Examination assistance device, examination assistance system, and examination assistance method

Abstract

An examination assistance device (300) acquires: a target image, which is an image of a label (94) attached to a piece of target equipment (44) imaged by a portable terminal (400); and a target state, which is the imaging state of the portable terminal (400) when the target image is imaged. The examination assistance device (300) extracts, as a target character string, a character string of the model name of the target equipment (44) from the target image. The examination assistance device (300) generates a corrected character string obtained by correcting the target character string, on the basis of the type of the target equipment (44), which is estimated on the basis of the target state, and information relating to the model name of the equipment. The examination assistance device (300) transmits the corrected character string to the portable terminal (400).

Description

Survey support device, survey support system, and survey support method

[0001] The present disclosure relates to an investigation support device, an investigation support system, and an investigation support method that support the investigation of facilities installed in a building.

[0002] When considering replacing air conditioning, ventilation, lighting, or other equipment installed in a building, an inspector visits the building to inspect the existing equipment. At that time, the inspector checks the model name and serial number written on the nameplate attached to the equipment to confirm the existing equipment.

[0003] International Publication No. 2021 / 229795 (Patent Document 1) discloses a maintenance management work support system that photographs nameplates in building facilities with a mobile device camera and converts the characters written on the nameplates into character codes using an OCR (Optical Character Recognition) service.

[0004] On the other hand, Japanese Patent Application Laid-Open No. 2002-373197 (Patent Document 2) discloses a technology for correcting garbled characters and misrecognized data, such as addresses, during OCR reading and analysis using misrecognition correction patterns from a pre-registered place name database, etc.

[0005] International Publication No. 2021 / 229795 Japanese Patent Application Laid-Open No. 2002-373197

[0006] When using the maintenance management work support system disclosed in Patent Document 1 to photograph labels such as nameplates attached to equipment with a mobile device camera and attempt to extract character strings such as model names from the photographed label image, erroneous character string recognition can occur. Incorrect character strings must be manually corrected, but this correction process is time-consuming.

[0007] Patent Literature 2 discloses a technology for automatically correcting misrecognition of extracted character strings. However, no consideration has been given to how to correct character strings extracted from photographed images of labels attached to building facilities, and there is still room for improvement in this correction method.

[0008] The present disclosure has been made to solve such problems, and the purpose of the present disclosure is to provide an investigation support technology that can accurately correct character strings extracted from photographed images of labels attached to equipment.

[0009] The investigation support device of the present disclosure is a device that supports the investigation of equipment installed in a building. A label attached to the equipment indicates the model name of the equipment. The investigation support device includes a storage device, an acquisition device, and a processing device. The storage device stores information regarding the model name of the equipment associated with the type of equipment. The acquisition device acquires a target image, which is an image of the label attached to the target equipment, captured by a mobile device equipped with a camera, and a target state, which is the photographed state of the mobile device when the target image was captured. The processing device processes the target image. The processing device extracts a character string of the model name of the target equipment from the target image as a target string. The processing device generates a corrected string by correcting the target string based on the type of target equipment estimated based on the target state and information regarding the model name of the equipment. The processing device transmits the corrected string to the mobile device.

[0010] The investigation support system of the present disclosure is a system that supports investigations of equipment installed in a building. A label attached to the equipment indicates the model name of the equipment. The investigation support system includes a mobile terminal, a storage device, an acquisition device, and a processing device. The mobile terminal is provided with a camera. The storage device stores information regarding the model name of the equipment associated with the type of equipment. The acquisition device acquires a target image, which is an image of a label attached to a target equipment photographed by the mobile terminal, and a target state, which is the photographed state of the mobile terminal when the target image was photographed. The processing device processes the target image. The processing device extracts a character string of the model name of the target equipment from the target image as a target string. The processing device generates a corrected string by correcting the target string based on the type of target equipment estimated based on the target state and information regarding the model name of the equipment. The processing device transmits the corrected string to the mobile terminal. The mobile terminal displays the corrected string.

[0011] The investigation support method disclosed herein is a method executed by a computer to support an investigation of equipment installed in a building. A label attached to the equipment describes a model name of the equipment. The investigation support method includes the steps of: storing information about the model name of the equipment associated with a type of equipment; acquiring a target image, which is an image of the label attached to the target equipment captured by a mobile device equipped with a camera, and a target state, which is the photographed state of the mobile device when the target image was captured; extracting a character string representing the model name of the target equipment from the target image as a target string; generating a corrected character string by correcting the target string based on the type of the target equipment estimated based on the target state and the information about the model name of the equipment; and transmitting the corrected character string to the mobile device.

[0012] According to the present disclosure, it is possible to accurately correct character strings extracted from a photographed image of a label attached to equipment.

[0013] FIG. 1 is an overall configuration diagram of a research support system according to a first embodiment. FIG. 2 is a hardware configuration diagram of the research support system. FIG. 3 is a functional block diagram of the research support system. FIG. 4 is a diagram showing an example of model name information. FIG. 5 is a diagram showing an example of serial number information. FIG. 6 is a diagram for explaining shooting state correspondence information. FIG. 7 is a flowchart of processing executed by the research support system. FIG. 8 is a flowchart of first processing. FIG. 9 is a flowchart of second processing. FIG. 10 is a functional block diagram of the research support system according to the second embodiment. FIG. 11 is a flowchart of processing executed by the research support system. FIG. 12 is a functional block diagram of the research support system according to the third embodiment. FIG. 13 is a flowchart of processing executed by the research support system. FIG. 14 is a functional block diagram of the research support system according to the fourth embodiment. FIG. 15 is a flowchart of processing executed by the research support system. FIG. 16 is a functional block diagram of the research support system according to the fifth embodiment. FIG. 17 is a diagram showing an example of a nameplate. FIG. 18 is a diagram showing an example of a nameplate.

[0014] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. While several embodiments will be described below, it was originally intended that the configurations described in each embodiment be combined as appropriate. Note that identical or corresponding parts in the drawings will be designated by the same reference numerals, and their description will not be repeated.

[0015] [First Embodiment] Fig. 1 is an overall configuration diagram of a survey support system 100 according to the first embodiment. The survey support system 100 is a system that supports surveys of equipment (equipment 41 to 44 described below) installed in a building 1. The building 1 may be any type of building, such as an office building, commercial building, factory, school, or hospital. The types of equipment installed in the building 1 include lighting, air conditioning equipment (indoor units, outdoor units), and ventilation equipment.

[0016] In this embodiment, a plurality of pieces of equipment, such as equipment 44, are installed in building 1. A nameplate 94 is attached to equipment 44 as a label on which information about equipment 44 is written. The nameplate 94 lists, as information about equipment 44, a model name T1 of equipment 44, a serial number N1 of equipment 44, and information that the manufacturer of equipment 44 is "Company A." The model name T1 of equipment 44 is "AB-IOE45FG." The serial number N1 of equipment 44 is "1AW23456."

[0017] The investigation support system 100 includes a mobile terminal 400 and an investigation support device 300. The mobile terminal 400 is a terminal used by an equipment investigator 5 to investigate equipment such as equipment 44 installed in a building 1, and includes a camera 427 that photographs nameplates (labels) for the investigation of the equipment.

[0018] The researcher 5 photographs the nameplate 94 with the camera 427 of the mobile terminal 400. The mobile terminal 400 can be wirelessly connected to the research support device 300 via the network NW. The mobile terminal 400 transmits the nameplate image (label image) of the nameplate 94 photographed by the camera 427 to the research support device 300.

[0019] The research support device 300 uses an OCR function to extract the character string of the model name T1 and the character string of the serial number N1 from the nameplate image. The research support device 300 has a function to automatically correct character strings extracted due to erroneous recognition. The display unit 432 of the mobile terminal 400 displays the corrected character strings of the model name T1 and the serial number N1. Details of the processing executed by the research support system 100 will be described using Figures 3 and subsequent figures.

[0020] 2 is a hardware configuration diagram of the research support system 100. The research support device 300 is a computer including a processor 321 as a processing device, a memory 322, a storage device 323, and a communication device 324 as an acquisition device. These are connected to each other via a bus 326 so that they can communicate with each other.

[0021] The processor 321 is, for example, a central processing unit (CPU). The memory 322 may be configured to include read-only memory (ROM) and random access memory (RAM). The storage device 323 is a non-volatile storage device, and may be, for example, a hard disk drive (HDD) or a solid state drive (SSD).

[0022] The processor 321 loads a program stored in the ROM or storage device 323 into the RAM and executes it to realize various functions of the research support device 300. The ROM or storage device 323 stores a program that describes the processing procedures of the research support device 300. The storage device 323 stores data such as the read result information 121 described below.

[0023] The RAM serves as a working area when the processor executes a program, and temporarily stores the program, data required for executing the program, etc. The survey support device 300 can be wirelessly connected to the mobile terminal 400 via the communication device 324.

[0024] The mobile terminal 400 is a mobile terminal such as a smartphone. The mobile terminal 400 is a computer including a processor 421, a memory 422, a storage device 423, a communication device 424, a camera 427, a gyro sensor 428 as a sensor, a GPS (Global Positioning System) receiver 429, an input unit 431, and a display unit 432. These are connected to each other via a bus 426 so as to be able to communicate with each other.

[0025] The processor 421 is, for example, a CPU. The memory 422 may be configured to include a ROM and a RAM. The storage device 423 may be a non-volatile storage device, for example, an HDD or SSD. The processor 421 loads a program stored in the ROM or storage device 423 into the RAM and executes it to realize various functions of the mobile terminal 400. The ROM or storage device 423 stores a program in which processing procedures for the mobile terminal 400 are written. The mobile terminal 400 can be wirelessly connected to the research support device 300 via the communication device 424.

[0026] The display unit 432 is, for example, a display. The input unit 431 is, for example, a keyboard, a mouse, or a touch panel display integrated with the display unit 432. The camera 427 captures an image of a nameplate (label) attached to the equipment.

[0027] Gyro sensor 428 is capable of measuring the tilt of portable terminal 400 and measures the shooting direction of camera 427. GPS receiver 429 measures the position of portable terminal 400 based on signals received from GPS satellites. GPS receiver 429 also measures the strength of signals received from GPS satellites (hereinafter also referred to as "GPS reception strength").

[0028] 3 is a functional block diagram of the investigation support system 100. The investigation support system 100 includes an imaging unit 111, a character recognition unit 112, an imaging state acquisition unit 113, an imaging equipment estimation unit 114, a model name candidate extraction unit 115, a character recognition result correction unit 116, a recognition result display unit 117, a recognition result manual correction unit 118, and a read result registration unit 119. Specific processing details of these units will be described later using the flowcharts of FIGS.

[0029] The research support system 100 stores various data in the storage device 323, including read result information 121 (see FIG. 9), shooting state correspondence information 122 (see FIG. 6), and model name information 123. The model name information 123 includes model name information 81 (see FIG. 4) and serial number information 82 (see FIG. 5). Note that the model name information 123 may be a single piece of data describing the notation rules for the serial number corresponding to the model name.

[0030] 4 is a diagram showing an example of the model name information 81. The storage device 323 stores "information related to the model name of the equipment" associated with the type of equipment. The model name information 81 includes information related to the model name of the equipment. The "information related to the model name of the equipment" included in the model name information 81 is a list of character strings of the model names of the equipment recorded for each type of equipment.

[0031] 4, the model name information 81 records the model name (character string) of the equipment, as well as the type of equipment, the capacity of the equipment, and the manufacturer of the equipment. For example, the model name information 81 records information such as the type of equipment, such as indoor unit, outdoor unit, lighting, equipment X (e.g., ventilation equipment), etc.

[0032] The model name information 81 contains a list of indoor unit model name character strings such as "AB-IOE45FG" (manufacturer: Company A), "AB-IOE50FG" (manufacturer: Company A), "AB-IOE56FG" (manufacturer: Company A), "JK-LMN50PQ" (manufacturer: Company B), etc. The model name information 81 contains a list of outdoor unit model name character strings such as "RS-TU224XY" (manufacturer: Company A).

[0033] Some of the characters included in the character string of the model name represent product specifications, etc. For example, the numbers "45," "50," "56," and "224" included in the character string of the model name of the indoor unit and the outdoor unit indicate that the cooling capacity is "4.5 kW," "5.0 kW," "5.6 kW," and "22.4 kW," respectively.

[0034] Other examples of indoor unit model names include PL-ZRP40EA7, PLFY-EP112EMG6, PDFY-P22GMG6, etc. Other examples of outdoor unit model names include PUZ-ZRMP40KA8-BS, PUHY-SGP450SDMG6-BS, PUSY-P80SMH2-BS, etc.

[0035] The model name list may be updated, for example, once a month by checking the latest equipment of each manufacturer in the model name information 81. If an unregistered model name is recorded during an on-site equipment inspection, this may be recorded as a new model name in the model name information 81.

[0036] 5 is a diagram showing an example of the serial number information 82. The storage device 323 stores "information related to the serial number of the equipment" associated with the type of equipment. The serial number information 82 includes information related to the serial number of the equipment. The "information related to the serial number of the equipment" included in the serial number information 82 is a notation rule for the equipment serial number, indicating whether each character in the string of characters in the equipment serial number is of a character type including alphabets and numbers.

[0037] In this example, if the character string is a number, the notation rule is "A," and if the character string is an English letter, the notation rule is "B." The serial number information 82 records the notation rule for the equipment serial number, as well as the type of equipment and the equipment manufacturer.

[0038] For example, the serial number information 82 records information such as indoor unit, outdoor unit, lighting, and equipment X as the type of equipment. The serial number information 82 records "ABBAAAA" (manufacturer: Company A), "AABBAAAA" (manufacturer: Company B), and the like as the notation rule for the serial number of the indoor unit. The serial number information 82 records "ABBBAB" (manufacturer: Company A), and the like as the notation rule for the serial number of the outdoor unit.

[0039] Taking the notation rule "ABBAAAA" for the serial number of Company A's indoor unit as an example, the eight characters are composed of, from the front, numbers, letters, letters, numbers, numbers, numbers, numbers. The serial number N1 "1AW23456" shown in the example of Figure 1 conforms to this notation rule. Also, for example, "1GY00332" conforms to this notation rule.

[0040] In this example, the description is based on the assumption that the notation rules differ for each manufacturer, but if the notation rules differ for each model name, the model name corresponding to the notation rules may also be written in the manufacturing number information 82.

[0041] Also, the model name information 81 and the serial number information 82 may be stored as a single piece of data. For example, in Fig. 4, a column for "Serial number notation rules" may be provided between the "Model name" column and the "Capability" column, and the information may be entered in that column. In Fig. 4, the "Model name" column lists model names such as "AB-IOE45FG," "AB-IOE50FG," "AB-IOE56FG," and "JK-LMN50PQ." When combining model name information 81 and serial number information 82 into one piece of data, the notation rules for the serial number corresponding to the model name can be listed in the "Model name" and "Serial number notation rules" columns, such as "AB-IOE45FG" and "ABBAAAA," "AB-IOE50FG" and "ABBAAAA," "AB-IOE56FG" and "ABBAAAA," and "JK-LMN50PQ" and "AABBAAAA."

[0042] 6 is a diagram for explaining the photographing state correspondence information 122. The storage device 323 stores the photographing state correspondence information 122 indicating the relationship between the "photographing state" of the mobile terminal 400 when an image of a nameplate (label) attached to the target equipment is photographed and the "type of target equipment" estimated from the photographing state.

[0043] Here, the image of the nameplate that is the subject of the photograph is also referred to as the "target image." The facility to which the nameplate that is the subject of the photograph is attached is also referred to as the "target facility." The photographing state of the mobile terminal 400 when the target image is photographed is also referred to as the "target state."

[0044] The shooting state includes the shooting direction of the camera 427 when the target image was captured, whether or not the flash of the camera 427 was used when the target image was captured, and the signal strength (GPS reception strength) received by the GPS receiver 429 when the target image was captured.

[0045] The survey support device 300 estimates the type of target equipment from the image capturing state in accordance with the image capturing state correspondence information 122. This will be specifically described below with reference to FIG.

[0046] When a flash is used in the photograph, the investigation support device 300 estimates that the type of target equipment is the first type (ceiling-mounted equipment) installed above the ceiling, including ventilation equipment and ceiling-mounted indoor units (No. 1).

[0047] As shown in the figure, for example, assume that equipment 43 is installed in the attic space above ceiling 61. Equipment 43 is, for example, ventilation equipment. To photograph a nameplate 93 attached to equipment 43, it is necessary to look into the attic space and photograph the side or top of equipment 43. In this case, since the attic space is dark, the flash of camera 427 is turned on when photographing. Therefore, as described above, if a flash is used when photographing, the type of equipment in question is estimated to be attic-installed equipment, such as ventilation equipment or a ceiling-mounted indoor unit.

[0048] When the GPS reception strength is above a threshold (predetermined strength) during imaging, the survey support device 300 estimates that the type of target equipment is a second type (outdoor equipment) that is installed outdoors, including an outdoor unit (No. 2).

[0049] The GPS reception strength is stronger (exceeds the threshold) when mobile terminal 400 is outdoors than when it is indoors. Therefore, if the GPS reception strength is equal to or greater than the threshold, it is determined that photography is being performed outdoors, that is, the type of target facility is outdoor equipment (including outdoor units).

[0050] When the photographing direction (photographing state) is in a third direction that is diagonally downward from the horizontal (for example, in the range of 60 degrees diagonally downward from the horizontal), the survey support device 300 estimates that the type of the target equipment is a third type that includes an outdoor unit and has a nameplate attached to the bottom of the side of the equipment (No. 3). Because outdoor units generally have a nameplate attached to the bottom of the side, when the photographing state is as described above, there is a possibility that the type of the target equipment is an outdoor unit.

[0051] As shown in the figure, a nameplate 92 is attached to the lower part of the side of the equipment 42. In this example, the equipment 42 is an outdoor unit. In this case, the researcher 5 tilts the mobile terminal 400 diagonally downward from the horizontal to photograph the nameplate 92. In this case, it is estimated that the target equipment is equipment with a nameplate attached to the lower part of the side.

[0052] When the shooting direction (shooting state) is a fourth direction that is diagonally upward from the horizontal (for example, in the range of 60 degrees diagonally upward from the horizontal), the survey support device 300 estimates that the type of target equipment is the fourth type, in which a nameplate is attached to the top of the side of the equipment (No. 4).

[0053] As shown in the figure, a nameplate 91 is attached to the upper part of the side of the equipment 41. In this case, the researcher 5 tilts the mobile terminal 400 diagonally upward from the horizontal to photograph the nameplate 91. In this case, the type of the target equipment is estimated to be equipment with a nameplate attached to the lower part of the side.

[0054] When the shooting direction (shooting state) is a fifth direction including vertically upward (for example, a direction in which the shooting is directed toward the ceiling and does not include the fourth direction), the investigation support device 300 estimates that the equipment is of the fifth type, including the indoor unit, in which a nameplate is attached to the bottom surface (No. 5).

[0055] As shown in the figure, a nameplate 94 is attached to the bottom surface of the equipment 44. In this example, the equipment 44 is an indoor unit. In this case, the investigator 5 points the mobile terminal 400 toward the ceiling and takes a photo of the nameplate 94. In this case, the type of the target equipment is estimated to be equipment with a nameplate attached to the bottom surface (such as an indoor unit).

[0056] In addition, in No. 3 to No. 5 above, it is necessary to investigate in advance where the nameplate is attached for each type of equipment and create the photographing condition correspondence information 122. For this purpose, when investigating (photographing) the nameplate, the photographing direction of the mobile terminal 400 may be stored along with the read results of the model name and serial number, and information on the correspondence between the type of equipment and the position of the nameplate may be accumulated. Furthermore, the photographing direction also depends on the height of the investigator 5. For this reason, height information of the investigator 5 may also be stored, and the photographing direction may be corrected according to the height.

[0057] 7 is a flowchart showing the process executed by the survey support system 100. This process may be started, for example, when the nameplate is photographed by the camera 427 of the mobile terminal 400.

[0058] When this process starts, the survey support system 100 executes a first process (see FIG. 8) in S101, then executes a second process (see FIG. 9) in S102, and then ends the process. These processes are executed by the processor 321 in the survey support device 300, and by the processor 421 in the mobile terminal 400.

[0059] 8 is a flowchart of the first process. When the first process starts, in S201, the photographing unit 111 of the survey support device 300 (communication device 324) acquires a target image (photographed data) photographed by the mobile terminal 400. The target image is an image of a nameplate (label) attached to the target facility.

[0060] 1 , an investigator 5 photographs a nameplate 94 of equipment 44 (indoor unit) using a mobile terminal 400. The mobile terminal 400 and the research support device 300 are connected for communication, and the research support device 300 acquires the image photographed by the mobile terminal 400.

[0061] In the following process, the research support device 300 processes the target image. In S202, the character recognition unit 112 of the research support device 300 uses an OCR function to extract a character string of the model name of the target equipment from the target image as a first target character string (hereinafter also referred to as the "target character string of the model name"), and extracts a character string of the serial number of the target equipment from the target image as a second target character string (hereinafter also referred to as the "target character string of the serial number").

[0062] For example, on the nameplate 94 shown in FIG. 1, the model name T1 is "AB-IOE45FG." Here, "IO" is the English letter "eye-oh." In contrast, the character string extracted by OCR is "AB-10E45FG." "10" is the numeral "one-zero." In other words, "IO (eye-oh)" is erroneously recognized as "10 (one-zero)."

[0063] Furthermore, in the nameplate 94 shown in FIG. 1, the serial number N1 is "1AW23456." Here, "1" is the numeral "1." In contrast, the character string extracted by OCR is "IAVV23456." "I" is the English character "i." In other words, "1" is erroneously recognized as "I" and "W" is erroneously recognized as "VV."

[0064] In S203, the image capture state acquisition unit 113 of the research support device 300 (communication device 324) acquires the image capture state (target state) of the mobile terminal 400 when the target image was captured. In this example, information that the image capture direction is vertically upward (facing the ceiling) is acquired as the image capture state of the mobile terminal 400. This corresponds to the state (No. 5) in FIG. 6 in which the nameplate 94 of the equipment 44 (indoor unit) is being captured.

[0065] In S204, the photographed equipment estimation unit 114 of the investigation support device 300 estimates the type of the target equipment based on the target state and the photographed state correspondence information 122. In the above example, since the photographed state corresponds to No. 5 in Fig. 6, the target equipment is estimated to be equipment with a nameplate on the underside, including an indoor unit, etc. At this time, the estimated type of equipment may be displayed on the screen and the investigator 5 may select it.

[0066] In S205, the model name etc. candidate extraction unit 115 of the research support device 300 extracts from the model name information 81 those items that correspond to the type of target equipment as a candidate list of target strings for model names (also referred to as the "model name candidate list"), and also extracts from the manufacturing number information 82 those items that correspond to the type of target equipment as a candidate list of notation rules for target strings for manufacturing numbers (also referred to as the "manufacturing number notation rule candidate list"), and then ends the first processing.

[0067] In the above example, "AB-IOE45FG," "AB-IOE50FG," "AB-IOE56FG," "JK-LMN50PQ," and the like, which correspond to the indoor unit in the model name information 81, are extracted as a candidate list of model names. "ABBAAAA," and the like, which correspond to the indoor unit in the serial number information 82, are extracted as a candidate list of serial number notation rules.

[0068] 9 is a flowchart of the second process. When the second process starts, the research support device 300 generates a string by correcting the target string of the model name based on the candidate list of model names. In this embodiment, the research support device 300 generates a corrected string by correcting the target string based on the type of target equipment estimated based on the target state and information about the model name of the equipment (candidate list of model names).

[0069] More specifically, in S301, the character recognition result correction unit 116 of the research support device 300 selects the character string from the candidate list of model names that has the highest similarity to the target character string of the model name, and sets this as the corrected character string of the model name. The similarity of the character strings may be calculated using a known technique. For example, the similarity of the character strings may be calculated using the Rebelstein distance or the Jaro-Winkler distance.

[0070] In this example, for the target character string "AB-10E45FG" in the model name, the candidate list for the model name includes "AB-IOE45FG," "AB-IOE50FG," "AB-IOE56FG," "JK-LMN50PQ," etc. Of these, "AB-IOE45FG," which matches a total of seven characters, "AB" and "E45FG," has the highest similarity to the target character string in the model name, and therefore "AB-IOE45FG" is selected as the corrected character string for the model name. As a result, the "10 (one / zero)" incorrectly recognized by OCR is corrected to "IO (eye-oh)."

[0071] Next, in S302, the character recognition result correction unit 116 of the research support device 300 generates a corrected string for the manufacturing number by correcting each character of the target string of the manufacturing number based on the candidate list of manufacturing number notation rules and the notation rules for the manufacturing number extracted from the target string of the model name corrected in S301. Specifically, the target string of the manufacturing number is corrected based on the type of target equipment estimated based on the target state of the research support device 300 and information about the equipment's manufacturing number (notation rules for the equipment's manufacturing number).

[0072] In this example, the model name "AB-IOE45FG" of Company A is selected, and therefore "ABBAAAA" is selected as the corresponding serial number notation rule. In other words, the serial number notation rule for the target character string "IAVV23456" of the serial number is "ABBAAAA."

[0073] Because the number of characters in the notation rule is eight and the target string is nine, it is determined that one character was recognized as two during OCR. In this case, it is determined that "VV" was misrecognized as "W," and "VV" is corrected to "W."

[0074] Of the target strings, "AW" is an English letter (B) and "23456" is a number (A), so they follow the notation rules. On the other hand, "I" (アイ) is not a number (A), so it is determined that there is a character misrecognition and is corrected to the number "1 (ichi)." As a result, the corrected serial number string "1AW23456" is generated from the target string "IAVV23456."

[0075] In S301 and S302, multiple correction character strings for the model name and serial number may be generated instead of one. For example, for the model name, "AB-IOE45FG," which has the highest similarity, may be presented as a first correction character string candidate, and "AB-IOE50FG" or "AB-IOE56FG," which have the next highest similarity, may be presented as a second correction character string candidate, and the investigator 5 may select one. In this way, priority may be assigned to the most similar, and correction character string candidates for the model name may be presented in order of priority.

[0076] Furthermore, in the corrected character string of the serial number, multiple candidates may be extracted based on patterns that are easily misrecognized, and the investigator 5 may select one. For example, since "N" and "M" are easily misrecognized, if a character string contains "M," a character string in which "M" is changed to "N" may be presented as the second candidate. The same applies to the above-mentioned "W" and "VV," which are easily misrecognized.

[0077] The research support device 300 transmits the corrected character string for the model name and the corrected character string for the serial number to the mobile terminal 400. As a result, the recognition result display unit 117 of the mobile terminal 400 displays the corrected character string for the model name and the corrected character string for the serial number in S303.

[0078] In this example, the display unit 432 of the mobile terminal 400 displays the model name "AB-IOE45FG" and serial number "1AW23456" as the "read result." These are the corrected and accurate model name and serial number. Below that, it also displays that the manufacturer is "Company A." This may be the manufacturer listed in the model name information 81, or the manufacturer listed on the nameplate 94 may be extracted as a character string using an OCR function and displayed.

[0079] In S304, the recognition result manual correction unit 118 of the portable terminal 400 accepts corrections to the corrected character string for the model name and the corrected character string for the serial number from the investigator 5. In the above example, the model name "AB-IOE45FG" and the serial number "1AW23456" displayed on the display unit 432 can be corrected using the input unit 431.

[0080] When the user modifies one or both of the model name and serial number using input unit 431 and taps the "Send" button, the modified character string after the modification is sent from mobile terminal 400 to research support device 300. When the user taps the "Send" button without modifying either, the modified character string is sent from mobile terminal 400 to research support device 300 as is.

[0081] In S305, the read result registration unit 119 of the portable terminal 400 stores the corrected character strings of the model name and serial number transmitted from the portable terminal 400 in the read result information 121, and ends the second process.

[0082] When the corrected character strings for the model name and serial number are saved, the equipment type and manufacturer information are also stored in the read result information 121. At this time, the location where the image was taken by the mobile terminal 400 may also be stored. For example, the read result information 121 records the model name "AB-IOE45FG" and serial number "1AW23456" along with the location (photography location) "2F Room B1," the equipment type "indoor unit," and the manufacturer "Company A." The photo location may be manually entered by the investigator 5, or the device may be configured to automatically determine the photo location.

[0083] As described above, the investigation support system 100 is a system that supports the investigation of equipment (equipment 41-44) installed in building 1. The model name of the equipment is written on a label (nameplate) attached to the equipment. The investigation support system 100 includes a mobile terminal 400 and an investigation support device 300. The investigation support device 300 includes a storage device 323, a communication device 324 as an acquisition device, and a processor 321 as a processing device. The mobile terminal 400 is provided with a camera 427. The storage device 323 stores information related to the model name of the equipment associated with the type of equipment. The model name information 81 includes information related to the model name of the equipment. The communication device 324 acquires a target image, which is an image of a label attached to the target equipment (equipment 44 in this example) photographed by the mobile terminal 400, and a target state, which is the photographed state of the mobile terminal 400 when the target image was photographed. The processor 321 processes the target image. The processor 321 extracts a character string representing the model name of the target equipment from the target image as a target character string. Processor 321 generates a corrected character string by correcting the target character string based on the type of target equipment estimated based on the target state and information related to the model name of the equipment. Processor 321 transmits the corrected character string to mobile terminal 400. Mobile terminal 400 displays the corrected character string.

[0084] In this way, the type of target equipment is estimated based on the photographed image (photographed image) of the label (nameplate) attached to the equipment, and the character string of the model name extracted from the photographed target image (OCR read result) is corrected based on the photographed image and information about the model name of the equipment and presented to the investigator. In this way, the character string extracted from the photographed image of the label attached to the target equipment can be accurately corrected (reading accuracy improved). Furthermore, because only the automatically acquired photographed state is used when correcting the OCR read result, the reading accuracy of the character string can be improved without imposing any additional work burden on the investigator. If the character string is misrecognized, the investigator ultimately needs to manually correct it, but improving the reading accuracy can reduce the effort required for the investigator to manually correct it.

[0085] The storage device 323 further stores imaging state correspondence information 122 indicating the relationship between the imaging state and the type of equipment. The "information relating to the equipment model name" included in the model name information 81 is a list of character strings for the equipment model names recorded for each type of equipment. The processor 321 estimates the type of target equipment based on the target state and the imaging state correspondence information 122. The processor 321 extracts, from the list of equipment model name character strings, those that correspond to the type of target equipment as a candidate list of target character strings. The processor 321 generates a corrected character string by correcting the target character string based on the candidate list.

[0086] The type of target equipment is estimated based on the photographed state of the target image (photographed image) of the label attached to the equipment, a list of character strings for the equipment's model name is narrowed down to a candidate list based on the type of target equipment, and the character strings for the model name extracted from the photographed target image (OCR read result) are corrected based on the candidate list and presented to the investigator. In this way, the character strings extracted from the photographed image of the label attached to the target equipment can be corrected with high accuracy (improved reading accuracy).

[0087] The processor 321 selects the character string from the candidate list that has the highest similarity to the target character string, and sets it as the corrected character string. By replacing the character string with the actual product model name, the accuracy of reading the character string can be further improved.

[0088] The mobile terminal 400 is further provided with a GPS receiver 429. The GPS receiver 429 measures the position of the mobile terminal 400. The shooting conditions include whether a flash was used by the camera 427 when the target image was captured and the signal strength of the signal received by the GPS receiver 429 when the target image was captured. If a flash was used, the processor 321 estimates that the type of target equipment is a first type installed in the attic, including ventilation equipment. If the signal strength is equal to or greater than a predetermined strength, the processor 321 estimates that the type of target equipment is a second type installed outdoors, including an outdoor unit. Because the type of equipment can be narrowed down to equipment installed in the attic based on whether a flash was used and to equipment installed outdoors based on the GPS reception strength, the accuracy of reading character strings can be further improved.

[0089] The mobile terminal 400 is further provided with a gyro sensor 428 as a sensor. The gyro sensor 428 measures the shooting direction of the camera 427. The shooting state further includes the shooting direction of the camera 427 when the target image was captured. When the shooting direction is a third direction obliquely downward from the horizontal, the processor 321 estimates that the type of the target equipment is a third type, including an outdoor unit, in which a label is attached to the bottom of the side of the equipment. When the shooting direction is a fourth direction obliquely upward from the horizontal, the processor 321 estimates that the type of the target equipment is a fourth type, including an outdoor unit, in which a label is attached to the top of the side of the equipment. When the shooting direction is a fifth direction including a vertically upward direction, the processor 321 estimates that the type of the target equipment is a fifth type, including an indoor unit, in which a label is attached to the bottom of the equipment. Since the type of equipment can be narrowed down based on the position of the label attached to the equipment, the accuracy of reading character strings can be further improved.

[0090] The label attached to the equipment further describes the equipment's serial number. The storage device 323 further stores information related to the equipment's serial number associated with the type of equipment. The serial number information 82 includes information related to the equipment's serial number. The processor 321 further extracts a character string of the serial number of the target equipment from the target image as a second target character string. The processor 321 generates a second corrected character string by correcting the second target character string based on the type of target equipment estimated based on the target state and information related to the equipment's serial number. The processor 321 transmits the second corrected character string to the mobile terminal 400.

[0091] In this way, the type of target equipment is estimated based on the photographed state of the target image (photographed image) of the label attached to the equipment, and the character string of the serial number extracted from the photographed target image (OCR read result) is corrected based on the photographed state and information related to the equipment's serial number, and is presented to the investigator. In this way, the character string extracted from the photographed image of the label attached to the target equipment can be corrected with high accuracy (improved reading accuracy).

[0092] The "information related to the equipment serial number" included in the serial number information 82 is a notation rule for the equipment serial number, which indicates whether each character in the string of characters in the equipment serial number is a character type including alphabetic characters and numeric characters. The processor 321 corrects each character in the second target string based on the notation rule. This improves the reading accuracy for easily confused alphanumeric characters, such as I and 1, or O and 0.

[0093] 10 is a functional block diagram of a survey support system 101 according to a second embodiment. Below, differences from the survey support system 100 according to the first embodiment will be described, and descriptions of parts that share functions will be omitted.

[0094] The survey support system 101 differs from the survey support system 100 in that it further includes a same property information extraction unit 221. This processing will be explained using a flowchart.

[0095] 11 is a flowchart of the processing executed by the survey support system 101. This flowchart differs from the flowchart of the processing executed by the survey support system 100 in that it includes S402. In this flowchart, the survey support device 300 executes the first processing described above in S401, then executes the processing of S402, and then executes the second processing described above in S403.

[0096] The storage device 323 stores, in the read result information 121, information on the model names of equipment installed in areas in building 1 for which model name investigation has been completed. Herein, "area" may refer to a room or section in building 1, or may refer to the entire space of a certain floor. Hereinafter, an area in building 1 for which model name investigation has been completed will also be referred to as a "first area," equipment installed in an area in building 1 for which model name investigation has been completed will also be referred to as "first equipment," and an area in building 1 different from the first area (an area for which investigation has not been completed) will also be referred to as a "second area."

[0097] In S402, if the target equipment is installed in an area where the investigation has not been completed (second area), the identical property information extraction unit 221 of the investigation support device 300 corrects the model name candidate list based on the model name information of the first equipment installed in the first area where the investigation has been completed. At this time, the correction may be made using manufacturer information.

[0098] In this example, the list of candidate model names is "AB-IOE45FG" (Company A), "AB-IOE50FG" (Company A), "AB-IOE56FG" (Company A), "JK-LMN50PQ" (Company B), etc. In contrast, the read result information 121 records information about the first equipment in the first area (e.g., room B1 on the second floor) for which model name investigation in building 1 has been completed.

[0099] The read result information 121 records that the manufacturer of all the equipment that has been investigated is "Company A." For example, in this embodiment, the investigation support device 300 excludes from the candidate list any model names other than the manufacturer of the first equipment for which investigation has been completed (i.e., narrows the candidate list to the manufacturer of the equipment for which investigation has been completed (Company A)). As a result, Company B's "JK-LMN50PQ" and the like are excluded from the candidate list of model names, and the candidate list of corrected model names is narrowed down to Company A's "AB-IOE45FG," "AB-IOE50FG," "AB-IOE56FG," and the like.

[0100] Within the same building, the equipment installed in the building tends to be standardized to products from the same manufacturer. As described above, if a product from manufacturer A is installed in a first area that has already been surveyed, there is a high possibility that a product from manufacturer A will also be installed in a second area that has not yet been surveyed. For this reason, in this embodiment, the model names of the manufacturers that have been installed in surveyed areas are preferentially narrowed down.

[0101] Furthermore, instead of narrowing down the candidates to the model names of manufacturers that have already been investigated as described above, candidates may be prioritized and presented to the researcher 5. For example, the model name of "Company A" may be presented as the first candidate, and model names other than "Company A" may be presented as second and subsequent candidates, and the researcher 5 may select one.

[0102] Furthermore, if it is the same building, there is a possibility that the equipment installed in this building will be standardized to the same manufacturer and the same model (the model name, excluding the capacity, is the same). Also, there is a possibility that the air conditioning capacity, ventilation air volume, lighting model name, and number of units will be similar in each area. For example, if the model name of the air conditioning that has been investigated is "AB-IOE45FG" (capacity is 4.5 kW), the search may be narrowed down to the same model "AB-IOEXXXXXX", or to the model name "XXXXXX45XX" that indicates the same capacity, and these may be presented to the researcher 5 as high-priority candidates.

[0103] For example, if the manufacturer of the equipment in the other two or three areas is all Company A, the list of candidate model names for the target equipment may be limited to Company A. However, since manufacturers may differ for air conditioning (outdoor unit / indoor unit), ventilation equipment, and lighting equipment (type of equipment), whether or not to unify the models may be determined for each type of equipment.

[0104] Alternatively, if there is an area currently being investigated and a similar area that has already been investigated, the search may be narrowed down by the model name of the equipment installed in the similar area that has already been investigated.Whether two areas (rooms) are similar may be determined by calculating the similarity between the two areas using line segment extraction using a differential filter or the like and template matching.Alternatively, if there are two or more areas that have already been investigated and the model names of the equipment in these areas are the same, the model names of the equipment in the area currently being investigated may be narrowed down by the model name of this equipment.

[0105] As described above, the storage device 323 further stores information on the model name of the first equipment installed in the first area for which the model name investigation in the building 1 has been completed. When the target equipment is installed in a second area of ​​the building 1 that is different from the first area, the processor 321 modifies the candidate list based on the information on the model name of the first equipment stored in the storage device 323. For example, the processor 321 may exclude model names other than those of the manufacturer of the first equipment from the candidate list.

[0106] The type of target equipment is estimated based on the photographed image (photographed image) of the label attached to the equipment, and the list of equipment model name strings is narrowed down to a candidate list based on the type of target equipment. Based on the candidate list, the model name string (OCR read result) extracted from the photographed target image is corrected and presented to the investigator. Because the candidate list is further narrowed down based on the model name information of equipment installed in the area where the model name survey has been completed, the string extracted from the photographed image of the label attached to the target equipment can be corrected with even greater accuracy (further improving reading accuracy). Furthermore, because only the automatically acquired photographed state and the survey results of the area where the survey has been completed are used when correcting the OCR read result, the string reading accuracy can be improved without imposing a workload on the investigator. If the string is misrecognized, the investigator must ultimately correct it manually, but improving reading accuracy reduces the investigator's manual correction effort.

[0107] 12 is a functional block diagram of a survey support system 102 according to a third embodiment. Below, differences from the survey support system 100 according to the first embodiment will be described, and descriptions of parts that share functions will be omitted.

[0108] The investigation support system 102 differs from the investigation support system 100 in that it further includes an indoor unit model name estimation unit 231 and an indoor unit number input unit 232. This processing will be explained using a flowchart.

[0109] 13 is a flowchart of the processing executed by the research support system 102. This flowchart differs from the flowchart of the processing executed by the research support system 100 in that it includes steps S602 to S604. In this flowchart, the research support device 300 executes the first processing described above in step S601, then executes steps S602 to S604, and then executes the second processing described above in step S605.

[0110] The storage device 323 stores, in the read result information 121, information on the model name of an outdoor unit (hereinafter also referred to as a "first outdoor unit") that is installed in building 1 and for which a model name investigation has been completed. In the third embodiment, the investigation support device 300 modifies the candidate list of model names of indoor units (hereinafter also referred to as a "first indoor unit") that are installed in building 1 in correspondence with the first outdoor unit, based on the information on the model name of the first outdoor unit. Here, the investigation support device 300 is configured to allow the investigator 5 to input the number of first indoor units installed from the input unit 431.

[0111] After the first process (S601) is completed, the indoor unit model name estimation unit 231 of the investigation support device 300 proceeds to S603 if the type of the target equipment is an indoor unit (YES in S602), and proceeds to S605 (second process) if the type of the target equipment is not an indoor unit (NO in S602).

[0112] In S603, the indoor unit number input unit 232 of the survey support device 300 acquires the number of first indoor units installed corresponding to the first outdoor unit for which the survey has been completed (the number of first indoor units entered by the surveyor 5 from the input unit 431).

[0113] In S604, the indoor unit model name estimation unit 231 of the research support device 300 modifies (narrows) the model name candidate list based on the model name information of the first outdoor unit and the number of first indoor units. Specifically, the research support device 300 first determines the outdoor unit's capacity from the model name information 81 based on the model name information of the first outdoor unit. Next, the research support device 300 estimates the capacity of the first indoor unit based on the capacity of the first outdoor unit and the number of first indoor units. Furthermore, the research support device 300 narrows down the model name candidate list extracted in the first process (S601) to model names of indoor units that are made by the same manufacturer as the outdoor unit and have the estimated capacity based on the model name information of the first outdoor unit, and modifies the candidate list for the first indoor unit.

[0114] For example, suppose a list of candidate model names for the first indoor unit includes "AB-IOE45FG" (Company A), "AB-IOE50FG" (Company A), "AB-IOE56FG" (Company A), "JK-LMN50PQ" (Company B), etc. Read result information 121a records information about the first outdoor unit for which model name investigation in building 1 has been completed. The model name of the first outdoor unit is "RS-TU224XY," the serial number is "1ABC2D," and the manufacturer is "Company A." Investigator 5 also inputs from input unit 431 that the number of first indoor units installed corresponding to this first outdoor unit is four.

[0115] Of the model name "RS-TU224XY" of the first outdoor unit, "224" is a numerical value (=22.4 kW) that indicates the capacity of the first outdoor unit (hereinafter, a numerical value that indicates capacity will be referred to as the "capacity display value"). The research support device 300 extracts the information on the model name "RS-TU224XY" of the first outdoor unit from the model name information 81, and determines that the capacity display value of the outdoor unit is 224.

[0116] Next, the investigation support device 300 estimates the capacity of the first indoor unit based on the capacity display value of the first outdoor unit, "224," and the number of first indoor units, "four." Specifically, the capacity display value of the first indoor unit is calculated using the formula: Capacity display value of the first indoor unit × Number of first indoor units = Capacity display value of the first outdoor unit × Range coefficient α of indoor unit capacities that can be connected to the outdoor unit (hereinafter referred to as "coefficient α"). In this example, if the outdoor unit capacity display value is 224, the number of connected indoor units is four, and coefficient α is 0.5 to 1.3, the capacity display value of the indoor unit is 224 ÷ 4 × (0.5 to 1.3), which is 28 to 72.

[0117] Furthermore, the research support device 300 modifies the candidate list for the first indoor unit based on the model name information of the first outdoor unit (model name "RS-TU224XY," manufacturer "Company A") and the capacity display value (28 to 72) of the first indoor unit. For example, the candidate list for the model name of the first indoor unit, such as "AB-IOE45FG" (Company A), "AB-IOE50FG" (Company A), "AB-IOE56FG" (Company A), and "JK-LMN50PQ" (Company B), is narrowed down to those with indoor unit capacity display values ​​between 28 and 72. In this case, the candidate list for the model name is narrowed down to "AB-IOE45FG" (Company A), "AB-IOE50FG" (Company A), "AB-IOE56FG" (Company A), and "JK-LMN50PQ" (Company B). In this case, information on the model name of the first outdoor unit may be used. For example, a match with the manufacturer (Company A) may be extracted. Furthermore, the search may be narrowed down to the model name of an indoor unit that corresponds to the model name of the outdoor unit (matching series or release year).

[0118] Furthermore, if the area (room) is known, the possible range of indoor unit capacity can be estimated based on the correspondence between indoor unit capacity and room area. For example, the relationship that the air-conditioning area A1 per 2.8 kW of indoor capacity is 11 to 22 square meters can be utilized. If the room area is 88 square meters and four indoor units are installed, the total indoor unit capacity can be estimated as 88 ÷ (11 to 22) × 2.8 ÷ 4 = 2.8 to 5.6. In this case, the model names can be narrowed down to "AB-IOE45FG" (Company A), "AB-IOE50FG" (Company A), and "AB-IOE56FG" (Company A). The relationship between indoor unit capacity and room area, as described above, can be calculated based on the room area, number of indoor units, and capacity extracted from accumulated past survey data.

[0119] As described above, storage device 323 further stores information on the model name of the first outdoor unit that is installed in building 1 and for which model name investigation has been completed. Processor 321 modifies the candidate list of first indoor units that are installed in building 1 in correspondence with the first outdoor unit, based on the information on the model name of the first outdoor unit stored in storage device 323.

[0120] The type of target equipment is estimated based on the photographed image (photographed image) of the label attached to the equipment, and the list of equipment model name strings is narrowed down to a candidate list based on the type of target equipment. Based on the candidate list, the model name string (OCR read result) extracted from the photographed target image is corrected and presented to the investigator. Because the candidate list is further narrowed down based on the model name information of the outdoor unit for which model name investigation has been completed, the string extracted from the photographed image of the label attached to the target equipment can be corrected with even greater accuracy (further improving reading accuracy). Furthermore, because only the automatically acquired photographed state and the investigation results of the outdoor unit for which investigation has been completed are used when correcting the OCR read result, the string reading accuracy can be improved without imposing a workload on the investigator. If the string is misrecognized, the investigator must ultimately correct it manually, but improving reading accuracy reduces the investigator's manual correction effort.

[0121] The investigation support device 300 further includes an input unit 431. The input unit 431 inputs the number of first indoor units. The processor 321 determines the capacity of the first outdoor unit based on the model name information of the first outdoor unit. The processor 321 estimates the capacity of the first indoor unit based on the capacity of the first outdoor unit and the number of first indoor units. The processor 321 corrects the candidate list of first indoor units based on the model name information of the first outdoor unit and the capacity of the first indoor unit. This allows the outdoor unit capacity to be determined based on the model name information of the outdoor unit whose model name has been investigated, and the indoor unit capacity to be estimated based on the outdoor unit capacity and the number of indoor units, thereby further improving reading accuracy.

[0122] 14 is a functional block diagram of a survey support system 103 according to a fourth embodiment. Below, differences from the survey support system 100 according to the first embodiment will be described, and descriptions of parts that share functions will be omitted.

[0123] The survey support system 103 differs from the survey support system 100 in that it further includes a room size estimation unit 241, an indoor capacity estimation unit 242, and an indoor unit number input unit 243. This process will be described using a flowchart.

[0124] 15 is a flowchart of the processing executed by the research support system 103. This flowchart differs from the flowchart of the processing executed by the research support system 100 in that it includes steps S702 to S705. In this flowchart, the research support device 300 executes the first processing described above in step S701, then executes steps S702 to S705, and then executes the second processing described above in step S706.

[0125] The storage device 323 stores, in the read result information 121, information on the model names of lighting (hereinafter also referred to as "third lighting") installed in an area in the building 1 where a survey of lighting model names has been completed (hereinafter also referred to as "third area"). In the third embodiment, the survey support device 300 modifies the candidate list for the model names of indoor units (also referred to as "third indoor units") installed in the third area based on the information on the model names of the third lighting. Here, the survey support device 300 is configured to allow the surveyor 5 to input the number of installed third indoor units from the input unit 431.

[0126] After the first process (S701) is completed, if the type of target equipment in the third area for which the investigation of the number of third lights and the model names of the third lights has been completed is an indoor unit (third indoor unit) (YES in S702), the investigation support device 300 proceeds to S703; if this is not the case (NO in S702), the investigation support device 300 proceeds to S706. For example, suppose the number of third lights in the third area is "N1" and the model name is "XXXXXXXXX." The model name includes a number that indicates the capacity of the third light.

[0127] In S703, the indoor unit number input unit 243 of the survey support device 300 acquires the number of third indoor units installed in the third area (the number of third indoor units input by surveyor 5 from the input unit 431). For example, assume that the number of third indoor units input by surveyor 5 is "N2."

[0128] In S704, the room size estimation unit 241 of the survey support device 300 estimates the size of the third area based on the information on the number of third lights and the model names of the third lights. For example, the capacity of the third lights is identified from the model name of the third lights. For example, the size of the third area is calculated using the formula: capacity of third lights × number of third lights (N1) × coefficient γ = size of third area. In this case, it is assumed that the size of the third area is estimated to be "A1" square meters.

[0129] In S705, the indoor capacity estimation unit 242 of the investigation support device 300 modifies (narrows) the candidate list for the third indoor unit based on the size of the third area and the number of third indoor units. Specifically, the investigation support device 300 estimates the capacity of the third indoor unit based on the size of the third area and the number of third indoor units. The investigation support device 300 then modifies the candidate list for the third indoor unit based on the capacity of the third indoor unit.

[0130] For example, suppose that a list of candidate model names for the third indoor unit includes "AB-IOE45FG" (Company A), "AB-IOE50FG" (Company A), "AB-IOE56FG" (Company A), "JK-LMN50PQ" (Company B), etc. The read result information 121 records information about the third lighting fixture for which the investigation has been completed.

[0131] For example, the capacity of the third indoor unit is calculated using the formula: Capacity of the third indoor unit × Number of third indoor units (N2) × Coefficient β (2 to 4) = Size of the third area (A1). In this example, the capacity of the third indoor unit = A1 ÷ (N2 × (2 to 4)) = 2.5 kW to 5.0 kW is calculated.

[0132] Furthermore, the research support device 300 modifies the candidate list for the third indoor unit based on the capacity of the third indoor unit (25 to 50). For example, from the candidate list of model names for the third indoor unit, such as "AB-IOE45FG" (Company A), "AB-IOE50FG" (Company A), "AB-IOE56FG" (Company A), and "JK-LMN50PQ" (Company B), the research support device 300 narrows the candidate list of model names to "AB-IOE45FG" (Company A), "AB-IOE50FG" (Company A), and "JK-LMN50PQ" (Company B), which correspond to the capacity of the third indoor unit being 25 to 50.

[0133] In the above, the model names of the indoor units are narrowed down based on the number of lights, the model names of the lights, and the number of indoor units. However, this is not limiting, and the model names of the lights may be narrowed down based on the model names of the indoor units, the number of indoor units, and the number of lights.

[0134] Furthermore, the above-mentioned method of estimating the size of a room from the number and model name of lighting fixtures may be derived by extracting property data with the same model name from accumulated past survey data and learning based on the relationship between the number of lighting fixtures and the size of the room. Furthermore, if the area of ​​the area (room) is known, the possible range of the indoor unit's capacity may be estimated from the correspondence relationship between the indoor unit's capacity and the area.

[0135] As described above, the memory device 323 further stores information on the model name of a third lighting device installed in a third area for which investigation of the model names of lighting devices in the building 1 has been completed. The processor 321 modifies the candidate list for a third indoor unit installed in the third area based on the information on the model name of the third lighting device stored in the memory device 323.

[0136] The type of target equipment is estimated based on the photographed image (photographed image) of the label attached to the equipment, and the list of equipment model name strings is narrowed down to a candidate list based on the type of target equipment. Based on the candidate list, the model name string (OCR read result) extracted from the photographed target image is corrected and presented to the investigator. The candidate list is further narrowed down based on the model name information of the lighting installed in the area where the model name investigation has been completed, so the string extracted from the photographed image of the label attached to the target equipment can be corrected with even greater accuracy (further improving reading accuracy). Furthermore, because only the automatically acquired photographed state and the investigation results of the lighting in the area where the investigation has been completed are used when correcting the OCR read result, the string reading accuracy can be improved without imposing a workload on the investigator. If the string is misrecognized, the investigator must ultimately correct it manually, but improving reading accuracy reduces the investigator's manual correction effort.

[0137] The input unit 431 inputs the number of third indoor units. The processor 321 estimates the size of the third area based on the number of third lights. The processor 321 estimates the capacity of the third indoor unit based on the size of the third area and the number of third indoor units. The candidate list for the third indoor unit is revised based on the capacity of the third indoor unit. This allows the area size to be estimated based on the number of lights in the area for which model name investigation has been completed, and the capacity of the indoor unit to be estimated based on the area size and the number of indoor units, further improving reading accuracy.

[0138] 16 is a functional block diagram of a survey support system 104 according to a fifth embodiment. Below, differences from the survey support system 100 according to the first embodiment will be described, and descriptions of parts that share functions will be omitted.

[0139] The research support system 104 differs from the research support system 100 in that it further includes a model name etc. candidate updating unit 251 and a model name etc. position acquiring unit 252. This processing will be explained using a flowchart.

[0140] The position on the nameplate where the model name and serial number are written varies depending on the type of equipment, manufacturer, etc. Figure 17 is a diagram showing an example of the nameplate 92.

[0141] The nameplate 92 shown in Fig. 17 is a nameplate attached to an indoor unit of Company A, similar to the nameplate 94 shown in Fig. 1. On the nameplate 94, "AB-IOE45FG" is written as the model name T1, and "1AW23456" is written as the serial number N1. On the other hand, on the nameplate 92, "AB-IOE50FG" is written as the model name T2, and "1AW45678" is written as the serial number N2. The nameplate 94 and nameplate 92 are indoor units of Company A, and the model name and serial number are written in the same position on the nameplate.

[0142] On the nameplate 92, the model name T2 is written at the position (X1, Y1), and the serial number N2 is written at the position (X2, Y2). On the nameplate 94, the model name T1 is written at the position (X1, Y1), and the serial number N1 is written at the position (X2, Y2).

[0143] Here, the position (X, Y) is a numerical value indicating the horizontal and vertical position on the nameplate. For example, X may indicate a relative position where the left edge of the nameplate is 0 and the right edge of the nameplate is 1, and Y may indicate a relative position where the top edge of the nameplate is 0 and the bottom edge of the nameplate is 1.

[0144] Fig. 18 is a diagram showing an example of a nameplate 93. The nameplate 93 shown in Fig. 18 is a nameplate attached to an indoor unit of Company B. On the nameplate 93, "JK-LMN50PQ" is written as the model name T3, and "98AB7654" is written as the serial number N3. On the nameplate 93, the model name T3 is written at position (X3, Y3), and the serial number N3 is written at position (X4, Y4).

[0145] On the nameplate 93 attached to Company B's indoor unit, the model name T3 is written toward the left (X3), and the serial number N3 is written toward the right (X4). On the other hand, on the nameplate 92 attached to Company A's indoor unit, the model name T2 is written near the center (X1), and the serial number N2 is also written near the center (X2). Furthermore, the model name and serial number are written in different positions along the Y-axis on nameplate 92 and nameplate 93. In this way, the positions where the model name and serial number are written on the nameplate vary depending on the type of equipment or manufacturer, etc.

[0146] 19 is a flowchart of the processing executed by the research support system 104. This flowchart differs from the flowchart of the processing executed by the research support system 100 in that it includes steps S802 and S803. In this flowchart, the research support device 300 executes the first processing described above in step S801, then executes the processing of steps S802 and S803, and then executes the second processing described above in step S804.

[0147] In the fifth embodiment, the information about the model name of the equipment includes information about the position of the model name on a nameplate (label) attached to the equipment. The information about the serial number of the equipment includes information about the position of the serial number on a nameplate (label) attached to the equipment. In other words, the model name information 123 records the position of the model name along with the model name, and records the position of the serial number along with the serial number. Alternatively, the read result information 121 may record the position of the model name along with the model name, and record the position of the serial number along with the serial number.

[0148] After the first process (S801) is completed, the model name etc. position acquisition unit 252 of the research support device 300 acquires information on the position of the model name on the nameplate (label) attached to the equipment and information on the position of the serial number on the nameplate attached to the equipment in S802. In S803, the model name etc. candidate update unit 251 of the research support device 300 modifies (narrows down) the list of model name candidates based on the information on the position of the model name on the nameplate attached to the equipment and the information on the position of the serial number on the nameplate attached to the equipment.

[0149] For example, suppose that a list of candidate equipment model names includes "AB-IOE45FG," "AB-IOE50FG," "AB-IOE56FG," "JK-LMN50PQ," etc. The model name of the target equipment is written in the position (X1, Y1), and the serial number of the target equipment is written in the position (X2, Y2). The model name character string read by OCR is "AB-10E50FG," and the serial number character string is "IAVV45678."

[0150] The read result information 121 (or model name information 123) records information on the position of the model name on the nameplate corresponding to each model name and information on the position of the serial number on the nameplate. In this example, it is recorded that the model name positions of "AB-IOE45FG," "AB-IOE50FG," "AB-IOE56FG," etc. are all (X1, Y1) and the serial number positions of all (X2, Y2). It is also recorded that the model name position of "JK-LMN50PQ," etc. is (X3, Y3) and the serial number position is (X4, Y4).

[0151] In S803, because the model name and serial number positions obtained in S802 are (X1, Y1) and (X2, Y2), the research support device 300 narrows down the list of candidate model names to those closest to these positions. Specifically, it excludes products such as "JK-LMN50PQ," which have model name positions (X3, Y3) and serial number positions (X4, Y4), and narrows down the list to products such as "AB-IOE45FG," "AB-IOE50FG," and "AB-IOE56FG," which have model name positions (X1, Y1) and serial number positions (X2, Y2).

[0152] As in the first embodiment, in the second process, the model name character string is corrected from "AB-10E50FG" to "AB-IOE50FG", and the serial number character string is corrected from "IAVV45678" to "1AW45678".

[0153] The position of the model name and serial number on the nameplate varies depending on the type of equipment, manufacturer, model name, etc., so this information can be used to further narrow down the list of candidate model names. The positions of the model name and serial number on the nameplate may be specified manually by the investigator 5 or may be specified automatically. The positions of the model name and serial number on the nameplate may be recorded when recording the survey results for the target equipment. The positions may be recorded at least once for each model name.

[0154] As described above, the information about the model name of the equipment includes information about the position of the model name on the label attached to the equipment. The information about the serial number of the equipment includes information about the position of the serial number on the label attached to the equipment. The processor 321 modifies the candidate list based on the information about the position of the model name on the label attached to the equipment and the information about the position of the serial number on the label attached to the equipment.

[0155] The system estimates the type of target equipment based on the photographed image (photographed image) of the label attached to the equipment, narrows down the list of model name strings for the equipment based on the type of target equipment to create a candidate list, and then corrects the model name string (OCR read result) extracted from the photographed target image based on the candidate list and presents it to the investigator. Because the candidate list is further narrowed down based on information on the position of the model name and serial number, the string extracted from the photographed image of the label attached to the target equipment can be corrected with even greater accuracy (further improving reading accuracy). Furthermore, because only the automatically acquired photographed state and the automatically or manually identified position of the model name and serial number are used to correct the OCR read result, the string reading accuracy can be improved without imposing a workload on the investigator. If a string is misrecognized, the investigator must ultimately correct it manually, but improving reading accuracy reduces the investigator's manual correction effort.

[0156] The embodiments disclosed herein are intended to be combined as appropriate within the scope of any technical inconsistency. The embodiments disclosed herein should be considered to be illustrative and not restrictive in all respects. The technical scope of the present disclosure is defined by the claims, not the description of the above-mentioned embodiments, and is intended to include all modifications within the meaning and scope of the claims.

[0157] 1 Building, 5 Investigator, 41 to 44 Equipment, 61 Ceiling, 81 Model name information, 82 Serial number information, 91 to 94 Nameplate, 100 to 104 Investigation support system, 111 Photography unit, 112 Character recognition unit, 113 Photography status acquisition unit, 114 Photography equipment estimation unit, 115 Model name candidate extraction unit, 116 Character recognition result correction unit, 117 Recognition result display unit, 118 Recognition result manual correction unit, 119 Reading result registration unit, 121 Reading result information, 122 Photography status correspondence information, 123 Model name information, 221 Same property information extraction unit, 231 Indoor unit model name estimation unit, 232 Indoor unit number input unit, 241 Room size estimation unit, 242 Indoor capacity estimation unit, 243 Indoor unit number input unit, 251 Model name candidate update unit, 252 Model name etc. position acquisition unit, 300 survey support device, 321, 421 processor, 322, 422 memory, 323, 423 storage device, 324, 424 communication device, 326, 426 bus, 400 portable terminal, 427 camera, 428 gyro sensor, 429 GPS receiver, 431 input unit, 432 display unit, N1 to N3 serial number, T1 to T3 model name.

Claims

1. An investigation support device that supports the investigation of equipment installed in a building, wherein a label affixed to the equipment indicates the model name of the equipment, and the investigation support device comprises: a storage device that stores information about the model name of the equipment associated with the type of the equipment; an acquisition device that acquires a target image, which is an image of the label affixed to the target equipment, taken by a mobile device equipped with a camera, and a target state, which is the photographed state of the mobile device when the target image was taken; and a processing device that processes the target image, wherein the processing device: extracts a character string that is a character string of the model name of the target equipment from the target image as a target string; generates a corrected character string by correcting the target string based on the type of the target equipment estimated based on the target state and information about the model name of the equipment; and sends the corrected character string to the mobile device.

2. The investigation support device described in claim 1, wherein the storage device further stores imaging state correspondence information indicating the relationship between the imaging state and the type of equipment, the information regarding the equipment model name is a list of strings of the equipment model names recorded for each type of equipment, and the processing device estimates the type of the target equipment based on the target state and the imaging state correspondence information, extracts those from the list of equipment model name strings that correspond to the type of target equipment as a candidate list of target strings, and generates the corrected string by correcting the target string based on the candidate list.

3. The research support device according to claim 2, wherein the processing device selects a character string from the candidate list that has the highest similarity to the target character string, and sets the selected character string as the corrected character string.

4. The survey support device of claim 2 or claim 3, wherein the mobile terminal is further provided with a GPS receiver that measures the position of the mobile terminal, and the shooting status includes whether or not a flash was used on the camera when the target image was taken and the signal strength of the signal received by the GPS receiver when the target image was taken, and the processing device, if the flash was used, estimates that the type of the target equipment is a first type that is installed in the attic and includes ventilation equipment, and if the signal strength is equal to or greater than a predetermined strength, estimates that the type of the target equipment is a second type that is installed outdoors and includes an outdoor unit.

5. The survey support device of claim 4, wherein the mobile terminal is further provided with a sensor that measures the shooting direction of the camera, and the shooting state further includes the shooting direction of the camera when the target image was shot, and the processing device: if the shooting direction is a third direction that is diagonally downward from the horizontal, estimates that the type of the target equipment is a third type in which a label is attached to the bottom of the side of the equipment, including the outdoor unit, if the shooting direction is a fourth direction that is diagonally upward from the horizontal, estimates that the type of the target equipment is a fourth type in which a label is attached to the top of the side of the equipment, and estimates that the type of the target equipment is a fifth type in which a label is attached to the bottom of the equipment, including an indoor unit, if the shooting direction is a fifth direction including a vertically upward direction.

6. A survey support device as claimed in any one of claims 2 to 5, wherein a label attached to the equipment further describes the serial number of the equipment, the storage device further stores information relating to the serial number of the equipment associated with the type of the equipment, and the processing device further extracts a character string of the serial number of the target equipment from the target image as a second target character string, generates a second corrected character string by correcting the second target character string based on the type of the target equipment estimated based on the target state and information relating to the equipment's serial number, and transmits the second corrected character string to the mobile terminal.

7. The research support device described in claim 6, wherein the information regarding the equipment's serial number is a notation rule for the equipment's serial number indicating whether each character in the string of characters in the equipment's serial number is a character type including alphabets and numbers, and the processing device corrects each character in the second target string based on the notation rule.

8. A survey support device as described in any one of claims 2 to 7, wherein the storage device further stores information on the model name of a first piece of equipment installed in a first area of ​​the building for which a survey of model names has been completed, and the processing device modifies the candidate list based on the information on the model name of the first piece of equipment stored in the storage device when the target equipment is installed in a second area of ​​the building that is different from the first area.

9. The research support device according to claim 8, wherein said processing device excludes model names other than those of the manufacturer of said first equipment from said candidate list.

10. A research support device as described in any one of claims 2 to 7, wherein the storage device further stores information on the model name of a first outdoor unit that is installed in the building and for which a model name investigation has been completed, and the processing device modifies the candidate list of a first indoor unit that is installed in the building in correspondence with the first outdoor unit based on the information on the model name of the first outdoor unit stored in the storage device.

11. The research support device of claim 10, further comprising an input unit for inputting the number of the first indoor units, wherein the processing device: estimates the capacity of the first outdoor unit based on information about the model name of the first outdoor unit; estimates the capacity of the first indoor unit based on the capacity of the first outdoor unit and the number of the first indoor units; and modifies the candidate list of the first indoor units based on information about the model name of the first outdoor unit and the capacity of the first indoor units.

12. The investigation support device described in any one of claims 2 to 7, wherein the storage device further stores information on the model name of a third lighting device installed in a third area of ​​the building for which an investigation of the model names of the lighting devices has been completed, and the processing device modifies the candidate list of a third indoor unit installed in the third area based on the information on the model name of the third lighting device stored in the storage device.

13. The survey support device of claim 12, further comprising an input unit for inputting the number of the third indoor units, wherein the processing device: estimates the size of the third area based on the number of the third lights; estimates the capacity of the third indoor units based on the size of the third area and the number of the third indoor units; and modifies the candidate list of the third indoor units based on the capacity of the third indoor units.

14. A survey support device as described in any one of claims 2 to 7, wherein the label attached to the equipment further describes the serial number of the equipment, the storage device further stores information about the serial number of the equipment associated with the type of equipment, the information about the model name of the equipment includes information about the position of the model name on the label attached to the equipment, and the information about the serial number of the equipment includes information about the position of the serial number on the label attached to the equipment, and the processing device modifies the candidate list based on the information about the position of the model name on the label attached to the equipment and the information about the position of the serial number on the label attached to the equipment.

15. An investigation support system that supports the investigation of equipment installed in a building, wherein a label attached to the equipment indicates the model name of the equipment, comprising: a mobile terminal equipped with a camera; a storage device that stores information about the model name of the equipment associated with the type of the equipment; an acquisition device that acquires a target image that is an image of the label attached to the target equipment photographed by the mobile terminal and a target state that is the photographed state of the mobile terminal when the target image was photographed; and a processing device that processes the target image, wherein the processing device extracts a character string that is the model name of the target equipment from the target image as a target string, generates a corrected string by correcting the target string based on the type of the target equipment estimated based on the target state and information about the model name of the equipment, and sends the corrected string to the mobile terminal, which displays the corrected string.

16. A survey support method executed by a computer to support the survey of equipment installed in a building, wherein a label attached to the equipment indicates the model name of the equipment, the survey support method comprising the steps of: storing information about the model name of the equipment associated with the type of the equipment; acquiring a target image, which is an image of the label attached to the target equipment, taken by a mobile device equipped with a camera, and a target state, which is the photographed state of the mobile device when the target image was taken; extracting a character string of the model name of the target equipment from the target image as a target string; generating a corrected string by correcting the target string based on the type of the target equipment estimated based on the target state and information about the model name of the equipment; and transmitting the corrected string to the mobile device.

Images