in the entire image is calculated as the likelihood. Will be done. The accuracy of matching between numbers with many similar shapes is better than matching based on the sum of absolute values of pixel value differences.

The meter reading device of the embodiment of the present disclosure includes a setting reception unit that accepts a setting of a drum image cutout range from the captured image in association with information that identifies the meter or a camera device that images the meter. The setting receiving unit receives the cutout range over a plurality of digits based on a reference vertical position in the image.

In the meter reading device of the present disclosure, the cutout range from the captured image captured by the meter is constant as long as the camera device is fixed, and is therefore stored as setting information. The median value is calculated from the degree of rotation corresponding to the reading direction of the drum numbers (vertical direction of the image). Therefore, the vertical position of the cropping range in the image should be set in common with all digits, or with reference to the same vertical position for each upper and lower digit.

In the meter reading device of the embodiment of the present disclosure, the setting receiving unit sets the cutting range so that the numerical value calculated by the calculating unit based on the cutting range matches the numerical value visually recognized for the meter. Execute the fine adjustment process.

In the meter reading device of the present disclosure, it is required that the numerical value output by the output unit matches the visual result of the mechanical counter of the meter. Fine adjustment is performed on the set cutout range on the meter reading device side so as to match the visual recognition result so as to increase the possibility of matching with the template image corresponding to the visual recognition result.

In the meter reading device of the embodiment of the present disclosure, the setting receiving unit receives the numerical value by visual recognition by designating a cutout range in the template image.

In the meter reading device of the present disclosure, the visual recognition result is accepted by designation in the template image. As a result, a reading result that matches the visual result of the operator with respect to the meter can be obtained.

According to the present disclosure, not only the number of the drum but also the intermediate value indicated by the meter can be read from the image captured by the mechanical counter provided in the meter.

The meter reading device attaches a camera device to an existing meter and executes reading processing on an image obtained from the camera device. According to the present disclosure, it is possible to appropriately attach a camera device to various existing meters. This makes it possible to monitor fluctuations by period within a unit period (for example, one day) using a meter using an existing mechanical counter, and provide various information from the data using the existing counter. Becomes possible.

It is explanatory drawing which shows the outline of the meter reading system. It is a schematic perspective view of a camera device. It is a block diagram which shows the structure of the electrical part of a camera device. It is a block diagram which shows the structure of the meter reading system. It is a flowchart which shows an example of the acceptance processing procedure of setting information. It is a flowchart which shows an example of the acceptance processing procedure of setting information. It is a figure which shows the content example of the setting information reception screen. It is a figure which shows the content example of the setting information reception screen. It is a figure which shows the content example of the setting information reception screen. It is a figure which shows the content example of the setting information reception screen. It is a flowchart which shows an example of the meter reading processing procedure by a server apparatus. It is a flowchart which shows an example of the meter reading processing procedure by a server apparatus. It is a flowchart which shows an example of the detailed procedure of a matching process. An example of the captured image is shown. It is a schematic diagram which shows the cutout example of an image. It is a figure which shows the outline of the matching process. It is a figure which shows the calculation method of the likelihood. A schematic diagram of the likelihood distribution is shown.

Hereinafter, the meter reading system of the present disclosure will be specifically described with reference to the drawings showing the embodiment thereof.

FIG. 1 is an explanatory diagram showing an outline of the meter reading system 100. The meter reading system 100 includes a plurality of camera devices 1, a gateway device 2, a server device 3, and a client device 4.

The camera device 1 is provided for a plurality of meters M provided in the equipment in the factory. The gateway device 2 is installed in the factory and can communicate with a plurality of camera devices 1. The communication connection between the camera device 1 and the gateway device 2 is realized by, for example, short-range wireless communication. The camera device 1 and the gateway device 2 may be configured so that a communication connection can be made by wire.

The camera device 1 transmits the captured image captured by the meter M to the gateway device 2 by communication. The gateway device 2 transmits captured images transmitted from each of the plurality of camera devices 1 to the server device 3 via the network N.

The server device 3 executes a process described later on the captured image transmitted from the gateway device 2, and specifies a value indicated by one or a plurality of meters M reflected in the captured image. In addition to this, the server device 3 exerts a function of providing information on the measurement target of the meter M to the manager of the factory where the meter M is provided based on the values indicated by the meters M respectively. In the present embodiment, the server device 3 will be described as one server computer for ease of explanation, but it is realized by a virtual server that logically operates by a plurality of instances on one server computer. .. Further, the functions or processes may be distributed or superimposed on a plurality of server computers.

The network N includes a public network N1 and a carrier network N2. The public network N1 is the so-called Internet. The carrier network N2 is a network provided by a communication carrier that realizes wireless communication based on a standard such as a next-generation or next-generation high-speed mobile communication standard. The public network N1 includes an access point AP. The carrier network N2 includes a base station BS. The gateway device 2 can send and receive information to and from the server device 3 connected to the public network N1 by the access point AP or the base station BS.

Of the various meters installed in the factory, for the meter M that does not have a digital output function, the person in charge or an external meter reader has conventionally visually read the numerical value indicated by the meter M. In the present embodiment, the camera device 1 is attached to the meter M that does not have the digital output function, and the meter M is functioned as a meter reading device of the server device 3 based on the captured image captured by the camera device 1. A meter reading system 100 that reads the numerical values indicated by the above with high accuracy is realized. The server device 3 can specify the numerical value indicated by the meter M with high accuracy by the processing contents shown below. As a result, the work of the person in charge or the meter reader can be omitted by a simple procedure of attaching the camera device 1 to the existing meter M and installing the gateway device 2 in the factory.

FIG. 2 is a schematic perspective view of the camera device 1. As for the camera device 1, FIG. 2 shows a state in which the camera device 1 is fixed to the meter M. As shown in FIG. 2, the meter M to which the camera device 1 of the present disclosure is attached has a display unit M1 using a mechanical counter, and at least a part of the meter M is transparent so that the display unit M1 can be visually recognized. It is a type covered with M2. The camera device 1 has an attachment 10 for applying to various meters M, and an electrical component 15.

The attachment 10 is made of resin. The attachment 10 is composed of a first member 11, a second member 12, and a third member 13 whose positional relationship can be adjusted with each other. The first member 11 extends from one side of a flat plate portion having a mounting surface that comes into contact with the cover M2, and the engaging portion 111 with which the second member 12 engages and the eaves extending from the engaging portion 111. It has 112. A flat plate-shaped second member 12 that can move in a direction perpendicular to the mounting surface is engaged with the inside of the engaging portion 111. The second member 12 makes it possible to adjust the positional relationship in the depth direction perpendicular to the width of the attachment 10.

A rectangular rod-shaped third member 13 is attached to the second member 12 so as to be movable in the height direction of the meter M. An imaging unit 151 is provided at the end of the third member 13 toward the mounting surface side. The third member 13 makes it possible to adjust the positional relationship of the imaging unit 151 along the height direction of the attachment 10.

As shown in FIG. 2, the movable second member 12 and third member 13 of the attachment 10 exhibit a function of adjusting the distance and height of the image pickup unit 151 with respect to the display unit M1 of the meter M.

Before executing the meter reading process in the server device 3 described later, it is necessary to adjust the distance and height of the imaging unit 151 with respect to the meter M in order to set the accuracy. When the distance between the image sensor and the display unit M1 is short, the entire display unit M1 cannot be included in the image pickup range of the image sensor, barrel distortion, and distortion of the numbers on the drum of the display unit M1 become large. In some cases, the correction described later may not be enough. On the contrary, if the distance between the image sensor and the display unit M1 is too long, the resolution of the numbers on the display unit M1 becomes too low, and it is necessary to maintain an appropriate distance. The adjustable configuration of the attachment 10 makes it possible to easily adjust the distance and height and capture an appropriate correctable image. Further, the eaves of the first member of the attachment 10 can reduce the reflection of the ambient light of the meter M on the imaging unit 151.

A lighting unit 153 (see FIG. 3) including a plurality of LEDs (Light Emitting Diodes) is provided on the mounting surface side of the first member 11, and the display unit M1 serves as indirect illumination for the imaging unit 151 (shown in the figure). figure). This enables the imaging unit 151 to capture a clearer captured image.

The attachment 10 is attached to the meter M by adhering the attachment surface and the cover M2 with an adhesive such as double-sided tape, or the hook attached to the cover M2 corresponds to the hook provided on the attachment surface. The engaging portion is engaged to fix the camera device 1 to the meter M.

FIG. 3 is a block diagram showing the configuration of the electrical component 15 of the camera device 1. The electrical component 15 is configured by mounting a control circuit 154, a battery 155, and a communication circuit 156 on a substrate. The control circuit 154 is connected to the image pickup unit 151 via the flexible wiring board 152 and to the illumination unit 153 via the flexible wiring board (which may be simply a signal line).

The image pickup unit 151 includes a visible light or infrared light image pickup element and a memory, and executes an image pickup in response to a control signal of the control circuit 154. The control circuit 154 can acquire the image data of the image captured by the image pickup unit 151 via the flexible wiring board 152.

The communication circuit 156 realizes a communication connection with the gateway device 2. The communication circuit 156 includes, for example, a short-range wireless communication module such as Bluetooth®, especially 2.4 GHz BLE (Bluetooth Low Energy).

The control circuit 154 includes a processor and a memory, and controls imaging execution by the imaging unit 151, lighting and extinguishing of the lighting unit 153, and data transmission / reception by the communication circuit 156 based on the program and setting information stored in the memory. In the setting information stored in the memory, device identification information for distinguishing the camera device 1 on which it is mounted from another camera device 1 is stored in advance. The device identification information may be, for example, a MAC address, or may be information previously given to each camera device 1 by an administrator who manages the operation of the meter reading system 100. The control circuit 154 associates the device identification information stored in the memory with the device identification information stored in the memory when the image data of the captured image is transmitted from the communication circuit 156 to the gateway device 2. The control circuit 154 executes imaging in the imaging unit 151 based on a set schedule or in response to an instruction from the gateway device 2, and transfers image data acquired from the imaging unit 151 from the communication circuit 156 to the gateway device 2. Send.

The camera device 1 may include other sensors in addition to the image pickup unit 151. The camera device 1 may include, for example, a temperature sensor and a humidity sensor. The camera device 1 may include a sensor that acquires information about the imaging environment, such as its own battery voltage and illuminance sensor. The control circuit 154 may transmit the image data including the data measured by these sensors in the metadata description portion of the image data.

Next, meter reading based on the image data transmitted from the camera device 1 will be described. FIG. 4 is a block diagram showing the configuration of the meter reading system 100. The gateway device 2 uses a device called a so-called IoT gateway including a first communication unit 22 and a second communication unit 23 having different protocols. The gateway device 2 includes a control unit 20 and a storage unit 21. The control unit 20 uses a CPU, a clock, or the like to perform a process of transmitting image data transmitted from the camera device 1 to the server device 3 based on a program and setting information stored in advance in the storage unit 21. The first communication unit 22 includes a short-range wireless communication module such as BLE for communication with the camera device 1. The second communication unit 23 includes a communication module for the next-generation mobile communication standard that realizes communication via the carrier network N2 included in the network N. The gateway device 2 may also include a device for Ethernet (registered trademark) and a communication module for wireless LAN.

The server device 3 uses a server computer. The server device 3 includes a control unit 30, a storage unit 31, and a communication unit 32. The control unit 30 is a processor using a CPU (Central Processing Unit) and / or a GPU (Graphics Processing Unit), and includes a built-in volatile memory, a clock, and the like. The control unit 30 may use a GPU or a separate graphic card to execute image processing as described later. The control unit 30 executes each process based on the meter reading program 30P stored in the storage unit 31 to make the general-purpose server computer function as a specific meter reading device.

The storage unit 31 stores the information referred to by the control unit 30 in addition to the meter reading program 30P using the hard disk. The storage unit 31 stores the template image of the drum for each type of the meter M or the type of the mechanical counter included in the meter M. The template image of the drum is a developed image of the drum, that is, an image in which numbers are continuously arranged along the rotation direction of the drum. The storage unit 31 stores the setting information for each meter M referred to by the control unit 30 when processing the captured image, which will be described later, in association with the identification information of the meter M. The setting information includes the above-mentioned template image, the number of digits of the mechanical counter, the correction amount regarding distortion / rotation of the image, and the information of the cutout range of the drum image from the image. Since the template image included in the setting information may differ for each digit of the mechanical counter, it is preferable to include identification information for specifying the template image for each digit. In the following description, a mechanical counter having drums of the same size will be described as an example, but for a meter M having a mechanical counter having drums of different sizes in the lower digits, the cutting range will be changed. May be set for each digit. When a plurality of display units M1, that is, a plurality of mechanical counters are imaged in the captured image, the storage unit 31 may store the setting information for each display unit M1. The meter reading program 30P stored in the storage unit 31 may be acquired and stored from the outside by the communication unit 32.

The communication unit 32 includes a network card. The control unit 30 can send and receive information to and from the client device 4 via the network N by the communication unit 32.

The client device 4 includes a control unit 40, a storage unit 41, a display unit 42, an operation unit 43, an audio input / output unit 44, and a communication unit 45. The control unit 40 includes a processor such as a CPU or GPU and a memory or the like. The control unit 40 may be configured as one piece of hardware (SoC: System On a Chip) in which a processor, a memory, a storage unit 41, and a communication unit 45 are integrated. The control unit 40 causes a general-purpose computer to function as a client device 4 for the user of the meter reading system 100 of the present embodiment based on the application program 40P stored in the storage unit 41.

The storage unit 41 includes a non-volatile memory such as a flash memory. The application program 40P is stored in the storage unit 41. The application program 40P may include a Web browser function. A general-purpose Web browser program stored in the storage unit 41 may be used. The storage unit 41 stores the data referred to by the control unit 40. The application program 40P may be one in which the application program 49P stored in the storage medium 49 is read by the control unit 40 by a reading unit (not shown) and installed in the storage unit 41. The application program 40P may be one in which the control unit 40 receives the application program (not shown) distributed by any server device via the network N by the communication unit 45 and installs it in the storage unit 41.

The display unit 42 includes a display device such as a liquid crystal panel or an organic EL display. The operation unit 43 is an interface that accepts user operations, and includes physical buttons and a touch panel device with a built-in display. The operation unit 43 can accept an operation on the screen displayed on the display unit 42 by a physical button or a touch panel.

The audio input / output unit 44 includes a speaker, a microphone, and the like. The voice input / output unit 44 includes a voice recognition unit, and can recognize the operation content from the input voice with a microphone and accept the operation.

The communication unit 45 is a wireless communication module that realizes transmission / reception of information to / from the server device 3 via the network N. The communication unit 45 may perform communication via the network N by wire using a network card.

A method of reading the value indicated by the meter M in the meter reading system 100 configured as described above will be described. In the present embodiment, the gateway device 2 stores in advance the device identification information of the camera device 1 that can be connected to the communication in the storage unit 21. The gateway identification information of the gateway device 2 itself may be additionally stored in the storage unit 21. The gateway device 2 establishes pairing for a plurality of camera devices 1 by the first communication unit 22 for the camera device 1, and temporarily assigns a different address to each of the camera devices 1 for which the pairing has been established. The control unit 20 of the gateway device 2 instructs a plurality of camera devices 1 to execute imaging and transmit image data in order in association with the device identification information stored in the storage unit 21.

When each of the camera devices 1 receives an instruction from the gateway device 2, the illumination unit 153 is turned on, the image pickup unit 151 executes imaging, the illumination unit 153 is turned off, and the image data obtained by the imaging is transmitted from the communication circuit 156 to the gateway. It is transmitted to the device 2. The control circuit 154 of the camera device 1 transmits the device identification information stored in the built-in memory together with the image data.

Each time the control unit 20 of the gateway device 2 receives the image data from the camera device 1 to the first communication unit 22, the control unit 20 associates the image data with the corresponding device identification information and sends the image data from the second communication unit 23 to the server device 3. Send. The control unit 20 may store image data from the camera device 1 using the built-in memory. If transmission from the second communication unit 23 is difficult, it may be stored and transmitted when communication becomes possible.

The control unit 20 of the gateway device 2 executes the above-mentioned imaging instruction, image data reception and transmission processing for one or a plurality of camera devices 1 capable of communication connection at set intervals.

Each time the server device 3 receives image data from the gateway device 2, the value indicated by the meter M is determined from the display unit M1 of the meter M for each meter M in association with the device identification information by a process described later. As will be described later, the control unit 30 cuts out a drum image of each digit of the mechanical counter from the image captured by the display unit M1 of the meter M in order to determine the numerical value. The cut-out range is stored as setting information for each camera device 1 as fixed data on the premise that the imaging range of the image captured by the same camera device 1 is invariant once fixed. The setting information is set via the client device 4 by an operation based on the visual inspection of the setting operator. The control unit 40 of the client device 4 accepts the settings as shown below.

5 and 6 are flowcharts showing an example of the procedure for accepting setting information. When the setting operator logs in to the service of the meter reading system 100 using the account information based on the application program 4P or the Web browser program of the client device 4, the control unit 30 of the server device 3 uses the service provided by the server device 3 to: Starts processing.

The control unit 30 receives the selection or input of the device identification information of the camera device 1 corresponding to the meter M to be set (step S301). The control unit 30 stores the received device identification information in association with the logged-in account information (step S302). The control unit 30 reads out the image data stored in the storage unit 31 in association with the received device identification information (step S303).

If the image data associated with the received device identification information is not yet stored in the storage unit 31 in step S303, the image data corresponding to the device identification information is collected from the server device 3 and the gateway device 2. You may. The server device 3 may transmit an imaging instruction for setting to the camera device 1 via the gateway device 2.

The control unit 30 transmits a setting information reception screen including an image of the read image data to the client device 4 (step S304).

The control unit 30 accepts the selection of the meter M type (model number, etc.) on the setting information reception screen (step S305). After that, the control unit 30 accepts various selections or inputs while displaying the preview screen on the client device 4. The control unit 40 of the client device 4 transmits / receives data to / from the server device 3 in the background while displaying the setting information reception screen, and receives the setting information while changing the screen.

On the setting information reception screen, the control unit 30 receives input or selection of the distortion / rotation correction amount of the image included in the screen by the operation unit 43 of the client device 4 (step S306). In step S306, the control unit 30 may accept the correction setting on the setting information reception screen while including the captured image of the meter M in the preview frame.

The control unit 30 stores the distortion / rotation correction amount received in step S306 in the storage unit 31 in association with the device identification information and the display unit identification information that identifies the display unit M1 (step S307). The display unit identification information may be information, a number, a symbol, or the like indicating the location of the display unit M1 on the meter M. The display unit identification information is information for identifying the display unit M1 when a plurality of mechanical counters are used for one meter M and all of them are to be read, and one mechanical counter for one meter M. If is used, it may not be used.

In this embodiment, the camera device 1 is attached to each meter M. Therefore, the relationship between the camera device 1 and the meter M is one-to-one, and it is not necessary to identify the meter M. However, since the process described below can be applied even when a plurality of meters M are shown in one image obtained from one camera device 1, the control unit 30 associates the information with the information for identifying the meters M. You may memorize it.

The control unit 30 accepts the selection or input of the vertical position in the image of the cutout range (step S308), and receives the horizontal position of the drum image of each digit and the interval between the drum images, respectively (step S309). .. Device identification information and display unit identification that identifies the display unit M1 using the information of the cutout range defined by the vertical position, horizontal position, and interval in the image received in steps S308 and S309 as setting information. It is stored in the storage unit 31 in association with the information (step S310).

The control unit 30 reads out the template image of the mechanical counter corresponding to the type of the meter M selected in step S305 (step S311). In step S311 the control unit 30 reads out a template image that differs from digit to digit depending on the type of mechanical counter. The control unit 30 displays the drum image of each digit of the template image in a comparable manner while displaying the rectangle indicating the cutout range set in step S310 on the captured image of the meter M in the setting information reception screen (). Step S312).

The control unit 30 receives, for each digit, the degree of rotation of the drum shown in the captured image of the meter M previewed by the operation unit 43 of the client device 4 at a position in the template image (step S313). ). In step S313, the control unit 30 displays a visual result by the operator of how much the drum of each digit is rotated in the image captured by the meter M that is being previewed, as a continuous image of the numbers developed by the drum. Accept by specifying the position in the rotation direction of a certain template image.

The control unit 30 cuts out the drum image of each digit of the mechanical counter of the meter M from the image of the image data read out in step S303 within the cutting range of the setting information stored in step S310 (step S314).

The control unit 30 cuts out from the template image an image of a position corresponding to the degree of rotation received in step S313 with a size included in the cutout range of step S310 (step S315).

The control unit 30 determines the likelihood of indicating the high possibility that the drum image of the range cut out from the actual captured image in step S314 matches the range cut out from the template image in step S315 for each digit. Calculate (step S316).

The control unit 30 determines the cutting range (vertical position, horizontal position, size in the image) of the drum image for each of the camera device 1 and the display unit M1 based on the information set in step S310. Fine adjustment is made every time (step S317). Each time the control unit 30 makes a fine adjustment, the control unit 30 cuts out a drum image from the imaging range according to the cutout range after the fine adjustment (step S318). Each time the control unit 30 makes fine adjustments, there is a high possibility that the drum image cut out from the actual captured image in step S318 within the range after fine adjustment matches the image cut out from the template image in step S315. The likelihood indicating the above is calculated for each digit (step S319).

The control unit 30 determines whether or not the fine adjustment of the cutting range has been executed until a predetermined condition is satisfied (step S320). The predetermined condition may be the number of times, or may be until the value indicating that the likelihood calculated for each digit is the highest and the probability of matching is the highest is reached.

If it is determined in step S320 that the predetermined condition is not satisfied (S320: NO), the control unit 30 returns the process to step S317.

When it is determined in step S320 that the predetermined condition is satisfied (S320: YES), the control unit 30 stores the cutout range having the highest likelihood in association with the device identification information and the display unit identification information ( Step S321), the process is terminated.

Although the above-mentioned processing has been described as being executed by the server device 3 as a meter reading device in the background, a part or all of the processing may be executed by the client device 4.

The processing procedure shown in the flowcharts of FIGS. 5 and 6 can assist the operator in adjusting the cropping range from the captured image set in step S308 to be accurate in order to calculate the likelihood.

7 to 10 are diagrams showing an example of the contents of the setting information reception screen.
FIG. 7 is a screen that accepts selection or input of a distortion / rotation correction amount in the image of the cutout range. The setting information reception screen 420 shown in FIG. 7 is a screen that is transmitted from the server device 3 to the client device 4 and displayed on the display unit 42 by the control unit 40. The setting information reception screen 420 shown in FIG. 7 shows a preview screen for correction of the image captured by the display unit M1 of the meter M, and a control for selecting the correction amount. The captured image included in the preview screen of the setting information reception screen 420 of FIG. 7 is rotating.

The setting information reception screen 420 of FIG. 7 has a slider-shaped control 421 for setting a correction amount for correcting the rotation distortion of the image, and a slider-shaped control 422 for setting a correction amount for the barrel-shaped distortion. including. In the preview of FIG. 7, a grid-like guide that assists in specifying the degree of rotation correction and barrel distortion correction may be displayed together.

As shown in FIG. 7, the setting information reception screen 420 includes a control 430 for changing the setting information to another. Each time the control 430 is selected, the setting target is cyclically switched from the setting information reception screen 420 for setting the correction amount in FIG. 7 to the setting information reception screen 420 in FIG. 8 and the setting information reception screen 420 in FIGS. 9 and 10. Be done.

FIG. 8 is a screen that accepts selection of a vertical position, a horizontal position, and an interval in the image of the cutout range. The setting information reception screen 420 of FIG. 8 is displayed after the correction amount is set in FIG. 7. On the setting information reception screen 420 of FIG. 8, the rotational distortion of the image captured by the display unit M1 of the meter M displayed in the preview is corrected.

On the setting information reception screen 420 shown in FIG. 8, an image of the display unit M1 of the meter M after correction of rotational distortion and barrel distortion, a preview screen of a cutout range from the image, and a control for moving the cutout range. 423 is shown. The preview cutout range after correcting the captured image with the correction amount received on the setting information reception screen 420 of FIG. 7 is indicated by a broken line rectangle superimposed on the image in FIG. The two-dot chain line in the figure indicates the vertical position of the cutout range in the image. When the control 423 is scrolled, the vertical position of the cut-out range indicated by the alternate long and short dash line is changed in the preview of the cut-out range. Since the reference line for reading the numbers on the drum is engraved on most of the display units M1, the two-dot chain line is set so as to coincide with the reference line. When the reference line differs depending on the digit (for example, the lower digit has a small drum), it may be set according to the reference line depending on the digit.

The setting information reception screen 420 of FIG. 8 includes a control 424 for selecting the horizontal position of the drum image of each digit and a control 425 for selecting the interval. The setting information reception screen 420 includes a control 426 for setting the size of the cutout range in the image. The aspect ratio of the rectangle in the cutout range is preset and does not have to be adjusted. It is preferable that the swing width can be set in order to cope with the deviation in the rotation axis direction caused by the gap between the drums of the meter M. The operator can set while checking the preview.

As described above, the cutout range is included in the setting information as fixed data on the premise that the image of the image data transmitted from the same camera device 1 has an invariant imaging range. Since each drum is interlocked in the mechanical counter and the intermediate value is calculated by the rotation position for at least the last digit of the mechanical counter, the rotation position is important. Therefore, as shown in FIGS. 8 and 9, the drums of all digits in the image are set to be cut out at the same vertical position in the image. Further, since the vertical deviation of the image corresponding to the rotation direction greatly affects the reading accuracy, the vertical position of the cutout range is set for each meter M, each mechanical counter, and each camera device 1. Since the mechanical counter has play in the rotation axis direction for each drum, it is permissible for the cutting range of all digits to deviate in the rotation axis direction.

9 and 10 are screens for accepting the rotation position of the drum in the captured image. On the setting information reception screen shown in FIG. 9, a preview of the cutout range including the image captured by the display unit M1 of the meter M and the frame of the cutout range superimposed on the image, and a preview of the cut out drum image of each digit. , A control 432 for adjusting the rotation position of each digit, and a box for showing the value corresponding to the likelihood are shown.

FIG. 9 shows a preview of the cutout range after the rotation degree is set by the operator, and FIG. 10 shows a preview of the cutout range after fine adjustment. The cut-out drum image displayed on the setting information reception screen 420 shown in FIG. 10 is cut out corresponding to the rotation degree set by the control 432 rather than the drum image in the setting information reception screen 420 shown in FIG. The likelihood with the image is high.

In the setting information reception screen 420 described above, the items that can be set are divided into different screens as shown in FIGS. 7, 8 and 9, respectively. However, it is not limited to this. For example, the total likelihood and likelihood shown in FIGS. 9 and 10 are affected by the size and horizontal position of the cutout range in FIG. Therefore, it is better if the setting information of FIG. 8 is accepted while checking the preview of the cutout range from the actual image of FIGS. 9 and 10 and the cutout from the template image and checking the likelihood. Therefore, it is preferable that the setting information reception screen 420 is included so that the preview of the cutout range from the actual images of FIGS. 9 and 10 can be compared with the cutout from the template image and the likelihood can be always confirmed. Further, the preview of the cutout range from the actual image and the comparison with the cutout from the template image were realized by displaying them side by side on the setting information reception screen 420 of FIGS. 9 and 10. However, not limited to this, the preview of the drum image cut out for each digit and the cutout range from the template image showing the degree of rotation are superimposed or displayed at the same position on the screen. May be good. As a result, the degree of matching can be visually confirmed by the operator, and it becomes easy to set the setting information having the best likelihood.

The meter reading process based on the setting information set in this way will be described with reference to the flowchart.

11 and 12 are flowcharts showing an example of the meter reading processing procedure by the server device 3. The server device 3 executes the following processing each time the image data is received from the gateway device 2.

The control unit 30 of the server device 3 identifies the device identification information of the camera device 1 that has captured the received image data (step S101), and stores the setting information including the drum template image associated with the device identification information. Read from unit 31 (step S102). The control unit 30 corrects distortion, rotation, color tone, and the like with respect to the image of the received image data based on the correction amount included in the setting information (step S103).

The control unit 30 selects one display unit M1 (mechanical counter) from the corrected image based on the setting information of the camera device 1 (step S104), and the following based on the setting information for the selected display unit M1. The processing of steps S105 to S113 is executed.

The control unit 30 obtains a drum image of each digit of the mechanical counter of the meter M from the image corrected in step S103 based on the cutout range included in the setting information read in step S102 in the image. It is cut out in a rectangular range having the same vertical position and the same size (step S105). The control unit 30 may resize the cut out range so that it corresponds to the resolution of the template image and includes substantially the same pixels of the same size.

The control unit 30 selects a digit (step S106). The control unit 30 specifies setting information such as a template image corresponding to the selected digit among the setting information read in step S102 (step S107). The control unit 30 selects a drum image of the selected digit from the drum image of each digit cut out in step S105 (step S108). The process of matching with the template image read out in step S102 is sequentially executed along the rotation direction of the drum (step S109).

The control unit 30 images the position of the drum by the cut-out drum image depending on which part of the template image the cut-out image of the selected digit is most likely to match by the matching process in step S109. A likelihood distribution indicating a high possibility of being present is calculated (step S110).

In step S110, the control unit 30 superimposes the drum image of the selected digit while shifting it little by little along the rotation direction of the drum of the template image, and sums the matching degree of the corresponding pixels with respect to all the pixels for each shift position. Is calculated and used as the likelihood. The details of the likelihood calculation method will be described later.

The control unit 30 specifies the position on the template image having the maximum likelihood based on the likelihood distribution calculated in step S110 (step S111).

The control unit 30 calculates an intermediate value between the numbers on the drum with the set number of digits (step S112), and stores the calculated value together with the identification number indicating the selected digit (step S113). ). In step S112, the control unit 30 may calculate the intermediate value as indicating an uncertain position such as between "0" and "1" and between "9" and "0".

When the control unit 30 determines whether or not the digits have been selected for all the digits included in the setting information (step S114) and determines that the processing has not been executed for all the digits (S114: NO), The process returns to step S106.

When it is determined that the processing has been executed for all digits (S114: YES), the value indicated by the selected display unit M1 is specified from the numerical values calculated for each digit (step S115), and the device identification information and the display unit are used. It is stored in association with the identification information that identifies M1 (step S116). In step S115, the numerical value of each digit may be specified by the numerical value calculated in the final digit, and the final value may be specified. The numerical value in step S116 may be stored and stored as a log in association with the time information indicating the time when the image data is received, or only the latest numerical data may be stored.

The control unit 30 determines whether or not the processing has been executed for all the display units M1 included in the image of the image data (step S117). If it is determined that all the processes have not been executed (S117: NO), the control unit 40 returns the processes to step S104, selects the next meter M (S104), and continues the processes.

When it is determined in step S117 that the processing has been executed for all the display units M1 (S117: YES), the control unit 40 ends the processing.

FIG. 13 is a flowchart showing an example of a detailed procedure of the matching process. The processing procedure shown in the flowchart of FIG. 13 corresponds to the details of the matching process of step S108 among the procedures shown in the flowcharts of FIGS. 11 and 12.

The control unit 30 selects one pixel of interest from the pixels included in the drum image with respect to the drum image of the selected digit (step S801), and the peripheral pixels of the selected pixel of interest, for example, the amount of change in the pixel values of up, down, left, and right. Is calculated (step S802).

The control unit 30 calculates and stores a vector having the direction of change and the amount of change as the magnitude (maximum 1) from the amount of change calculated in step S802 with respect to the pixel of interest as the feature amount of the pixel of interest (step S803). ). In step S803, for example, the control unit 30 has an upward or downward vector having a change amount of the pixel value around the pixel of interest in the vertical direction of the image, and a right or left vector having a change amount of the pixel value in the horizontal direction. Calculated by adding the orientation vectors. That is, the control unit 30 has an upward or downward vector indicating a change in the pixel values of the two pixels above and below the pixel of interest, and a vector pointing left or right indicating a change in the pixel values of the two pixels on the left and right of the pixel of interest. Calculate the sum. The calculation method of the change in the pixel value not only in the vertical and horizontal directions but also in the pixel of interest and the four pixels in the vertical and horizontal directions is not limited to this.

The control unit 30 determines whether or not the process has been executed for all the pixels (step S804), and if it is determined that the process has not been executed for all the pixels (S804: NO), the process proceeds to step S801. return. In this case, the control unit 30 selects the next pixel of interest (S801) and executes the processes of steps S802 and S803. The processes of steps S802 and S803 are repeatedly executed for all the pixels of the drum image of the selected digit.

When it is determined in step S804 that the processing has been executed for all the pixels (S804: YES), the control unit 30 advances the processing to the next step S805.

The control unit 30 sets the position in the template image to the initial position in the rotation direction of the drum (step S805), and features each pixel in the same range (number of pixels) as the range in which the drum image is cut out from the set position. Obtain the quantity (step S806). In step S806, the control unit 30 executes the processes of steps S802 and S803 in advance for the entire template image, and stores the template image in the storage unit 31 in association with the template image.

The control unit 30 has a pixel feature amount at a corresponding position (same coordinates within the range) between the drum image (captured image) of the selected digit and the range cut out from the template image, that is, the similarity of the vector. Is calculated for all pixels (step S807). In step S807, the control unit 30 calculates the inner product of the vectors of the pixels at the corresponding positions for all the pixels. When the vectors, which are the feature quantities of the corresponding pixels, have the same amount of change in the same direction, the magnitude of the vectors is calculated to be a maximum of 1, so it becomes "1", which is completely If the orientation is different, it will be "-1". When calculating the inner product, the control unit 30 may arbitrarily weight the angle (θ), cos θ, or the size so that a value as large as it is similar is calculated.

The control unit 30 totals the calculated similarity for all pixels (step S808). The sum is, for example, the sum. In step S808, the control unit 30 may perform processing such as totaling the number of pixels having a similarity of 0.5 or more, that is, a certain degree of similarity or more.

The control unit 30 uses the value totaled in step S808 as the likelihood and stores it in association with the set position (step S809).

The control unit 30 determines whether or not the set position is the final position (step S810), and if it is determined that the set position is not the final position (S810: NO), sets the position in the template image by a predetermined unit (for example, one pixel). Minutes) Shift (step S811) and return the process to step S806.

If it is determined in step S810 that it is the final position (S810: YES), the control unit 30 ends the matching process. The control unit 30 returns the process to step S109 shown in the flowchart of FIG.

According to the processing procedure shown in the flowchart of FIG. 13, erroneous detection in distinguishing numbers having similar shapes can be reduced as compared with executing matching processing using a method called SSD (Sum of Squared Difference). In the process shown in the flowchart of FIG. 13, it is possible to make a determination that reflects the degree of color change of the intermediate color for each pixel, which may occur due to out-of-focus or the like, in addition to the area of the character font of numbers. It is possible to regard the comparison between numbers having similar shapes such as "8" and "9" and "8" and "6" as having a remarkable difference.

The meter reading processing procedure shown in the flowcharts of FIGS. 11 to 13 will be described in detail with reference to specific examples. FIG. 14 shows an example of a captured image. FIG. 14A shows an example of an image captured by the camera device 1, and FIG. 14B shows a comparative image. As shown in FIG. 14, the image of the image data transmitted from the camera device 1 is adjusted to an appropriate distance by the movable part of the attachment 10 and the imaging is executed. Therefore, the image is compared with the comparative image shown in FIG. 14B. The image was taken from a long distance, and the numbers in each digit of the mechanical counter are sufficiently captured with relatively good resolution. In addition, each drum is clearly reflected by the indirect lighting of the lighting unit 153. As described above, since the position of the imaging unit 151 with respect to the display unit M1 can be adjusted by the attachment 10 for various meters M, the accuracy of the meter reading process can be improved.

FIG. 15 is a schematic view showing an example of cutting out an image. An image obtained by capturing an image of the display unit M1 of the meter M is shown, and a cutout range set for the meter M (a rectangle showing the range is shown by a broken line superimposed on the image. Shows the vertical position in the image.

FIG. 16 is a diagram showing an outline of the matching process. FIG. 16 shows a template image and a drum image cut out in the cutout range as shown in FIG. As shown in FIG. 16, the template image is an image of continuous numbers in which the drum is developed in the rotation direction. In the template image, the numbers "0" to "9" and the images from "0" applied to the drum are continuous so that the space between "9" and "0" can be specified.

In the matching process, the control unit 30 cuts out the template image by the same size as the drum image while gradually shifting it from the upper end (or lower end) in FIG. 16, and the pixel values of the pixels at the same positions in the cut out image and the drum image. The difference in the feature amount of is obtained, and the likelihood indicating the high possibility of matching the part is calculated from the total difference of the size of the drum image. The larger the difference, the larger the difference, and the smaller the difference, the smaller the difference. Therefore, the larger the total, the lower the possibility.

The accuracy of pattern matching with the drum image is affected by the resolution of the template image. If the resolution of the template image is such that one number fits in the height of 10 pixels, the intermediate value is calculated with an accuracy of about 0.1 with respect to the number of the drum, and 1 in the height of 40 pixels. If only one number is entered, it is calculated with an accuracy of about 0.025.

FIG. 17 is a diagram showing a method of calculating the likelihood. FIG. 17 shows an image in which the number “2” shown in the drum image is cut out and an image in which the portion “8” is cut out from the template image. The same range in the clipped image is enlarged, and the vector of the feature amount of each pixel in the range is indicated by the arrow in the pixel indicated by the rectangle. The vector points in the direction in which the brightness value (pixel value) increases, and its length corresponds to the amount of change.

In FIG. 17, among the pixel groups, the pixels to be matched at the same position in the image are surrounded by thick lines. By calculating the inner product of the vectors, which are the feature quantities calculated for each pixel to be associated, as shown in the lower part of FIG. 17, the similarity of the feature quantities between the corresponding pixels can be expressed. The control unit 30 calculates the inner product of the vectors showing the similarity calculated for each pixel as described above for all the pixels in the drum image, and takes the sum of the calculation results to obtain the likelihood.

By discriminating the similarity between pixels by the method shown in FIG. 17, it is possible to reduce erroneous detection in distinguishing numbers having similar shapes, as compared with executing matching processing using the SSD method.

FIG. 18 shows a schematic diagram of the likelihood distribution. FIG. 18 shows the likelihood distribution superimposed on the template image. The likelihood distribution indicates the likelihood with respect to the position along the rotation direction of the drum in the template image, that is, the distribution of the degree of similarity between the image cut out from the template image at each position and the drum image cut out from the captured image. ..

FIG. 18 shows an example of a numerical value specified from the likelihood distribution. Corresponding numerical values are shown as scales at positions along the rotation direction of the template image. The upper end of the range of the same size as the drum image surrounding each number corresponds to the value indicated by each number, and the others correspond to the intermediate values of the numbers indicated by the drum. The number of digits (significant figures) of the intermediate value is determined according to the resolution of the selected template image. In the example of FIG. 11, the highest peak position in the likelihood distribution is between "5.000" and "6.60", which is closer to "5.000". The control unit 30 outputs this digit as "5.175" based on the distance (number of pixels) from the position corresponding to "5.000" and "6.00" respectively.

There are 5 significant digits, and the numbers specified for each digit are "0.028", "0.295", "2.852", "8.518", and "5.175" in descending order. If, the control unit 30 has the most likely large digit and the second digit between "0" and "1", the third between "2" and "3", and the fourth digit between "8" and "9". , Will output "285.175" as the numerical value indicated by the meter M.

In this way, based on the image data from the camera device 1 attached after appropriately adjusting the distance and height from the meter M by the attachment 10, the value indicated by the meter M is read as high as including the intermediate value. It is possible to read with accuracy and without error. As described above, since the setting of the setting information is also supported in order to reduce the error, the setting information enables highly accurate and error-free reading.

In the above-described embodiment, the server device 3 which is the meter reading device of the present disclosure is realized as a different computer device capable of communicating with the camera device 1. However, not limited to this, the meter reading device may be configured in the camera device 1. For example, the control circuit 154 on the board may execute the meter reading process by the control unit 30 of the server device 3. Further, the reading process may be executed by the control unit 20 of the gateway device 2.

The embodiments disclosed as described above are exemplary in all respects and are not restrictive. The scope of the present invention is indicated by the scope of claims and includes all modifications within the meaning and scope equivalent to the scope of claims.

1 Camera device 10 Attachment 15 Electrical equipment 151 Imaging unit 3 Server device 30 Control unit (image processing unit)
31 Storage unit 32 Communication unit 30P Meter reading program 4 Client device 42 Display unit 420 Setting information reception screen 43 Operation unit

Claims (9)

A meter reading device that reads a number indicated by the mechanical counter from an image captured by an image of a meter including a mechanical counter using a rotary drum.
An image processing unit that cuts out a drum image of each digit of the number from the captured image in a range based on a reference vertical position in the image.
A calculation unit that calculates an intermediate value between numbers on a drum from a drum image corresponding to at least the last digit of a plurality of cut out drum images with a set number of two or more digits.
It is equipped with a numerical value shown in the drum image of each digit cut out, and an output unit that outputs a number having a number of digits larger than the number of digits of the mechanical counter based on the intermediate value.
The calculation unit has the highest possibility that the drum image cut out by the image processing unit matches the template image that is the developed image of the rotary drum stored in advance in the template image. A meter reading device that calculates the intermediate value based on the position of the rotary drum in the rotation direction. The calculation unit
The likelihood that the drum image cut out by the image processing unit matches the image cut out from the template image is increased while shifting the cutting range of the image from the template image in the direction of rotation of the drum. Calculate and
From the likelihood distribution obtained by calculating the likelihood for each cutout position, a value corresponding to the position of the cutout position from the template image having the highest likelihood in the rotation direction of the drum is calculated as the intermediate value. Item 1. The meter reading device according to item 1. The calculation unit
Regarding the drum image cut out by the image processing unit and each of the plurality of pixels included in the image cut out from the template image.
Calculate the amount of change in the pixel value between the pixel of interest and the peripheral pixels,
Find a vector with the direction of change and the amount of change,
The inner product of the vectors of the pixels at the same position between the drum image and the image cut out from the template image is calculated.
The meter reading device according to claim 2, wherein the calculated inner product is totaled over the entire drum image to calculate the likelihood. It is provided with a setting reception unit that accepts the setting of the cutting range of the drum image from the captured image in association with the information that identifies the meter or the camera device that images the meter.
The meter reading device according to claim 1, wherein the setting receiving unit receives the cutout range over a plurality of digits based on a reference vertical position in the image. The setting reception unit
The meter reading device according to claim 4, wherein the meter reading device executes a process of finely adjusting the cutout range so that the numerical value calculated by the calculation unit based on the cutout range matches the numerical value visually recognized by the meter. The meter reading device according to claim 5, wherein the setting receiving unit receives the numerical value by visual recognition by designating a cutout range in the template image. A camera device that captures a meter, including a mechanical counter with a rotating drum,
A meter reading device that receives image data of an image captured from the camera device and reads a number indicated by the mechanical counter.
A meter reading system that includes a client device that receives the numbers read by the meter reading device.
The camera device is
It comprises an imaging unit attached via an attachment that is adjustablely positioned in the distance to the mechanical counter and in the height direction of the meter.
The meter reading device is
An image processing unit that cuts out a drum image of each digit of the number from the image of the image data in a range based on a reference vertical position in the image.
A calculation unit that calculates an intermediate value between numbers on a drum from a drum image corresponding to at least the last digit of a plurality of cut out drum images with a set number of two or more digits.
An output unit that outputs a number having a number larger than the number of digits of the mechanical counter based on the value of the number shown in the drum image of each digit cut out and the intermediate value.
It is provided with a setting reception unit that accepts the setting of the cutting range of the drum image from the image in association with the information that identifies the meter or the camera device that images the meter.
The setting receiving unit is a meter reading system that receives the setting of the cutting range by the display unit and the operating unit included in the client device. A meter reading method for reading a number indicated by a mechanical counter from an image captured by an image of a meter including a mechanical counter using a rotary drum.
From the captured image, a drum image of each digit of the number is cut out in a range based on a reference vertical position in the image.
Of the plurality of cut out drum images, for the drum image corresponding to at least the last digit, the template image in the range in which the drum image is most likely to match in the template image which is the developed image of the drum stored in advance. Determine the position of the drum in the direction of rotation within
Based on the determined position, the intermediate value between the numbers on the drum is calculated with the set number of digits of 2 or more.
A meter reading method including a process of outputting a number having a number larger than the number of digits of the mechanical counter based on the value of the number shown in the drum image of each digit cut out and the intermediate value. A computer program that causes a computer to execute a process of reading a number indicated by the mechanical counter from an image captured by an image of a meter including a mechanical counter using a rotary drum.
On the computer
From the captured image, a drum image of each digit of the number is cut out in a range based on a reference vertical position in the image.
Of the plurality of cut out drum images, for the drum image corresponding to at least the last digit, the template image in the range in which the drum image is most likely to match in the template image which is the developed image of the drum stored in advance. Determine the position of the drum in the direction of rotation within
Based on the determined position, the intermediate value between the numbers on the drum is calculated with the set number of digits of 2 or more.
A computer program that executes a process of outputting a number having a number of digits larger than the number of digits of the mechanical counter based on the value of the number shown in the drum image of each digit cut out and the intermediate value.

Images