Imaging apparatus and imaging method

The photographing device addresses the challenge of accurately reading numerical values by capturing multiple images under varied lighting conditions and using OCR to combine and identify digits, effectively reducing light interference.

JP2024158116APending Publication Date: 2024-11-08SHARP KK

Patent Information

Application Number
JP2023073055
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-04-27
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

Existing photographing devices struggle to accurately specify numerical values from images taken by a camera due to interference from reflected light, particularly in environments with varying lighting conditions.

Method used

A photographing device equipped with a camera and multiple lights, which captures multiple images under different lighting conditions, uses OCR (Optical Character Recognition) to generate and combine digits from these images, minimizing the impact of reflected light by identifying and extracting areas with less noise.

Benefits of technology

The device accurately specifies numerical values by reducing the influence of reflected light, ensuring high accuracy in recognizing multi-digit numbers from meter readings.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024158116000001_ABST
    Figure 2024158116000001_ABST
Patent Text Reader

Abstract

To provide an imaging apparatus capable of accurately identifying numerical values from a photographic image captured by a camera.SOLUTION: An imaging apparatus 100 comprises a camera 153, a plurality of lights 1551, 1552, and a control unit 151. The control unit acquires a plurality of photographic images by captured by a camera in a plurality of photography environments in which the lighting states of the plurality of lights are different from each other, and identifies a plurality of digits using an OCR function based on the plurality of photographic images.SELECTED DRAWING: Figure 5
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to a technology for an imaging device for reading displayed numerical values ​​and characters. [Background technology]

[0002] Conventionally, meters for measuring the amount of water, gas, etc., have been installed in homes and offices. For example, JP 2017-3536 A (Patent Document 1) discloses a meter device. According to Patent Document 1, a transparent plate is attached to a cover body to form an enclosed space. A circuit board is attached to the ceiling part of the inner surface of the cover body that forms this enclosed space, a pinhole camera is fixed to the center of the circuit board, a cylinder is lightly pressed into the inside of the cover body so as not to fall out, four grooves are formed in the vertical direction on the outer circumferential surface of the cylinder, a power supply line of a side-emitting type LED is stored in the groove, the upper end of the power supply line is connected to the circuit board, and the LED is held in contact with the upper surface of the transparent plate or the upper end is held by the groove. The LED is arranged on the lower outer side with respect to the camera so that the reflected image on the surface of the transparent plate and the reflected image on the cover plate are not reflected by the camera. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2017-3536 A Summary of the Invention [Problem to be solved by the invention]

[0004] An object of the present invention is to provide an imaging device that can accurately identify a numerical value from an image captured by a camera. [Means for solving the problem]

[0005] According to one aspect of the present invention, there is provided an imaging device including a camera, a plurality of lights, and a control unit. The control unit obtains a plurality of captured images by taking images with the camera in a plurality of imaging environments in which the lighting conditions of the plurality of lights are different, and identifies a multi-digit number based on the plurality of captured images using an OCR function. Effect of the Invention

[0006] As described above, according to the present invention, an imaging device capable of accurately identifying a numerical value from an image captured by a camera is provided. [Brief description of the drawings]

[0007] [Figure 1] FIG. 1 is a conceptual diagram showing an overall configuration of a network system according to a first embodiment. [Diagram 2] FIG. 2 is a conceptual diagram showing a state in which the imaging device according to the first embodiment is attached to a water meter. [Diagram 3] 3A to 3C are conceptual diagrams showing a procedure for attaching the imaging device according to the first embodiment to a water meter. [Figure 4] 3 is a conceptual diagram showing a procedure for attaching the imaging device according to the first embodiment to a gas meter. FIG. [Diagram 5] 1 is a block diagram showing main functions of an imaging device according to a first embodiment. [Figure 6] 5 is a flowchart showing information processing in the imaging device according to the first embodiment. [Figure 7] 3A to 3C are conceptual diagrams showing an operating state of an LED light according to the first embodiment and an image captured by a camera. [Figure 8] 4 is a captured image in which reflected light of an LED light according to the first embodiment is captured. [Figure 9] FIG. 2 is a block diagram showing main functions of a center server according to the first embodiment. [Figure 10] 10 is a flowchart showing information processing in the imaging device according to the second embodiment. [Figure 11] 13A to 13C are conceptual diagrams showing an operating state of an LED light according to a third embodiment and an image captured by a camera. [Figure 12] 13 is a flowchart showing information processing in the imaging device according to the third embodiment. [Figure 13] 13A to 13C are conceptual diagrams showing an operating state of an LED light according to a fourth embodiment and an image captured by a camera. [Figure 14] 13 is a flowchart showing information processing in the imaging device according to the fourth embodiment. [Figure 15] 13A to 13C are conceptual diagrams showing the operating state, recognition result, and accuracy of an LED light according to a fifth embodiment. [Figure 16] 13 is a flowchart showing information processing in the imaging device according to the fifth embodiment. [Figure 17] 13 is a flowchart showing information processing in the imaging device according to the sixth embodiment. [Figure 18] 13 is a flowchart showing information processing in the imaging device according to the seventh embodiment. [Figure 19] 23 is a flowchart showing information processing in the imaging device according to the eighth embodiment. [Figure 20] 23 is a flowchart showing information processing in the imaging device according to the tenth embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0008] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. In the following description, the same components are denoted by the same reference numerals. Their names and functions are also the same. Therefore, detailed description thereof will not be repeated. <First embodiment> <Overall configuration of network system 1>

[0009] First, the overall configuration of a network system 1 according to the present embodiment will be described with reference to Fig. 1. The network system 1 according to the present embodiment mainly includes a meter 200 as a measuring device, an imaging device 100, a network 500, a center-side network control device 400, a center server 300, and the like.

[0010] In this embodiment, the meter 200 is a meter for measuring the amount of water, gas, or the like used, and displays the cumulative amount of usage up to that point as the measurement result.

[0011] The photographing device 100 is a device that can be attached later near the meter 200. As shown in Fig. 2, in this embodiment, the photographing device 100 is attached to face the measuring device in the meter reading box 10, i.e., the numeric display unit 230 of the meter 200, and reads the measurement value displayed by the meter 200 by using a camera and an OCR (Optical Character Reader) function, and transmits the measurement value to the server 300 by using a communication antenna.

[0012] Returning to FIG. 1, the server 300 acquires measurement values ​​from the image capturing device 100 via a network 500 such as the Internet, and calculates the usage and fees for water, gas, etc. for each home, office, or building.

[0013] In this way, even if the meter 200 is not previously connected to the network, the network system 1 according to this embodiment can read the measurement value of the meter 200 and provide it to the server 300 by retrofitting the photographing device 100 to the meter 200. In particular, the photographing device 100 according to this embodiment can photograph the numeric display section 230 of the meter 200 while irradiating the numeric display section 230 with light, and can accurately recognize the numeric value based on the photographed image while reducing the influence of reflected light. Such functions will be described in detail below. <How to attach the imaging device>

[0014] First, with reference to FIG. 3, a case will be described in which the imaging device 100 is attached to a water meter 200A that serves as a meter 200 and that measures the amount of water used.

[0015] 3(A), water meter 200A is preferably placed in meter box 10 that is not exposed to sunlight. Display unit 230 is provided to enable an operator to check the amount of water used when reading the meter. In this embodiment, display unit 230 has white numbers, a black back surface, and a front surface covered with a glass surface.

[0016] In this embodiment, the photographing device 100 is attached to the front of the display panel 231 of the water meter 200A. More specifically, as shown in Figs. 3(A) and 3(B), the photographing device 100 is held in a state in which the camera 153 of the photographing device 100 faces the display unit 230 on the front of the water meter 200A. In this embodiment, the camera 153 is configured to photograph the entire display panel 231 including the display unit 230 of the water meter 200A. Since the entire display panel 231 is photographed, the display unit 230 can be included in the photographed image even if the photographing device 100 is attached in a rotated state. In this state, the photographing device 100 is fixed to the water meter 200A by a jig such as a cable tie, as shown in Fig. 3(C).

[0017] As will be described later, the meter 200 to be imaged is not limited to the water meter 200 A. For example, as shown in Fig. 4, the meter 200 may be a gas meter 200B for measuring the amount of gas used.

[0018] 4(A), the gas meter 200B is also provided with a display unit 230 that is used by an operator to check the amount of gas used when reading the meter. In this embodiment, the numbers on the display unit 230 are white, the back surface is black, and the front surface is covered with a glass surface.

[0019] In this embodiment, the photographing device 100 is attached to the front of the gas meter 200B. More specifically, as shown in Fig. 4(A) and Fig. 4(B), the photographing device 100 is held in a state in which the camera 153 of the photographing device 100 faces the display unit 230 on the front of the gas meter 200B. In this embodiment, the camera 153 is configured to photograph the entire display panel 231 including the display unit 230 of the gas meter 200B. In this state, as shown in Fig. 4(C), the photographing device 100 is fixed to the gas meter 200B with a jig such as a cable tie.

[0020] In particular, in this embodiment, the numerical value displayed on the numerical value display section 230 can be accurately recognized without being hindered by reflected light from the light, as will be described later. <Configuration of imaging device>

[0021] An embodiment of the configuration of the imaging device 100 according to the present embodiment will be described with reference to Fig. 5. The imaging device 100 according to the present embodiment is composed of an OCR meter reading unit 150 for capturing an image of a numeric display section to obtain a numeric value, and an NCU (Network Control Unit) 110 for transmitting the meter reading value to the center server 300.

[0022] First, the OCR meter reading unit 150 mainly includes a microcomputer 151, a memory 152, a camera 153, a communication interface 158, a power supply 159, an operation interface 154, an LED light 155, and the like.

[0023] The microcomputer 151 controls each part of the OCR meter reading unit 150 according to a program stored in the memory 152. Alternatively, the CPU of the microcomputer 151 controls each part of the OCR meter reading unit 150 according to a program in the memory stored in the microcomputer 151. The microcomputer 151 according to this embodiment recognizes a multi-digit numeric string as a meter reading value from an image captured by the camera 153, for example, by executing an application program of an OCR function.

[0024] The memory 152 stores the control program, images captured by the camera 153, the recognition results and matching rates for each digit, the final meter reading value and meter reading date and time, information regarding noise caused by reflected light from the LED light 155, and the like.

[0025] The camera 153 photographs the display unit 230 of the meter 200 located in front of the image capturing device 100 in accordance with instructions from the microcomputer 151 , and stores the photographed image data in the memory 152 .

[0026] The communication interface 158 transmits meter reading values, reflected light noise information, and the like to the NCU 110 in accordance with instructions from the microcomputer 151, and receives commands to obtain meter reading values ​​from the NCU 110 and inputs them to the microcomputer 151.

[0027] The power supply 159 is realized by various batteries and supplies power to each part of the OCR meter reading unit 150.

[0028] The operation interface 154 is realized by various switches and sensors, and receives instructions from an administrator or a service person and inputs them to the microcomputer 151 .

[0029] The LED light 155 includes a plurality of LED lights 1551 and 1552 as described below, and emits light based on instructions from the microcomputer 151.

[0030] On the other hand, the NCU 110 mainly includes a communication interface 118, a control microcomputer 111, a memory 112, a communication module 116, a wireless communication antenna 117, a power supply 119, an operation interface 114, an LED light 115, and the like.

[0031] The microcomputer 111 controls each part of the NCU 110 according to a program stored in the memory 112. Alternatively, the CPU of the microcomputer 111 controls each part of the NCU 110 according to a program in the memory stored in the microcomputer 111. The microcomputer 111 according to this embodiment executes a meter reading program, for example, to cause the OCR meter reading unit 150 to photograph the meter reading value and upload the meter reading value to the server 300.

[0032] The memory 112 stores a control program, a history of meter reading values ​​obtained from the OCR meter reading unit 150, information on reflected light, and the like.

[0033] The communication interface 118 is realized by a connector of the wired cable 130 or the like, and performs data communication with the OCR meter reading unit 150. The communication interface 118 transmits a command to obtain a meter reading value to the OCR meter reading unit 150 according to an instruction from the microcomputer 111, and receives information such as the meter reading value and noise of reflected light from the OCR meter reading unit 150.

[0034] The communication module 116 is a communication module for a wireless communication antenna 117, and transmits meter reading values ​​to the server 300 via a base station and receives various instructions from the server 300 in accordance with instructions from the microcomputer 111.

[0035] The power supply 119 is realized by various batteries and supplies power to each part of the NCU 110.

[0036] The operation interface 114 is realized by various switches and sensors, and receives instructions from an administrator or a service person and inputs them to the microcomputer 111 .

[0037] The LED light 115 displays the control state, communication state, operation state, etc. of the NCU 110 according to instructions from the microcomputer 111 .

[0038] Note that the meter 200 to be photographed measures the usage amounts of water and gas and displays the cumulative usage amounts up to the present on the numerical display unit 230 of the display panel 231. <Operation of the OCR Meter Reading Unit>

[0039] Next, with reference to FIGS. 6 and 7, the information processing of the OCR meter reading unit 150 according to the present embodiment will be described. When the microcomputer 151 receives an instruction to photograph the display panel 231 of the meter 200 from the NCU 110 via the communication interface 158, as shown in FIG. 7(A), for example, while one of the LED lights 1551 on the left side is lit (step S102), the camera 153 is made to take a photograph (step S104). Note that even if the photographing device 100 is attached to the meter 200 in a rotated state in order to photograph the entire display panel 231, the entire display unit 230 can be included in the photographed image.

[0040] The microcomputer 151 analyzes the photographed image and identifies, as an area with less influence of reflected light, an area on the side with less noise, which is often the area 1532 farther from the LED light 1551 (step S106).

[0041] Here, the microcomputer 151 determines that the noise is large when there is a large set portion of bright pixels equal to or greater than a predetermined value. More specifically, as shown in FIG. 8(A), the microcomputer 151 searches for a block in which pixels with a luminance equal to or greater than a predetermined value are continuous based on the grayscale image, and determines that the area with a block of a size equal to or greater than a predetermined size has large noise, or determines that the area without a large-sized white block has small noise.

[0042] Alternatively, as shown in FIG. 8(B), the microcomputer 151 creates a binary image by inverting black and white, searches for a block of black pixels, and determines that the area with a black block of a size equal to or greater than a predetermined size has large noise, or determines that the area with only small-sized black blocks has small noise.

[0043] The method for identifying areas where the influence of reflected light is large and areas where the influence of reflected light is small is not particularly limited.

[0044] Returning to FIG. 6, the microcomputer 151 cuts out the image data on the side with less noise and stores it in the memory 152 (step S108).

[0045] Next, as shown in FIG. 7B, the microcomputer 151 turns on the other LED light, for example the right LED light 1552 (step S112) and causes the camera 153 to take an image (step S114).

[0046] The microcomputer 151 analyzes the captured image and identifies the area 1531 on the side with less noise as an area less affected by reflected light (step S116).

[0047] The microcomputer 151 cuts out the image data on the side with less noise and stores it in the memory 152 (step S118).

[0048] As shown on the right side of FIG. 7C, the microcomputer 151 combines the image data with less noise stored in the memory 152 to generate one image data of the original size (step S122).

[0049] The microcomputer 151 uses the OCR function to identify a multi-digit number from the image data as the meter reading value (step S124), and transmits the meter reading value to the NCU 110 via the communication interface 158 (step S126). <Center server configuration>

[0050] Next, a description will be given of the configuration of the center server 300. As shown in Fig. 9, the center server 300 includes a CPU 310, a memory 320, an operation unit 340, and a communication interface 360 ​​as main components.

[0051] CPU 310 controls each unit of server 300 by executing a program stored in memory 320. For example, CPU 310 executes a program stored in memory 320 to perform various processes while referring to various data.

[0052] Memory 320 is realized by various RAMs, various ROMs, etc., and may be included in server 300, may be detachable from various interfaces of server 300, or may be a recording medium of another device accessible from server 300. Memory 320 stores programs executed by CPU 310, data generated by execution of programs by CPU 310, input data, other databases used for the service according to this embodiment, and the like.

[0053] For example, the memory 320 stores meter reading data for each home. The meter reading data includes, for each customer, the identification information of the customer, the customer's address, the usage history, the worker in charge, and the communication address of the terminal, and other correspondences.

[0054] The operation unit 340 receives commands from a service manager, operator, service technician, etc., and inputs the commands to the CPU 310 .

[0055] The communication interface 360 ​​transmits data from the CPU 310 to the center-side network control device 400, the photographing device 100, and other devices via the Internet, a carrier network, a router, etc. Conversely, the communication interface 360 ​​receives data from the center-side network control device 400, the photographing device 100, and other devices via the Internet, a carrier network, a router, etc., and passes the data to the CPU 310. <Second embodiment>

[0056] In the above embodiment, an area less affected by reflected light is specified for each image capture, and an image of that area is extracted. However, it is also possible to first capture images under a plurality of conditions, and then to specify an area less affected by reflected light for each captured image, and extract an image of that area.

[0057] In this embodiment, as shown in FIG. 10, when the microcomputer 151 receives a command from the NCU 110 via the communication interface 158 to photograph the display panel 231 of the meter 200, as shown in FIG. 7(A), the microcomputer 151 causes the camera 153 to photograph one side, for example the left LED light 1551, with the light turned on (step S202), and stores the photographed image in the memory 152 as a first photographed image (step S204).

[0058] Next, as shown in FIG. 7(B), the microcomputer 151 turns on the other LED light 1552, for example the right LED light (step S212), and causes the camera 153 to take a photograph, thereby storing the photographed image in the memory 152 as a second photographed image (step S214).

[0059] The microcomputer 151 analyzes the first captured image and identifies the side with less noise as an area less affected by reflected light (step S222). The microcomputer 151 cuts out the image on the side with less noise and stores it in the memory 152 (step S224).

[0060] The microcomputer 151 analyzes the second captured image and identifies the side with less noise as an area less affected by reflected light (step S226). The microcomputer 151 cuts out the image on the side with less noise and stores it in the memory 152 (step S228).

[0061] The microcomputer 151 combines the images with less noise stored in the memory 152 to generate one image data (step S232).

[0062] The microcomputer 151 uses the OCR function to identify a multi-digit number from the image data as the meter reading value (step S234), and transmits the meter reading value to the NCU 110 via the communication interface 158 (step S236). <Third embodiment>

[0063] In the above embodiment, the OCR meter reading unit 150 is provided with the LED lights 1551 and 1552 on the left and right sides of the camera 153. However, three or more LED lights 1551 and 1552 may be provided.

[0064] In the above embodiment, the OCR meter reading unit 150 extracts half the image with less noise from the captured image. However, the OCR meter reading unit 150 may extract an image of an area smaller than half the captured image and combine them.

[0065] In this embodiment, as shown in Fig. 11, a form will be described in which the OCR meter reading unit 150 has four LED lights 1551, 1552, 1553, and 1554 around the camera 153. Also, a form will be described in which the OCR meter reading unit 150 divides the captured image into four areas 1531, 1532, 1533, and 1534 corresponding to the positions of the LED lights 1551, 1552, 1553, and 1554, and determines the area with less noise.

[0066] The information processing of the OCR meter reading unit 150 according to this embodiment will be described with reference to Fig. 12. When the microcomputer 151 receives a command to capture an image of the display panel 231 of the meter 200 from the NCU 110 via the communication interface 158, it causes the camera 153 to capture an image (step S104) with the upper left LED light 1551 and the lower right LED light 1554 turned on (step S102), as shown in Fig. 11(A).

[0067] The microcomputer 151 analyzes the captured image and identifies the upper right area 1532 and the lower left area 1533, which have low noise, as areas that are less affected by reflected light (step S106). In many cases, the reflected light is small in areas where the front light is not turned on.

[0068] The microcomputer 151 cuts out image data of the upper right area 1532 and the lower left area 1533, which have small noise, and stores them in the memory 152 (step S108).

[0069] Next, as shown in FIG. 11(B), the microcomputer 151 turns on the upper right LED light 1552 and the lower left LED light 1553 (step S112) and causes the camera 153 to take a picture (step S114).

[0070] The microcomputer 151 analyzes the captured image and identifies the upper left area 1531 and the lower right area 1534, which have low noise, as areas less affected by reflected light (step S116).

[0071] The microcomputer 151 cuts out image data of the upper left area 1531 and the lower right area 1534, which have small noise, and stores them in the memory 152 (step S118).

[0072] As shown in FIG. 11C, the microcomputer 151 combines the image data of the four areas stored in the memory 152 to create one piece of image data of the original size (step S122).

[0073] The microcomputer 151 uses the OCR function to identify a multi-digit number from the image data as the meter reading value (step S124), and transmits the meter reading value to the NCU 110 via the communication interface 158 (step S126). <Fourth embodiment>

[0074] In the above embodiment, as shown in Figures 2, 3, 7, and 11, the camera 153 of the OCR meter reading unit 150 is configured to capture an image of the entire display panel 231 including the display unit 230 of the meter 200. However, as shown in Figure 13, the camera 153 may be configured to capture an image of only the vicinity of the display unit 230 of the meter 200.

[0075] As shown in FIG. 14, when the microcomputer 151 receives a command from the NCU 110 via the communication interface 158 to photograph the display unit 230 of the meter 200, it causes the camera 153 to photograph the display unit 230 of the meter 200 (step S104) while turning on one side of the LED light 1551, for example on the left side, as shown in FIG. 13(A) (step S102).

[0076] The microcomputer 151 analyzes the captured image and identifies the side with less noise as an area less affected by reflected light (step S106).

[0077] The microcomputer 151 cuts out the image data on the side with less noise and stores it in the memory 152 (step S108).

[0078] Next, as shown in FIG. 13(B), the microcomputer 151 causes the camera 153 to capture an image of the display unit 230 of the meter 200 (step S114) while turning on the other LED light, for example the right LED light 1552 (step S112).

[0079] The microcomputer 151 analyzes the captured image and identifies an area 1534 with low noise as an area that is less affected by reflected light (step S116).

[0080] The microcomputer 151 cuts out the image data on the side with less noise and stores it in the memory 152 (step S118).

[0081] As shown in FIG. 13(C), the microcomputer 151 combines the image data stored in the memory 152 to create one image data of the original size (step S122).

[0082] The microcomputer 151 uses the OCR function to identify a multi-digit number from the image data as the meter reading value (step S124), and transmits the meter reading value to the NCU 110 via the communication interface 158 (step S126). <Fifth embodiment>

[0083] In the above embodiment, images with less noise are extracted first, and then images are created by combining these images, and then the meter reading value consisting of a multi-digit number is recognized. However, it is also possible to first recognize the numbers from the captured image, and then extract and combine only the numbers with high resolution, accuracy, and matching rate to specify the meter reading value consisting of a multi-digit number.

[0084] Hereinafter, the information processing of the OCR meter reading unit 150 according to this embodiment will be described with reference to Fig. 15 and Fig. 16. When the microcomputer 151 receives a command to capture an image of the display unit 230 of the meter 200 from the NCU 110 via the communication interface 158, it causes the camera 153 to capture an image (step S304) with the left LED light 1551 and the right LED light 1552 turned on (step S302), as shown in Fig. 15(A).

[0085] The microcomputer 151 uses the OCR function to analyze the numbers (step S306). At this time, the microcomputer 151 also obtains the accuracy, likelihood, matching rate, etc. of the recognition result for each number. Note that it is preferable that the number of digits of the target meter reading value is set in advance in the program.

[0086] As shown in FIG. 15(A), the microcomputer 151 determines whether or not the accuracy is at a predetermined level, for example 80% or higher, for each number, that is, for each digit (step S308).

[0087] The microcomputer 151 determines whether the accuracy of all the digits is equal to or higher than a predetermined level (step S310).

[0088] If the accuracy of any of the numbers related to the photograph for all digits is above a predetermined level (YES in step S310), the microcomputer 151 identifies a multi-digit number combining numbers with high accuracy as the meter reading value (step S312) and transmits the meter reading value to the NCU 110 via the communication interface 158 (step S314).

[0089] If a number with a high degree of certainty has not yet been recognized for any of the digits (NO in step S310), the microcomputer 151 changes the lighting conditions (step S322). For example, as shown in FIG. 15B, the microcomputer 151 turns on the left LED light 1551 (step S322) and causes the camera 153 to take a picture (step S324).

[0090] The microcomputer 151 uses the OCR function to analyze the numbers (step S326). At this time, the microcomputer 151 also obtains the accuracy, likelihood, matching rate, and the like of the recognition result for each number.

[0091] As shown in FIG. 15(B), the microcomputer 151 determines for each number, that is, for each digit, whether or not the accuracy is at a predetermined level, for example, 80% or higher (step S328).

[0092] The microcomputer 151 determines whether the accuracy of all the digits is equal to or higher than a predetermined level (step S330).

[0093] If the accuracy of any of the numbers related to the photograph for all digits is above a predetermined level (YES in step S330), the microcomputer 151 identifies a multi-digit number combining numbers with high accuracy as the meter reading value (step S312) and transmits the meter reading value to the NCU 110 via the communication interface 158 (step S314).

[0094] If a number with a high degree of certainty has not yet been recognized for any of the digits (NO in step S330), microcomputer 151 further changes the lighting conditions (step S342). For example, as shown in FIG. 15C, microcomputer 151 turns on right LED light 1552 (step S342) and causes camera 153 to take a picture (step S344).

[0095] The microcomputer 151 uses the OCR function to analyze the numbers (step S346). At this time, the microcomputer 151 also obtains the accuracy, likelihood, matching rate, and the like of the recognition result for each number.

[0096] As shown in FIG. 15(C), the microcomputer 151 determines whether or not the accuracy is at a predetermined level, for example 80% or higher, for each number, that is, for each digit (step S348).

[0097] The microcomputer 151 determines whether the accuracy of all the digits is equal to or higher than a predetermined level (step S350).

[0098] If the accuracy of any of the numbers related to the photograph for all digits is above a predetermined level (YES in step S350), the microcomputer 151 identifies a multi-digit number combining numbers with high accuracy as the meter reading value (step S312) and transmits the meter reading value to the NCU 110 via the communication interface 158 (step S314).

[0099] If a number with a high degree of certainty has not yet been recognized for any of the digits (NO in step S350), the microcomputer 151 repeats the process from step S302. Sixth embodiment

[0100] If a number with a high degree of certainty has not yet been recognized for any of the digits (NO in step S350), the microcomputer 151 may cause the camera 153 to capture an image (step S364) while changing the illumination conditions (step S362) as shown in Fig. 17. For example, the microcomputer 151 may cause the camera 153 to capture an image (step S364) while turning on both LED lights 1551 and 1552 and lowering the illuminance of one or both of them (step S362). Then, it is preferable that the microcomputer 151 repeats the capturing of images by the camera 153 and the recognition of numbers by OCR while changing the illumination conditions until the certainty of all the digits becomes equal to or higher than a predetermined level. <Seventh embodiment>

[0101] Alternatively, if a number with a high degree of certainty has not yet been recognized for any of the digits (NO in step S350), as shown in FIG. 18, the microcomputer 151 may send data to the NCU 110 via the communication interface 158 indicating that the meter reading failed (step S372). <Eighth embodiment>

[0102] 19, when the accuracy of the numbers is equal to or higher than a predetermined level (steps S308 and S310), the microcomputer 151 may store the lighting state and illuminance of each of the LED lights 1551, 1552, etc. as the imaging conditions at that time in the memory 152 (step S380). Then, thereafter, instead of steps S302, 322, 342, etc., the microcomputer 151 may take images by controlling the LED lights 1551, 1552, etc. to the lighting state and illuminance with which the accuracy of the meter reading value recognition result was high. In other words, thereafter, it is preferable that the microcomputer 151 recognizes the numbers for each digit while taking images with lighting states and illuminance with which the accuracy of the number recognition is high, without taking images with lighting states and illuminance with which the accuracy of the number recognition is low.

[0103] Alternatively, the lighting state of the LED light 155 and the digits of the numerical value to be recognized may be set in advance. For example, the microcomputer 151 may recognize the two-digit numerical value on the right side when the left LED light 1551 is turned on, and may recognize the two-digit numerical value on the left side when the right LED light 1552 is turned on. <Ninth embodiment>

[0104] Similarly, in the above-mentioned form in which images are first combined, it is also preferable that the microcomputer 151 stores the areas with less reflected light, the lighting state, and the illuminance, and thereafter photographs are taken under the lighting state and illuminance that provided a high recognition accuracy, and an image of the corresponding area is cut out for each photograph, and these cut-out images are combined before recognizing the multi-digit number.

[0105] Alternatively, the lighting state of the LED light 155 and the area to be extracted from the captured image may be set in advance. For example, the microcomputer 151 may always extract an image on the opposite side of the lit LED light 155, combine multiple images thus obtained, and recognize a multi-digit number from the combined image. <Tenth embodiment>

[0106] In the above embodiment, the OCR meter reading unit 150 and the NCU 110 are configured as separate units and communicate with each other. However, both may be included in the same device. In other words, the imaging device may be the NCU, or the NCU may include the OCR meter reading unit 150.

[0107] In the above embodiment, the OCR meter reading unit 150 performs illumination, photography, and numerical recognition, but some or all of these functions may be mounted on the NCU 110 or the server 300. For example, the illumination function and numerical recognition function of the OCR meter reading unit 150 may be mounted on the NCU 110, and only a camera having a photography function may be provided on a device other than the NCU 110. Alternatively, a device having a camera and illumination may send photographed image data to the NCU 110, so that the NCU 110 executes numerical recognition, accuracy calculation, etc., or the server 300 executes numerical recognition, accuracy calculation, etc., by sending the photographed image data to the server 300.

[0108] For example, the NCU 110 may control the LED lights 1551, 1552, etc. and the camera 153 of the OCR meter reading unit 150 to perform the following processing, or the NCU 110 itself may be equipped with the LED lights 1551, 1552, etc. and the camera 153 to perform the following processing.

[0109] As shown in FIG. 20, the microcomputer 111 of the NCU 110 causes the camera 153 to capture an image (step S504) while turning on the LED lights 1551, 1552, . . . under a first illumination condition (step S502) via the communication interface 118.

[0110] The microcomputer 111 acquires the captured image from the camera 153 via the communication interface 118, and analyzes the numbers from the captured image using the OCR function (step S506). At this time, the microcomputer 111 also acquires the accuracy of the recognition result for each number. Note that the number of digits of the meter reading value to be targeted is set in advance in the program.

[0111] As shown in FIG. 15(A), the microcomputer 111 determines whether or not the accuracy is at a predetermined level, for example 80% or higher, for each number, that is, for each digit (step S508).

[0112] The microcomputer 111 determines whether the accuracy of all the digits is equal to or higher than a predetermined level (step S510).

[0113] If the accuracy of any of the numbers related to the photograph for all digits is above a predetermined level (YES in step S510), the microcomputer 111 identifies a multi-digit number combining numbers with high accuracy as the meter reading value (step S512) and transmits the meter reading value to the center server 300 via the communication module 116 (step S514).

[0114] If a number with a high degree of certainty has not yet been recognized for any of the digits (NO in step S510), the microcomputer 111 changes the lighting conditions to LED lights 1551, 1552, etc. via the communication interface 118 (step S522) and causes the camera 153 to take a photo (step S524).

[0115] The microcomputer 111 uses the OCR function to analyze the numbers from the captured image (step S526). At this time, the microcomputer 111 obtains the accuracy of the recognition result for each number.

[0116] As shown in FIG. 15(B), the microcomputer 111 determines whether the accuracy is at a predetermined level, for example 80% or higher, for each number, that is, for each digit (step S528).

[0117] The microcomputer 111 determines whether the accuracy of all the digits is equal to or higher than a predetermined level (step S530).

[0118] If the accuracy of any of the numbers related to the photograph for all digits is above a predetermined level (YES in step S530), CU111 identifies a multi-digit number combining numbers with high accuracy as the meter reading value (step S512) and transmits the meter reading value to the center server 300 via the communication module 116 (step S514).

[0119] If a number with a high degree of certainty has not yet been recognized for any of the digits (NO in step S530), microcomputer 111 further changes the lighting conditions of LED lights 1551, 1552, etc. (step S542) and causes camera 153 to take a photo (step S544).

[0120] The microcomputer 111 uses the OCR function to analyze the numbers from the captured image (step S546). At this time, the microcomputer 111 also obtains the accuracy of the recognition result for each number.

[0121] As shown in FIG. 15(C), the microcomputer 111 determines whether the accuracy is at a predetermined level, for example 80% or higher, for each number, that is, for each digit (step S548).

[0122] The microcomputer 111 determines whether the accuracy of all the digits is equal to or higher than a predetermined level (step S550).

[0123] If the accuracy of any of the numbers related to the photograph for all digits is above a predetermined level (YES in step S550), the microcomputer 111 identifies a multi-digit number combining numbers with high accuracy as the meter reading value (step S512) and transmits the meter reading value to the center server 300 via the communication module 116 (step S514).

[0124] If a number with a high degree of certainty has not yet been recognized for any of the digits (NO in step S550), microcomputer 111 may further change the lighting conditions of LED lights 1551, 1552, etc. (step S362) and cause camera 153 to take a photograph (step S364).

[0125] Such processing may be executed by the CPU 310 of the center server 300. <Tenth embodiment>

[0126] In the above embodiment, the subject of imaging is the display unit of a meter reading device such as a water meter or a gas meter, but the subject of imaging may be a display unit other than a meter reading device.

[0127] In addition, in the above embodiment, the subject to be photographed is a multi-digit Arabic number such as a meter reading value, but the subject to be photographed may be another type of number, or may be letters or symbols. <Summary>

[0128] In the above embodiment, a photographing device is provided that includes a camera, a plurality of lights, and a control unit. The control unit obtains a plurality of photographed images by photographing with the camera under a plurality of photographing environments in which the lighting conditions of the plurality of lights are different, and identifies a multi-digit number based on the plurality of photographed images using an OCR function.

[0129] Preferably, the control unit identifies numbers using an OCR function for each of the plurality of captured images, and identifies the multi-digit number by combining numbers with good recognition results.

[0130] Preferably, the control unit stores one or more shooting environments in which numbers are easily recognized using the OCR function, and thereafter, takes pictures with the camera under the one or more shooting environments.

[0131] Preferably, the control unit extracts an image of an area where light is not reflected much from each of the multiple captured images, and identifies a multi-digit number using an OCR function for an image obtained by combining these images.

[0132] Preferably, the control unit stores one or more shooting environments in which light from the light source is less reflected in a captured image, and thereafter, the control unit performs shooting with the camera under the one or more shooting environments.

[0133] Preferably, the control unit executes shooting with the camera while changing the lighting states of the plurality of lights until all of the multi-digit numbers can be identified.

[0134] In the above embodiment, a photographing method is provided that includes the steps of photographing with a camera with a first light of the multiple lights turned on, photographing with a camera with a second light of the multiple lights turned on, and a step in which a control unit identifies a multi-digit number using an OCR function based on the multiple photographed images.

[0135] The embodiments disclosed herein should be considered to be illustrative and not restrictive in all respects. The scope of the present invention is defined by the claims, not the above description, and is intended to include all modifications within the scope and meaning equivalent to the claims. [Explanation of symbols]

[0136] 1: Network system 100: Imaging device 110 :NCU 111: Microcomputer 112: Memory 114: Operation interface 115: LED light 116: Communication module 117: Wireless communication antenna 118: Communication interface 119: Power supply 130: Wired cable 150: OCR Meter Reading Unit 151: Microcomputer 152: Memory 153: Camera 154: Operation interface 155: LED light 158: Communication interface 159: Power supply 200: Meter 200A: Water meter 200B: Gas meter 230: Numeric display section 231:Display panel 300: Center server 310: CPU 320: Memory 322: Step 340:Operation unit 342: Step 360: Communication interface 400: Center side network control device 500 : Network 1551: LED light 1552: LED light 1553: LED light 1554: LED light

Claims

1. Camera and Multiple lights and A control unit, The control unit of the photographing device obtains a plurality of photographed images by taking pictures with the camera in a plurality of photographing environments in which the lighting states of the plurality of lights are different, and identifies a multi-digit number based on the plurality of photographed images using an OCR function.

2. The photographing device according to claim 1 , wherein the control unit identifies numbers for each of the plurality of photographed images using the OCR function, and identifies the multi-digit number by combining numbers with good recognition results.

3. 3. The photographing device according to claim 2, wherein the control unit stores one or more photographing environments in which numbers are easily recognized using the OCR function, and thereafter photographs are taken with the camera under the one or more photographing environments.

4. The photographing device according to claim 1 , wherein the control unit extracts an image of an area where the light of the light is less reflected from each of the plurality of photographed images, and identifies the multi-digit number using the OCR function for an image obtained by combining the extracted images.

5. The photographing device according to claim 4 , wherein the control unit stores one or more photographing environments in which the light of the light is less reflected in a photographed image, and thereafter photographs are taken with the camera under the one or more photographing environments.

6. The photographing device according to claim 1 , wherein the control unit executes photographing with the camera while changing the lighting states of the plurality of lights until all of the multi-digit numbers can be identified.

7. taking an image with a camera while turning on a first light of the plurality of lights; taking an image with the camera while turning on a second light of the plurality of lights; The control unit identifies a multi-digit number using an OCR function based on a plurality of captured images.

Citation Information

Patent Citations

  • Meter device

    JP2017003536A

Cited By

  • Magnetic disk device

    CN121662093A