JP2017028407A - Program, device and method for imaging instruction - Google Patents

Abstract

PROBLEM TO BE SOLVED: To output an imaging instruction to image an object from an imaging direction enabling efficiently identifying the object.SOLUTION: An imaging instruction device S acquires a target image TI obtained by imaging an object T. The imaging instruction device S searches for an image which resembles the acquired target image TI from among images which are respectively obtained by imaging a plurality of objects from a respective plurality of imaging directions, for each of the plurality of objects. When a plurality of searched images exist, the imaging instruction device S calculates similarity among images which are obtained by imaging respective objects corresponding to the searched images from imaging directions which are different from the imaging direction of the searched image by predetermined angles to a predetermined direction. Based on the calculated similarity, the imaging instruction device S outputs an imaging instruction to image the object T from the imaging direction different by the predetermined angle from the imaging direction of the target image TI to the predetermined direction.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a shooting instruction program, a shooting instruction device, and a shooting instruction method.

  2. Description of the Related Art Conventionally, there is a technique for identifying an object corresponding to a subject from among a plurality of objects by comparing an image obtained by photographing the subject and an image obtained by photographing each of the plurality of objects. As related prior art, for example, candidates for an object corresponding to a subject are acquired based on the characteristics of each part of the subject extracted from an image, and when a plurality of candidates are acquired, the candidates are narrowed down. There is one that specifies the region and prompts the user to take another image. In addition, for example, there is a technique for performing pattern matching processing on each image captured by a plurality of cameras and acquiring information about an object present in each image.

JP 2006-101166 A JP 2002-83297 A

  However, in the above-described conventional technology, it is not possible to determine a shooting direction in which an object can be efficiently identified from a shot image. For example, an image effective for narrowing down the object corresponding to the target object cannot be obtained, and the number of times the target object is shot increases, leading to an increase in the work amount of the photographer.

  In one aspect, an object of the present invention is to provide a photographing instruction program, a photographing instruction device, and a photographing instruction method that output a photographing instruction from a photographing direction in which an object can be efficiently identified.

  According to one aspect of the present invention, a target image obtained by photographing a target object is acquired, and for each of a plurality of objects, the acquired target image is selected from among images obtained by photographing the object from each of a plurality of photographing directions. When similar images are searched and there are a plurality of searched images, images obtained by shooting objects corresponding to each of the searched images from different shooting directions by a predetermined angle from the shooting direction of the images. A shooting instruction program, a shooting instruction device, and a shooting instruction method for outputting a shooting instruction of the object from a shooting direction that differs from the shooting direction of the target image by the predetermined angle from the shooting direction of the target image based on the similarity of Is done.

  According to one aspect of the present invention, it is possible to output a shooting instruction from a shooting direction that can efficiently identify an object.

FIG. 1 is an explanatory diagram of an example of the imaging instruction method according to the embodiment. FIG. 2 is an explanatory diagram illustrating an example of the object specifying system 200. FIG. 3 is a block diagram illustrating a hardware configuration example of the imaging instruction apparatus S. FIG. 4 is a block diagram illustrating a hardware configuration example of the imaging apparatus C. FIG. 5 is an explanatory diagram showing an example of the data structure of the image information 500. FIG. 6 is an explanatory diagram showing an example of the data structure of the object information table 600. FIG. 7 is an explanatory diagram showing an example of the data structure of the shooting information table 700. FIG. 8 is an explanatory diagram showing an example of the data structure of the similarity information table 800. FIG. 9 is an explanatory diagram showing an example of the data structure of the comparison information table 900. FIG. 10 is an explanatory diagram showing an example of the data structure of the shooting instruction 1000. FIG. 11 is a block diagram illustrating a functional configuration example of the imaging instruction apparatus S. FIG. 12 is an explanatory diagram (part 1) illustrating the flow of the object specifying process corresponding to the target T. FIG. 13 is an explanatory diagram (part 2) illustrating the flow of the object specifying process corresponding to the target T. FIG. 14 is an explanatory diagram (part 3) illustrating the flow of the object specifying process corresponding to the target T. FIG. 15 is an explanatory diagram (part 1) illustrating an example of the object specifying process corresponding to the target object PC-1. FIG. 16 is an explanatory diagram (part 2) illustrating an example of the object specifying process corresponding to the target object PC-1. FIG. 17 is an explanatory diagram (part 3) illustrating an example of the object specifying process corresponding to the target object PC-1. FIG. 18 is an explanatory diagram (part 4) illustrating an example of the object specifying process corresponding to the target object PC-1. FIG. 19 is an explanatory diagram (part 5) illustrating an example of the object specifying process corresponding to the target object PC-1. FIG. 20 is an explanatory diagram (part 6) illustrating an example of the object specifying process corresponding to the target object PC-1. FIG. 21 is an explanatory diagram illustrating an example of a confirmation process for the object PC-1. FIG. 22 is a sequence diagram illustrating an example of the specific processing procedure. FIG. 23 is a flowchart illustrating an example of an image determination processing procedure. FIG. 24 is a flowchart illustrating an example of an image comparison processing procedure.

  Hereinafter, embodiments of a shooting instruction program, a shooting instruction device, and a shooting instruction method according to the present invention will be described in detail with reference to the drawings.

(One Example of Shooting Instruction Method According to Embodiment)
FIG. 1 is an explanatory diagram of an example of the imaging instruction method according to the embodiment. In FIG. 1, a photographing instruction device S outputs a photographing instruction for a target T, obtains a target image TI obtained by photographing the target T, and identifies an object corresponding to the target T based on the target image TI. It is. The object corresponding to the target T is the same type of object as the target T.

  Here, a case where an object corresponding to the target T is specified from among a plurality of objects with reference to the name of the target T, identification information attached to the target T, or the like can be considered. However, in this case, when the name of the target T is unknown or when identification information is not attached to the target T, the object corresponding to the target T cannot be specified. Further, when the name of the object T is examined or the identification information attached to the object T is searched, the work amount of the operator who specifies the object T is increased.

  On the other hand, by preparing an image obtained by photographing each of a plurality of objects from each of a plurality of photographing directions and searching for an image similar to the target image TI obtained by photographing the target T from the prepared images. A case where an object corresponding to the object T is specified can be considered. For example, an image similar to the target image TI is searched from the prepared images, an object on which a similar image is captured is determined to be a candidate for the target T, and an object that is a candidate for the target T Can be narrowed down to one, the object is identified as an object corresponding to the target T. However, in this case, if there are a plurality of similar images, it is not possible to narrow down the objects that are candidates for the target T to one, and the object corresponding to the target T cannot be specified.

  In addition, when an object that is a candidate for the target T cannot be narrowed down to one, it is similar to another target image TI that has captured the target T out of images obtained by capturing the object that is a candidate for the target T. It is conceivable that objects that are candidates for the target T are further narrowed down by searching for images to be performed. However, even in this case, if another target image TI obtained by photographing the target T is not a target image TI effective for narrowing down the object that is a candidate for the target T, an object that is a candidate for the target T is selected. The object cannot be narrowed down and the object corresponding to the object T cannot be specified. In addition, the photographing of the target T is repeated until a target image TI effective for narrowing down the objects that are candidates for the target T is obtained, resulting in an increase in the work amount of the photographer who photographs the target T. End up.

  Therefore, in the present embodiment, a target image TI effective for narrowing down objects that are candidates for the target T is obtained by outputting a shooting instruction from a shooting direction that can efficiently identify the target T. A method for instructing photographing to efficiently identify the target T will be described.

  In the example of FIG. 1, (1) the photographing instruction device S acquires a target image TI obtained by photographing the target T. The target object T is an object for specifying which of the plurality of objects corresponds to the object. The object T is, for example, a machine in a machine room. The imaging instruction device S receives, from the imaging device C, for example, a target image TI obtained by imaging the object T from one direction by the imaging device C. The photographing apparatus C is a computer having a photographing function. The imaging device C is, for example, a smartphone.

  (2) For each of the plurality of objects, the imaging instruction device S searches for an image similar to the acquired target image TI from images obtained by imaging the object from each of a plurality of imaging directions. The imaging instruction device S stores, for example, images P11 to P14, P21 to P24, P31 to P34, and P41 to P44 obtained by imaging each of the objects O1 to O4 from the front, rear, left, and right directions. And the imaging | photography instruction | indication apparatus S specifies the image P11, P21, P31 which image | photographed the objects O1-O3 from the front as an image similar to the target image TI.

  (3) When there are a plurality of searched images, the shooting instruction device S captures images corresponding to each of the searched images from different shooting directions by a predetermined angle from the shooting direction of the images. The similarity is calculated. Then, based on the calculated similarity, the photographing instruction device S outputs a photographing instruction for the target T from a photographing direction that differs from the photographing direction of the target image TI by a predetermined angle from the predetermined direction.

  The imaging instruction device S calculates, for example, the similarity between images from different imaging directions of 90 degrees, 180 degrees, and 270 degrees to the right from the respective imaging directions of the identified images. Specifically, the similarities of the images P12 and P22 from the shooting directions that are 90 degrees to the right from the respective shooting directions of the images P11 and P21, and the shooting directions that are 180 degrees to the right from the respective shooting directions of the images P11 and P31. The similarity between the images P13 and P33 is calculated.

  Then, the imaging instruction device S selects the angle “90 degrees” at which the smallest similarity is calculated, and the object T from different imaging directions by the angle “90 degrees” selected to the right from the imaging direction of the target image TI. The shooting instruction is output. Further, the imaging instruction device S is predetermined to the right from the imaging direction of the target image TI when the similarity between the images from the imaging directions different from each imaging direction by a predetermined angle in the right direction from the respective imaging directions is equal to or less than a threshold value. You may output the imaging | photography instruction | indication of the target object T from the imaging direction which differs by an angle.

  According to this, in the imaging instruction device S, there is a high possibility that the target image TI to be acquired next is similar to one of the images having relatively low similarity, and may not be similar to the other. The target image TI can be increased. As a result, the imaging instruction device S determines that one of the images with relatively low similarity is an image that is not similar to the target image TI acquired next, and the object from which the image is captured is the target object. It can be determined that the object may be excluded from the T candidate objects.

  In the example of FIG. 1, the imaging instruction device S outputs the imaging instruction of the object T from the imaging direction that is different from the imaging direction of the target image TI by the angle “90 degrees” to the right, and as a result A new target image TI taken can be acquired. That is, the shooting instruction device S is likely to be similar to one of the images P12 and P32 from the shooting directions 90 degrees to the right from the respective shooting directions of the images P11 and P31, and may not be similar to the other. It is possible to acquire a new photographed image TI with high performance. As a result, based on the similarity between the new target image TI and each of the images P12 and P32, the imaging instruction device S selects the object O1 corresponding to the image P12 whose similarity is equal to or less than the threshold as the candidate for the target T. It can be determined that the object may be excluded from the object.

  On the other hand, the user of the imaging device C, for example, moves the object T from the rear if there is no instruction to shoot the object T from the imaging direction that is different from the shooting direction of the target image TI by the angle “90 degrees”. There is a possibility of shooting. However, the target image TI obtained by photographing the target T from the rear is similar to any of the images P13, P23, and P33 obtained by photographing each of the objects O1 to O3 from the rear, and is thus excluded from objects that are candidates for the target T. It may not be effective for determining whether or not. For this reason, even if the user of the imaging device C captures the target T, there is no object that is determined to be an object that may be excluded from the objects that are candidates for the target T, and May not be narrowed down, leading to an increase in the amount of work for the user of the photographing apparatus C.

  In this way, the imaging instruction device S can easily identify an object that has been captured with an image that is not similar to the next acquired target image TI and that may be excluded from objects that are candidates for the target T. Thus, it is possible to efficiently narrow down objects that are candidates for the target T. In other words, the shooting instruction apparatus S obtains a target image TI that is effective for narrowing down the objects that are candidates for the target T, and outputs a shooting instruction from the shooting direction that enables the target T to be identified efficiently. can do. Then, the imaging instruction device S can efficiently narrow down objects that are candidates for the target T based on the next acquired target image TI, and can efficiently specify the object corresponding to the target T. .

  Here, the case has been described in which the shooting instruction apparatus S calculates the similarity between two images shot from different shooting directions by a predetermined angle from the shooting direction of the searched image. However, the present invention is not limited to this. For example, the imaging instruction device S may calculate the average value of the similarity for three or more images captured from a shooting direction that differs from the shooting direction of the searched image by a predetermined angle in a predetermined direction. Then, the photographing instruction device S outputs a photographing instruction for the target T from a photographing direction that differs by a predetermined angle from the photographing direction of the target image TI based on the average value of the similarity.

  In addition, here, a case has been described in which the photographing instruction device S stores images obtained by photographing an object from each of the four directions of front, rear, left, and right, but the present invention is not limited thereto. For example, the photographing instruction device S may store an image obtained by photographing an object from multiple directions such as front and rear, left and right, up and down, and tanning.

  Although the case where the shooting instruction device S outputs a shooting instruction of the object T from one shooting direction has been described here, the present invention is not limited thereto. For example, the photographing instruction device S may output a photographing instruction so as to photograph the object T from at least one of a plurality of photographing directions in which the calculated similarity is equal to or less than a threshold value.

(An example of the object specifying system 200)
Next, an example of the object specifying system 200 to which the imaging instruction device S shown in FIG. 1 is applied will be described.

  FIG. 2 is an explanatory diagram illustrating an example of the object specifying system 200. In FIG. 2, the object specifying system 200 includes a photographing instruction device S and photographing devices C-1 to Cn (n is a natural number of 2 or more). In the following description, the imaging devices C-1 to C-n may be referred to as “imaging device C”.

  The imaging instruction device S is, for example, a server, a PC (Personal Computer), a notebook PC, or the like. The photographing apparatus C is, for example, a notebook PC, a tablet terminal, a smartphone, a PDA (Personal Digital Assistant), or the like.

  In the object identification system 200, the imaging instruction device S and the imaging devices C-1 to C-n are connected via a wired or wireless network 210. The network 210 is, for example, a local area network (LAN), a wide area network (WAN), or the Internet.

<Specific Example of Object Identification System 200 (Part 1)>
For example, the photographing instruction device S may output a photographing instruction to the photographing device C, and the photographing device C may receive a photographing instruction from the photographing instruction device S.

  Specifically, the imaging instruction apparatus S receives image information 500 described later with reference to FIG. The imaging instruction apparatus S includes various tables such as an object information table 600, an imaging information table 700, a similarity information table 800, and a comparison information table 900, which will be described later with reference to FIGS. The imaging instruction apparatus S transmits an imaging instruction 1000 (to be described later in FIG. 10) to the imaging apparatus C based on the received image information 500 and information in various tables.

  Specifically, the imaging device C captures the target T and acquires a target image TI obtained by capturing the target T. The imaging apparatus C transmits image information 500 described later in FIG. 5 including the acquired target image TI to the imaging instruction apparatus S. As a result of transmitting the image information 500, the imaging apparatus C receives an imaging instruction 1000 (to be described later in FIG. 10) from the imaging instruction apparatus S. When the imaging apparatus C receives the imaging instruction 1000, the imaging apparatus C notifies the user of the imaging apparatus C of the imaging direction indicated by the imaging instruction 1000 via the display 406 described later in FIG.

<Specific Example of Object Identification System 200 (Part 2)>
Further, for example, the imaging instruction device S provides information of various tables to the imaging device C, and the imaging device C has a function as the imaging instruction device S, and the imaging instruction device S is based on the information of the various tables provided. May be operated to output the shooting instruction 1000.

  Specifically, the imaging instruction apparatus S includes the various tables described above, and provides information on the various tables to the imaging apparatus C in response to a request from the imaging apparatus C.

  Specifically, the imaging device C captures the target T, acquires the target image TI obtained by capturing the target T, and generates image information 500 described later in FIG. 5 including the target image TI. When the image capturing apparatus C generates the image information 500, the image capturing apparatus C generates an image capturing instruction 1000 which will be described later with reference to FIG. Then, the photographing apparatus C notifies the user of the photographing apparatus C of the photographing direction indicated by the photographing instruction 1000 via the display 406 described later in FIG.

<Specific Example of Object Identification System 200 (No. 3)>
In addition, for example, the photographing apparatus C has a function as the photographing instruction apparatus S, the photographing apparatus C has various tables, and the photographing apparatus C operates as the photographing instruction apparatus S based on the information of the various tables. 1000 may be output.

  Specifically, the photographing apparatus C has the various tables described above. The imaging device C captures the target T, acquires a target image TI obtained by capturing the target T, and generates image information 500 described later in FIG. 5 including the target image TI. When the image capturing apparatus C generates the image information 500, the image capturing apparatus C generates an image capturing instruction 1000 which will be described later with reference to FIG. Then, the photographing apparatus C notifies the user of the photographing apparatus C of the photographing direction indicated by the photographing instruction 1000 via the display 406 described later in FIG. In this case, the imaging instruction device S may not be included in the object specifying system 200.

  In the following description, in the specific example (part 1) of the object identification system 200 described above, the imaging instruction device S outputs the imaging instruction 1000 to the imaging device C, and the imaging device C receives the imaging instruction 1000 from the imaging instruction device S. A case of receiving will be described.

(Example of hardware configuration of the imaging instruction device S)
Next, a hardware configuration example of the imaging instruction apparatus S will be described with reference to FIG.

  FIG. 3 is a block diagram illustrating a hardware configuration example of the imaging instruction apparatus S. In FIG. 3, the imaging instruction apparatus S includes a CPU (Central Processing Unit) 301, a memory 302, an I / F (Interface) 303, a disk drive 304, and a disk 305. Each component is connected by a bus 300.

  Here, the CPU 301 governs overall control of the imaging instruction apparatus S. The memory 302 includes, for example, a ROM (Read Only Memory), a RAM (Random Access Memory), and a flash ROM. Specifically, for example, a flash ROM or ROM stores various programs, and a RAM is used as a work area for the CPU 301. The program stored in the memory 302 is loaded into the CPU 301 to cause the CPU 301 to execute the coded process. The memory 302 stores various tables such as an object information table 600, an imaging information table 700, a similarity information table 800, and a comparison information table 900, which will be described later with reference to FIGS.

  The I / F 303 is connected to the network 210 via a communication line, and is connected to another computer (for example, the photographing apparatus C shown in FIG. 2) via the network 210. The I / F 303 controls an internal interface with the network 210 and controls input / output of data from other computers. For example, a modem or a LAN adapter may be employed as the I / F 303.

  The disk drive 304 controls reading / writing of data with respect to the disk 305 according to the control of the CPU 301. The disk drive 304 is, for example, a magnetic disk drive. The disk 305 is a nonvolatile memory that stores data written under the control of the disk drive 304. The disk 305 may store various tables such as an object information table 600, an imaging information table 700, a similarity information table 800, and a comparison information table 900, which will be described later with reference to FIGS. The disk 305 is, for example, a magnetic disk or an optical disk.

  The imaging instruction apparatus S may include, for example, an SSD (Solid State Drive), a keyboard, a mouse, and a display in addition to the above-described components. Further, the imaging instruction device S may have an SSD or the like instead of the disk drive 304 and the disk 305.

(Hardware configuration example of photographing apparatus C)
Next, a hardware configuration example of the photographing apparatus C will be described with reference to FIG.

  FIG. 4 is a block diagram illustrating a hardware configuration example of the imaging apparatus C. In FIG. 4, the imaging apparatus C includes a CPU 401, a memory 402, an I / F 403, a disk drive 404, a disk 405, a display 406, a camera 407, and a compass 408. Each component is connected by a bus 400.

  Here, the CPU 401 controls the entire photographing apparatus C. The memory 402 includes, for example, a ROM, a RAM, a flash ROM, and the like. Specifically, for example, a flash ROM or ROM stores various programs, and the RAM is used as a work area of the CPU 401. The program stored in the memory 402 is loaded on the CPU 401 to cause the CPU 401 to execute the coded process.

  The I / F 403 is connected to the network 210 via a communication line, and is connected to another computer (for example, the imaging instruction apparatus S shown in FIG. 2) via the network 210. The I / F 403 controls an internal interface with the network 210 and controls input / output of data from other computers.

  The disk drive 404 controls reading / writing of data with respect to the disk 405 according to the control of the CPU 401. The disk 405 stores data written under the control of the disk drive 404. Examples of the disk 405 include a magnetic disk and an optical disk.

  The display 406 displays data such as a document, an image, and function information as well as a cursor, an icon, or a tool box. The display 406 can employ, for example, a CRT (Cathode Ray Tube), a liquid crystal display, an organic EL (Electroluminescence) display, or the like.

  The camera 407 acquires, as an image, information obtained by converting optical information obtained from a predetermined space through a photographing lens into an electrical signal by a photoelectric conversion element. The compass 408 acquires the shooting direction of the shooting apparatus C. For example, the compass 408 acquires the direction of the photographing lens of the camera 407 included in the photographing apparatus C as the photographing direction.

  The photographing apparatus C may not include the disk drive 404, the disk 405, the compass 408, and the like among the above-described components. In addition to the components described above, the photographing apparatus C may include, for example, an SSD, a scanner, a printer, and the like.

(Data structure of image information 500)
Next, an example of the data structure of the image information 500 that is generated by the imaging device C and transmitted to the imaging instruction device S will be described with reference to FIG.

  FIG. 5 is an explanatory diagram showing an example of the data structure of the image information 500. As shown in FIG. 5, the image information 500 has a field of the shooting direction in association with the field of the target image TI. The image information 500 is generated by setting information in each field when the object T is photographed by the photographing apparatus C.

  The target image TI is an image obtained by photographing the target T. The imaging direction is the direction of the imaging device C when the object T is imaged. The photographing direction is, for example, a direction in which the photographing lens of the photographing device C measured by the compass 408 built in the photographing device C that photographed the target T is directed. The photographing direction is not set when the photographing apparatus C does not have the compass 408, and may be empty data, for example.

(Data structure of object information table 600)
Next, with reference to FIG. 6, the data structure of the object information table 600 that is stored by the imaging instruction device S and used when the object corresponding to the target T is specified based on the image information 500 shown in FIG. An example will be described.

  FIG. 6 is an explanatory diagram showing an example of the data structure of the object information table 600. As shown in FIG. 6, the object information table 600 has an object name, related information, and an image address field in association with the object ID field. The object information table 600 stores object information as a record 600-a by setting information in each field for each object. a is an arbitrary natural number of 1 to n.

  The object ID is object identification information. The object name is a name assigned to the object identified by the object ID. The related information is information related to the object identified by the object ID. The related information is, for example, explanation of an object identified by the object ID, instruction manual, or management document information. The image address is an address of the photographing information table 700 that stores an image obtained by photographing an object identified by the object ID from each direction.

(Data structure of the shooting information table 700)
Next, referring to FIG. 7, a shooting information table 700 stored with the shooting instruction device S and corresponding to the image address of the object information table 600 shown in FIG. 6 is set. An example of the data structure will be described.

  FIG. 7 is an explanatory diagram showing an example of the data structure of the shooting information table 700. As illustrated in FIG. 7, the shooting information table 700 includes fields of an image and a shooting direction in association with an image ID field. In the shooting information table 700, shooting information is stored as a record 700-b by setting information in each field for each image. b is an arbitrary natural number of 1 to n.

  The image ID is identification information of an image obtained by photographing an object from one direction. An image is an image obtained by photographing an object from one direction, identified by an image ID. The shooting direction is the direction in which the image identified by the image ID is shot.

(Data structure of similarity information table 800)
Next, referring to FIG. 8, the photographing instruction device S uses the target image TI obtained by photographing the target T based on the image information 500 shown in FIG. 5 and the image of the photographing information table 700 shown in FIG. An example of the data structure of the similarity information table 800 in which the similarity is set will be described.

  FIG. 8 is an explanatory diagram showing an example of the data structure of the similarity information table 800. As illustrated in FIG. 8, the similarity information table 800 includes fields for similarity to an object, object ID, and shooting direction in association with an image ID field. In the similarity information table 800, similarity information is stored as a record 800-c by setting information in each field for each image. c is an arbitrary natural number of 1 to n.

  The image ID is identification information of an image obtained by photographing an object from one direction. The similarity with the object is the similarity between the image identified by the image ID and the object image TI obtained by photographing the object T. The object ID is identification information of an object corresponding to the image identified by the image ID. The shooting direction is the direction in which the image identified by the image ID is shot.

(Data structure of comparison information table 900)
Next, with reference to FIG. 9, comparison information in which the degree of similarity between images in the photographing information table 700 illustrated in FIG. 7 is set by the photographing instruction device S based on the similarity information table 800 illustrated in FIG. 8. An example of the data structure of the table 900 will be described.

  FIG. 9 is an explanatory diagram showing an example of the data structure of the comparison information table 900. As illustrated in FIG. 9, the comparison information table 900 includes fields of a comparison source ID, a comparison destination ID, and a similarity between objects in association with an angle difference field. The comparison information table 900 stores the comparison information as a record 900-d by setting information in each field for each combination of images. d is an arbitrary natural number of 1 to n.

  The angle difference is an angle in a predetermined direction from the shooting direction of an image obtained by shooting an object to the shooting direction of an image obtained by shooting the same object. The comparison source ID is an image shot from different shooting directions by an angle difference from a shooting direction of an image having the highest degree of similarity to the target image TI of the target T among images shot from the same object. Identification information. The comparison source ID may be identification information of an image shot from a shooting direction that differs by an angle difference from a shooting direction of an image whose similarity to the target image TI obtained by shooting the target T is equal to or greater than a threshold value. .

  The comparison destination ID is from the image capturing direction of the image having the highest similarity to the target image TI obtained by capturing the object T among images obtained by capturing an object different from the object corresponding to the image identified by the comparison source ID. This is identification information of an image shot from different shooting directions by an angle difference in a predetermined direction. The comparison destination ID may be identification information of an image shot from a shooting direction that differs by an angle difference in a predetermined direction from the shooting direction of an image whose similarity to the target image TI obtained by shooting the target T is equal to or greater than a threshold value. . The similarity between objects is the similarity between the image identified by the comparison source ID and the image identified by the comparison destination ID.

(Data structure of shooting instruction 1000)
Next, an example of the data structure of the shooting instruction 1000 generated by the shooting instruction apparatus S based on the comparison information table 900 shown in FIG. 9 will be described with reference to FIG.

  FIG. 10 is an explanatory diagram showing an example of the data structure of the shooting instruction 1000. As illustrated in FIG. 10, the imaging instruction 1000 includes fields for an instruction direction and target information in association with a determination result field. The shooting instruction 1000 is generated by setting information in each field when the object T is shot by the shooting apparatus C.

  The determination result is a determination result of whether or not an object corresponding to the target object T has been specified. The determination result is, for example, OK when an object corresponding to the target T is specified, or NG when there are a plurality of objects that are candidates for the target T. The designated direction is an imaging direction in which a new target image TI for further narrowing down objects that are candidates for the target T can be captured when there are a plurality of objects that are candidates for the target T. For example, the instruction direction is not set when the determination result is OK, and may be empty data.

  The target information is object information. The object information includes, for example, an object ID, an object name, and related information about the object. The target information is, for example, information on an object corresponding to the target T when the determination result is OK. The target information is information on an object that is a candidate for the target T when the determination result is NG, for example. For example, the target information is not set when the determination result is NG, and may be empty data.

(Functional configuration example of the imaging instruction device S)
Next, a functional configuration example of the imaging instruction device S will be described with reference to FIG.

  FIG. 11 is a block diagram illustrating a functional configuration example of the imaging instruction apparatus S. The imaging instruction apparatus S includes an acquisition unit 1101, a search unit 1102, a selection unit 1103, an output unit 1104, and a determination unit 1105. The acquisition unit 1101 to the determination unit 1105 are functions as a control unit. For example, the acquisition unit 1101 to the determination unit 1105 causes the CPU 301 to execute a program stored in a storage device such as the memory 302 and the disk 305 illustrated in FIG. The function is realized by F303. The processing result of each functional unit is stored in a storage area such as the memory 302 and the disk 305, for example.

  The acquisition unit 1101 acquires a target image TI obtained by photographing the target T. The target object T is an object in which which object among a plurality of objects corresponds is specified. The object T is, for example, a machine in a machine room. The target T may be a living thing. The target image TI is information obtained by converting optical information obtained from a predetermined space including the target T as a subject into an electrical signal. For example, the acquisition unit 1101 receives the image information 500 including the target image TI obtained by capturing the target T from the image capturing apparatus C via the network 210 using the I / F 303. Thereby, the acquisition unit 1101 can acquire the target image TI used for specifying the object corresponding to the target T.

  The acquisition unit 1101 acquires the shooting direction of the target image TI. The shooting direction is a direction from the shooting device C to the target T when the target T is shot. For example, the acquisition unit 1101 receives image information 500 including the direction from the imaging device C to the object T when the object T is imaged as an imaging direction from the imaging device C via the network 210 by the I / F 303. To do. Thereby, the acquisition unit 1101 can acquire the shooting direction used for determining whether or not the target image TI is shot from the shooting direction specified by the shooting instruction 1000.

  For each of the plurality of objects, the search unit 1102 searches for an image similar to the target image TI acquired by the acquisition unit 1101 from images obtained by capturing the object from each of a plurality of shooting directions. For example, the search unit 1102 searches for an image similar to the target image TI from the images stored in the imaging information table 700. Specifically, the search unit 1102 calculates the similarity between each of the images stored in the imaging information table 700 corresponding to each of the image addresses stored in the object information table 600 and the target image TI, A similarity information table 800 is generated.

  Then, the search unit 1102 has the largest similarity to the target image TI among the images stored in the shooting information table 700 corresponding to the image address for each of the image addresses stored in the object information table 600. Search for an image that is equal to or greater than the threshold. The similarity is calculated, for example, by analyzing the shape, color, pattern, etc. of the object shown in the image using edge detection or feature point extraction techniques. Thereby, the search unit 1102 can acquire an image of an object that is a candidate for the target T.

  For each of a plurality of angles, the selection unit 1103 sets the similarity between the images obtained by shooting the object corresponding to each of the images searched by the search unit 1102 from the shooting direction of the image in a predetermined direction by a different angle. Based on this, select one of a plurality of angles.

  For example, the selection unit 1103 calculates the similarity between images obtained by capturing an object corresponding to each of the images searched by the search unit 1102 from shooting directions different by 90 degrees, 180 degrees, and 270 degrees from the shooting direction of the image. calculate. Then, the selection unit 1103 selects an angle having the highest similarity among 90 degrees, 180 degrees, and 270 degrees.

  Specifically, the selection unit 1103 determines the shooting direction of each of the remaining images excluding the image searched by the search unit 1102 among the images obtained by shooting each of the plurality of objects. Specify how much the angle differs from the shooting direction to the right. Next, the selection unit 1103 calculates the similarity between the images of different objects that are identified as being different by the same angle, and generates the comparison information table 900. Then, the selection unit 1103 selects an angle associated with the smallest similarity among the similarities stored in the comparison information table 900.

  In addition, the selection unit 1103 may select any of the angles of 90 degrees, 180 degrees, and 270 degrees whose similarity is equal to or less than the threshold, or each of the angles whose similarity is equal to or less than the threshold. Accordingly, the selection unit 1103 can select an angle used for specifying the shooting direction specified by the shooting instruction 1000.

  The selection unit 1103 captures an object corresponding to each of the images searched by the search unit 1102 for each combination of a plurality of directions and a plurality of angles from the image capturing direction of the image by the angle. The similarity between images taken from the direction is calculated. Then, the selection unit 1103 selects one of a plurality of directions and one of a plurality of angles based on the calculated similarity.

  For example, the selection unit 1103 selects an object corresponding to each of the images searched by the search unit 1102 from the shooting direction of the image by 90 degrees to the right, 180 degrees, 270 degrees different shooting directions, and upward 90 degrees, 270 The similarity between images taken from different shooting directions is calculated. Then, the selection unit 1103 selects the direction and angle having the highest similarity among 90 degrees, 180 degrees, and 270 degrees in the right direction, and 90 degrees and 270 degrees in the upward direction.

  Further, the selection unit 1103 may select a direction and an angle whose similarity is equal to or less than a threshold value among 90 degrees, 180 degrees, 270 degrees in the right direction, and 90 degrees, 270 degrees in the upward direction. Accordingly, the selection unit 1103 can select the direction and angle used for specifying the shooting direction specified by the shooting instruction 1000.

  When there are a plurality of images searched by the search unit 1102, the output unit 1104 outputs a shooting instruction 1000 for the target T from a shooting direction that differs from the shooting direction of the target image TI by a predetermined angle in a predetermined direction. The shooting instruction 1000 is information for instructing the shooting direction of the target T and requesting transmission of the target image TI obtained by shooting the target T to the shooting instruction device S.

  The output unit 1104, for example, instructs to shoot the object T based on the similarity between the images obtained by shooting the objects corresponding to the searched images from a shooting direction different from the shooting direction of the image by a predetermined angle. 1000 is generated. The output unit 1104 displays an imaging instruction 1000 for the object T on a display, prints it out on a printer, transmits it to an external device via the I / F 303, or stores it in a storage area such as the memory 302 or the disk 305.

  Specifically, the output unit 1104 outputs a shooting instruction 1000 for the target T from a shooting direction that is different from the shooting direction of the target image TI by the angle selected by the selection unit 1103 in a predetermined direction. Accordingly, the output unit 1104 can output the shooting instruction 1000 so that a target image TI effective for narrowing down an object that is a candidate for the target T is acquired.

  The output unit 1104 outputs a shooting instruction 1000 for the target T from a shooting direction different from the shooting direction of the target image TI by the angle selected by the selection unit 1103 in the direction selected by the selection unit 1103. Accordingly, the output unit 1104 can output the shooting instruction 1000 so that a target image TI effective for narrowing down an object that is a candidate for the target T is acquired.

  When there is one image searched by the search unit 1102, the output unit 1104 outputs information on an object corresponding to the image searched by the search unit 1102. Thereby, the output unit 1104 can output information on the object corresponding to the target T.

  When there are a plurality of images searched by the search unit 1102 and the objects corresponding to the images searched by the search unit 1102 are the same, the output unit 1104 outputs information on the object. Thereby, the output unit 1104 can output information on the object corresponding to the target T.

  After the output unit 1104 outputs the shooting instruction 1000, the determination unit 1105 is instructed by the shooting instruction 1000 output by the output unit 1104 for the shooting direction of the target image TI obtained by shooting the target T acquired by the acquisition unit 1101. It is determined whether or not it matches the shooting direction. Thereby, the determination unit 1105 can determine whether or not the target image TI acquired by the acquisition unit 1101 is the target image TI shot according to the shooting instruction 1000.

  The output unit 1104 outputs a shooting instruction 1000 when the determination unit 1105 determines that they do not match. Thereby, when the target image TI acquired by the acquisition unit 1101 is not the target image TI captured according to the imaging instruction 1000, the output unit 1104 outputs the imaging instruction 1000 so that the target image TI is captured again. be able to.

(Flow of object identification process corresponding to the object T)
Next, the flow of the object specifying process corresponding to the target T performed by the imaging instruction device S will be described with reference to FIGS.

  12-14 is explanatory drawing which shows the flow of the specific process of the object corresponding to the target object T. FIG. In FIG. 12, (11) the photographing instruction device S receives a target image A obtained by photographing the target T from the photographing device C.

  (12) The imaging instruction apparatus S captures the object PC-1 stored in the imaging information table 700 corresponding to the received target image A and the object PC-1 from the imaging direction “0 degree forward”. The degree of similarity “0.91” with 1 is calculated.

  (13) The imaging instruction apparatus S includes the object ID “PC-1” of the object PC-1, the imaging direction “0 ° forward”, the image ID “1-1” of the image 1-1, and the similarity “0”. .91 ”is generated and added to the similarity information table 800.

  (14) The imaging instruction apparatus S also performs (11) to (11) on other images stored in the imaging information table 700 corresponding to the object PC-1 and the imaging information table 700 corresponding to the objects PC-2 and PC-3. The same processing as 13) is performed. Thereby, the imaging instruction device S can calculate the similarity to the target image A obtained by capturing the target T, which is used for narrowing down the objects that are candidates for the target T. Here, the description shifts to the description of FIG.

  In FIG. 13, (15) the imaging instruction apparatus S captures images 1-1, 2-1, and 3-2 that are images whose similarity is equal to or greater than a threshold among images stored in the similarity information table 800. PC-1, PC-2, and PC-3 are specified as objects that are candidates for the target T.

  (16) The imaging instruction apparatus S sequentially selects the objects corresponding to the first record in the similarity information table 800 among the objects that are candidates for the target T. Here, the imaging instruction device S specifies the object PC-1 corresponding to the record stored at the top of the similarity information table 800.

  (17) The photographing instruction device S specifies the photographing direction of the image 1-1. Further, the imaging instruction device S refers to the imaging information table 700, and among the images corresponding to the object PC-1, the remaining images 1-2 and 1 excluding the image 1-1 whose similarity is equal to or higher than a threshold value. -3 and 1-4 are specified.

  Next, in the shooting instruction device S, the shooting direction of each of the remaining images 1-2, 1-3, and 1-4 is a shooting direction that is slightly different in angle from the shooting direction of the image 1-1 to the right. Is identified. Then, the imaging instruction device S stores in the relative list 1300 a record in which the identification information of the object PC-1 and the remaining images 1-2, 1-3, and 1-4 and the specified angles are associated with each other. to add.

  (18) The imaging instruction apparatus S performs the same processing as (16) to (17) for the objects PC-2 and PC-3. As a result, the photographing instruction device S can convert the absolute photographing direction of each image into a relative photographing direction based on the photographing direction of an image whose similarity is equal to or greater than a threshold value. Here, the description shifts to the description of FIG.

  14, (19) the imaging instruction device S selects the image 1-2 of the object PC-1 based on the relative list 1300. Further, the imaging instruction device S selects an image 2-3 that is an image of the object PC-2 different from the object PC-1 and that is associated with the same angle as the selected image 1-2.

  Next, the imaging instruction device S calculates the degree of similarity between the selected image 1-2 and the selected image 2-3. Then, the imaging instruction apparatus S associates the angle associated with the image, the identification information of the images 1-2 and 2-3, and the similarity between the images, and adds them to the comparison information table 900.

  (20) Similarly, the imaging instruction device S selects the image 1-2 of the object PC-1 based on the relative list 1300. In addition, the imaging instruction device S selects an image 3-4 that is an image of the object PC-3 different from the object PC-1 and is associated with the same angle as the selected image 1-2.

  Next, the imaging instruction device S calculates the degree of similarity between the selected image 1-2 and the selected image 3-4. Then, the imaging instruction apparatus S associates the angle associated with the image, the identification information of the images 1-2 and 3-4, and the similarity between the images, and adds them to the comparison information table 900.

  (21) The imaging instruction apparatus S performs the same processing as (19) to (20) for other images. Thereby, the imaging instruction device S can calculate the degree of similarity between images captured from the same relative imaging direction.

  (22) The imaging instruction device S selects an angle associated with the smallest similarity among the angles stored in the comparison information table 900. Then, the photographing instruction device S generates a photographing instruction 1000 for the target T from a photographing direction that is different from the photographing direction of the target image A for photographing the target T by the angle selected to the right, and transmits it to the photographing device C. Thereby, the imaging instruction device S can specify a more effective imaging direction for narrowing down objects that are candidates for the target T based on the similarity.

  In addition, the imaging instruction device S may select each of the angles associated with the degree of similarity equal to or greater than the threshold among the angles stored in the comparison information table 900. Then, the imaging instruction device S generates an imaging instruction 1000 for the object T from the imaging direction that is different by the selected angle from the imaging direction of the object image A in which the object T is imaged, to the imaging device C. Send. Thereby, the imaging instruction apparatus S can transmit to the imaging apparatus C an imaging instruction for instructing a plurality of imaging direction candidates, and notify the user of the imaging apparatus C.

  As a result, the user of the photographing apparatus C can photograph the object T by selecting a photographing direction that is easy to photograph among the designated photographing directions. In addition, even if it is difficult to shoot from a specific shooting direction due to the physical situation where the target T is installed, it is only necessary to shoot the target T from any of the specified shooting directions. Convenience for the user of the photographing apparatus C can be improved.

(Example of object specifying process corresponding to the object PC-1)
Next, an example of an object specifying process corresponding to the target object PC-1 performed by the imaging instruction device S will be described with reference to FIGS.

  15-20 is explanatory drawing which shows an example of the specific process of the object corresponding to target object PC-1. In the following, the imaging instruction device S stores, as the imaging information table 700, images obtained by imaging each of the objects PC-1, PC-2, and PC-3 shown in FIG. The case where the object corresponding to the target object PC-1 is specified will be described.

  Here, the description shifts to the description of FIG. 16, and a shooting information table 700 in which images obtained by shooting each of the objects PC-1, PC-2, and PC-3 shown in FIG. 15 from a plurality of shooting directions are set. Will be described.

  As illustrated in FIG. 16, the shooting information table 700 is associated with the object PC-1 and includes an image 1-1 obtained by photographing the object PC-1 from 0 degrees ahead, an image 1-2 obtained by photographing 180 degrees from the rear, and the right. An image 1-3 taken from 90 degrees and an image 1-4 taken from 90 degrees to the left are stored.

  The shooting information table 700 is associated with the object PC-2, the image 2-1 obtained by shooting the object PC-2 from 0 degrees ahead, the image 2-2 shot from 180 degrees behind, and the image 2 shot from 90 degrees to the right. -3, image 2-4 photographed from 90 degrees left is stored.

  The shooting information table 700 is associated with the object PC-3, an image 3-1 obtained by shooting the object PC-3 from 0 degrees ahead, an image 3-2 taken from 180 degrees behind, and an image 3 taken from 90 degrees to the right. -3, the image 3-4 taken from the left 90 degrees is stored. Here, the description shifts to FIG.

  In FIG. 17, the photographing apparatus C displays a photographing instruction screen including a message that prompts photographing of the object PC-1. When the object PC-1 is photographed by a user's operation input, the image capturing apparatus C transmits a target image P-1 obtained by capturing the object PC-1 to the image capturing instruction apparatus S. Here, the description shifts to the description of FIG.

  In FIG. 18, the imaging instruction device S receives the target image P-1. The imaging instruction device S calculates the similarity between the target image P-1 and each of the images stored in the imaging information table 700. The similarity is, for example, in a range of 0 to 1 and indicates that the larger the value, the more similar. The imaging instruction device S calculates, for example, a similarity “0.98” between the target image P-1 and an image 1-2 obtained by imaging the object PC-1 from 180 degrees behind.

  In addition, the imaging instruction device S, for example, the similarity “0.98” between the target image P-1 and the image 2-1 obtained by imaging the object PC-2 from 0 degrees ahead, the target image P-1, and the object The degree of similarity “0.98” with the image 2-2 taken from 180 degrees behind PC-2 is calculated.

  In addition, the imaging instruction device S, for example, has a similarity “0.98” between the target image P-1 and the image 3-1 obtained by imaging the object PC-3 from 0 degrees ahead, the target image P-1, and the object The degree of similarity “0.98” with the image 3-2 taken from 180 degrees behind PC-3 is calculated.

  Next, the imaging instruction device S displays a record that associates the calculated similarity, the image from which the similarity is calculated, the object from which the image is captured, and the imaging direction of the image with the similarity. Sorted in descending order and added to the similarity information table 800.

  Then, based on the similarity information table 800, the imaging instruction device S has the image 1-2, the image 2-1, the image 2-2, the image 3-1, Specify 3-2. Objects PC-1 and PC in which the image 1-2, the image 2-1, the image 2-2, the image 3-1, and the image 3-2 identified as having a similarity equal to or greater than the threshold “0.8” -2, PC-3 are objects that are candidates for the object. Here, the description shifts to FIG.

  In FIG. 19, the imaging instruction device S captures images of the same object taken from the respective imaging directions of the specified image 1-2, image 2-1, image 2-2, image 3-1, and image 3-2. Specify how much the angle is different up to the shooting direction.

  The imaging instruction device S has, for example, the respective imaging directions “0 degrees front, 90 degrees right, 90 degrees left” of the images 1-1, 1-3, and 1-4, which are images of the same object as the image 1-2. It is specified that the angle “180 degrees, 270 degrees, 90 degrees” differs from the shooting direction “180 degrees backward” of the image 1-2 to the right.

  The photographing instruction device S has, for example, the respective photographing directions “180 degrees backward, 90 degrees right, 90 degrees left” of the images 2-2, 2-3, and 2-4 obtained by photographing the same object as the image 2-1. It is specified that the angle “180 degrees, 90 degrees, 270 degrees” differs from the shooting direction “0 degrees in front” of the image 2-1 to the right.

  The imaging instruction device S has, for example, the respective imaging directions “0 degrees front, 90 degrees right, 90 degrees left” of the images 2-1, 2-3, 2-4 obtained by imaging the same object as the image 2-2. It is specified that the angle “180 degrees, 270 degrees, 90 degrees” differs from the shooting direction “180 degrees backward” of the image 2-2 to the right.

  The imaging instruction device S has, for example, respective imaging directions “180 degrees backward, 90 degrees right, 90 degrees left” of the images 3-2, 3-3, and 3-4, which are the same object as the image 3-1. It is specified that the angle “180 degrees, 90 degrees, 270 degrees” differs from the shooting direction “0 degrees in front” of the image 3-1 to the right.

  The imaging instruction device S has, for example, the respective imaging directions “0 ° front, 90 ° right, 90 ° left” of the images 3-1, 3-3, 3-4 obtained by imaging the same object as the image 3-2. It is specified that the angle “180 degrees, 270 degrees, 90 degrees” differs from the shooting direction “180 degrees backward” of the image 3-2 to the right. The specified angle is a value that indicates a relative shooting direction when the shooting direction of the image specified as having a similarity of not less than the threshold value “0.8” is used as a reference. Here, the description shifts to the description of FIG.

  In FIG. 20, the imaging instruction device S calculates the similarity between images of different objects that are identified as being different by the same angle and that are captured from the same relative imaging direction. The imaging instruction device S calculates, for example, the similarity “0.02” between the image 1-4 and the image 3-3 that are specified to be different from each other by the same angle “90 degrees” in the right direction.

  Next, the imaging instruction apparatus S sorts the records in which the image IDs of the images, the angles specified for the images and the calculated similarity are associated with each other in ascending order of similarity, and compares the comparison information table 900. Add to Then, the imaging instruction device S selects the angle “90 degrees” associated with the smallest similarity “0.02” among the similarities stored in the comparison information table 900.

  Next, the imaging instruction device S generates an imaging instruction 1000 for the object PC-1 from an imaging direction different from the imaging direction of the target image P-1 by an angle “90 degrees” selected in the right direction. To C. When the imaging apparatus C receives the imaging instruction 1000, the imaging apparatus C displays an imaging instruction screen that displays the imaging direction instructed by the imaging instruction 1000 and displays a message that prompts the imaging of the object PC-1.

  When the object PC-1 is photographed by a user's operation input, the image capturing apparatus C transmits a target image P-1 obtained by capturing the object PC-1 to the image capturing instruction apparatus S. Thereby, the imaging instruction device S can narrow down objects that are candidates for the target PC-1 efficiently.

  The imaging instruction device S, for example, is a newly acquired target image P-1 and images 1-4, 2-3, 2- taken from the same relative imaging direction indicated by the selected angle “90 degrees”. The degree of similarity with 4, 3-3, 3-4 is calculated. Then, the imaging instruction device S selects an object from which an image having a similarity equal to or greater than a threshold is captured as an object candidate from among the objects PC-1, PC-2, and PC-3 that are object candidates. Narrow down as an object.

(An example of the confirmation process of the object PC-1)
Next, an example of the confirmation process of the target object PC-1 performed by the imaging instruction device S after specifying the object corresponding to the target object PC-1 will be described with reference to FIG.

  FIG. 21 is an explanatory diagram illustrating an example of a confirmation process for the object PC-1. In FIG. 21, the photographing instruction device S receives the target image TI photographed according to the photographing instruction from the photographing device C. The imaging instruction apparatus S is able to identify an object corresponding to the target object PC-1 because the candidate object for the target object PC-1 is narrowed down to one based on the received target image TI. .

  When the imaging instruction device S specifies the object corresponding to the target object PC-1, the imaging instruction device S acquires information on the location of the object to which the identification information of the specified object is attached from a storage area such as the memory 302 or the disk 305. The imaging instruction device S generates an input instruction including the acquired location information and transmits the input instruction to the imaging device C.

  When receiving the input instruction, the imaging apparatus C notifies the user of the location instructed by the input instruction. The photographing apparatus C transmits the identification information to the photographing instruction apparatus S when the pasted identification information is photographed by the user's operation input or when the identification information is input. When receiving the identification information, the imaging instruction apparatus S transmits information on the object identified by the identification information to the imaging apparatus C. When the imaging apparatus C receives the object information, the imaging apparatus C displays the object information. Thereby, after specifying the object corresponding to the target object PC-1, the imaging instruction apparatus S can confirm whether or not the object specified as the object corresponding to the target object PC-1 is correct.

  In addition, after the object corresponding to the object is identified and the type of the object is determined, the imaging instruction device S is one of a plurality of objects of the same type based on the identification information. And the management information specific to the object may be transmitted to the imaging apparatus C. Thereby, the imaging instruction device S can notify the user of the imaging device C of management information unique to the object.

(Specific processing procedure)
Next, an example of the specific processing procedure will be described with reference to FIG.

  FIG. 22 is a sequence diagram illustrating an example of the specific processing procedure. In FIG. 22, the photographing apparatus C acquires a target image TI obtained by photographing the target T in response to a user operation input, and generates image information 500 including the target image TI (step S2201). The photographing apparatus C transmits the generated image information 500 to the photographing instruction apparatus S (step S2202).

  When receiving the image information 500, the imaging instruction apparatus S executes image determination processing described later in FIG. 23 (step S2203). Then, the imaging instruction device S determines whether or not there are a plurality of objects identified as objects that are candidates for the target T based on the image information 500 (step S2204).

  If there are a plurality of identified objects (step S2204: YES), the imaging instruction device S performs an image comparison process described later with reference to FIG. 24 (step S2205). Then, the imaging instruction device S transmits an imaging instruction 1000 to the imaging device C (step S2206).

  Upon receiving the shooting instruction 1000, the shooting apparatus C displays the shooting direction specified by the shooting instruction 1000 (step S2207). Then, the photographing apparatus C returns to the process of step S2201.

  On the other hand, when there is one identified object (step S2204: No), the imaging instruction device S transmits information on the identified object to the imaging device C (step S2208).

  When receiving the object information, the imaging apparatus C displays the object information (step S2209). Thereby, the imaging instruction device S can specify an object corresponding to the target T, and can transmit information on the object corresponding to the target T to the imaging device C. Thereby, the imaging instruction device S can identify an object corresponding to the target T, and can notify the user of the imaging device C.

(Image determination processing procedure)
Next, an example of the image determination processing procedure shown in step S2203 of FIG. 22 will be described using FIG.

  FIG. 23 is a flowchart illustrating an example of an image determination processing procedure. In FIG. 23, the imaging instruction device S selects an image stored in the imaging information table 700 (step S2301). Next, the imaging instruction device S calculates the similarity between the selected image and the target image TI obtained by imaging the target T (step S2302). Then, the imaging instruction device S adds records to the similarity information table 800 while sorting in descending order of similarity (step S2303).

  Next, the imaging instruction device S determines whether or not all images have been selected (step S2304). If there is an unselected image (step S2304: NO), the imaging instruction device S returns to the process of step S2301.

  On the other hand, when all the images have been selected (step S2304: YES), the imaging instruction device S determines whether or not there are a plurality of images whose similarity is equal to or greater than a threshold among the images stored in the similarity information table 800. Is determined (step S2305). Here, when there is one image whose similarity is greater than or equal to the threshold (step S2305: No), the imaging instruction device S corresponds to the target T with an object in which an image whose similarity is greater than or equal to the threshold is captured. (Step S2306), and the image determination process ends.

  On the other hand, when there are a plurality of images having a degree of similarity equal to or greater than the threshold (step S2305: Yes), the imaging instruction device S regards an object from which an image having a degree of similarity equal to or greater than the threshold as a candidate for the target T. (Step S2307), and the image determination process ends. Thereby, the imaging instruction apparatus S can narrow down objects that are candidates for the target T.

(Image comparison processing procedure)
Next, an example of the image comparison processing procedure shown in step S2205 of FIG. 22 will be described using FIG.

  FIG. 24 is a flowchart illustrating an example of an image comparison processing procedure. In FIG. 24, the imaging instruction device S selects one of the objects that are candidates for the target T (step S2401). Next, the imaging instruction device S specifies a remaining image excluding an image whose similarity is equal to or greater than a threshold among images obtained by imaging the selected object (step S2402). Then, the shooting instruction apparatus S specifies the shooting direction of each of the remaining images so that the angle is different from the shooting direction of the image whose similarity is equal to or greater than the threshold value to the right (step S2403).

  Next, the imaging instruction device S adds the specified angle to the relative list 1300 in association with each of the remaining images (step S2404). Then, the imaging instruction device S determines whether or not all objects that are candidates for the target T have been selected (step S2405). Here, when there is an unselected object (step S2405: No), the imaging instruction device S returns to the process of step S2401.

  On the other hand, when all the objects have been selected (step S2405: Yes), the imaging instruction device S selects images associated with the same angle from the remaining images stored in the relative list 1300 (step S2405: Yes). S2406). Next, the imaging instruction device S calculates the similarity between the selected images (step S2407). Then, the imaging instruction device S sorts the records in which the selected images, the angles associated with the images in common and the calculated similarities are associated with each other while sorting the images in the descending order of similarity. It adds to the table 900 (step S2408).

  Next, the imaging instruction device S determines whether or not all images have been selected (step S2409). If there are unselected images (step S2409: NO), the imaging instruction device S returns to the process of step S2406.

  On the other hand, when all the images are selected (step S2409: Yes), the imaging instruction device S selects an angle corresponding to the smallest similarity among the angles stored in the comparison information table 900 (step S2410). ). Then, the imaging instruction device S ends the image comparison process. Thereby, the imaging instruction device S can specify a more effective imaging direction for narrowing down objects that are candidates for the target T based on the similarity.

  As described above, according to the photographing instruction device S, each of the plurality of objects is similar to the target image TI obtained by photographing the target T from the images obtained by photographing the object from the plurality of photographing directions. You can search for images. Here, according to the imaging instruction device S, when there are a plurality of searched images, an object corresponding to each of the searched images is imaged from a shooting direction different from the shooting direction of the image by a predetermined angle by a predetermined angle. The similarity between images can be calculated. According to the photographing instruction device S, it is possible to output a photographing instruction of the target T from a photographing direction that differs by a predetermined angle from the photographing direction of the target image TI based on the similarity. Thereby, the imaging instruction device S outputs an imaging instruction so that an effective target image TI is acquired for narrowing down the objects that are candidates for the target T, and based on the acquired target image TI. Objects that are candidates for the target T can be narrowed down efficiently. And the imaging | photography instruction | indication apparatus S can pinpoint the object corresponding to the target object T efficiently.

  According to the photographing instruction device S, for each of a plurality of angles, the similarity between images obtained by photographing an object corresponding to each searched image from a photographing direction different from the photographing direction by a predetermined direction from the photographing direction of the image is calculated. can do. Then, according to the photographing instruction device S, one of a plurality of angles is selected based on the calculated similarity, and the object from the photographing direction different from the photographing direction of the target image TI in the predetermined direction is selected. An imaging instruction for the object T can be output. Thereby, the imaging instruction device S can specify a more effective imaging direction for narrowing down objects that are candidates for the target T based on the similarity, and output an imaging instruction from the imaging direction. An object can be identified efficiently.

  According to the imaging instruction device S, for a combination of each of a plurality of directions and each of a plurality of angles, an object corresponding to each of the retrieved images is different from the imaging direction of the image by the angle by the angle. It is possible to calculate the degree of similarity between images taken from. Then, according to the photographing instruction device S, the direction selected from the photographing direction of the target image TI is selected based on the calculated similarity, by selecting one of the plurality of directions and one of the plurality of angles. It is possible to output a shooting instruction of the target T from different shooting directions corresponding to the selected angle. Thereby, the imaging instruction device S can be applied even when the imaging direction includes directions from front and rear, right and left, up and down, or diagonally.

  According to the imaging instruction device S, when there is one searched image, information on an object corresponding to the searched image can be output. As a result, the imaging instruction device S identifies the object information corresponding to the searched image as the object corresponding to the target T, and notifies the user of the imaging device C of the information on the object corresponding to the target T. can do.

  According to the imaging instruction device S, when there are a plurality of searched images and the objects corresponding to the searched images are the same, information on the objects can be output. As a result, even if there are a plurality of searched images, the shooting instruction device S does not output a shooting instruction when the number of objects that are candidates for the target T is narrowed down to one. The increase can be suppressed. Further, the photographing instruction device S does not output a photographing instruction, suppresses an increase in the number of times the user of the photographing device C photographs the object T, and increases the work amount of the user of the photographing device C. Can be suppressed.

  According to the photographing instruction device S, the photographing direction of the target image TI obtained by photographing the target T obtained after outputting the photographing instruction is obtained, and whether the obtained photographing direction matches the photographing direction designated by the photographing instruction. It can be determined whether or not. Then, according to the photographing instruction device S, when it is determined that they do not match, a photographing instruction can be output. As a result, when the user of the photographing apparatus C makes a mistake in the direction of photographing the target object T, the photographing instruction apparatus S notifies the user of the photographing apparatus C of the photographing instruction again to photograph the target object T. be able to.

  Note that the photographing instruction method described in this embodiment can be realized by executing a program prepared in advance on a computer such as a personal computer or a workstation. This photographing instruction program is recorded on a computer-readable recording medium such as a hard disk, a flexible disk, a CD-ROM, an MO, and a DVD, and is executed by being read from the recording medium by the computer. The present photographing instruction program may be distributed via a network such as the Internet.

  The following additional notes are disclosed with respect to the embodiment described above.

(Supplementary note 1)
Get the target image of the target,
For each of a plurality of objects, search for an image similar to the acquired target image from images obtained by capturing the object from each of a plurality of shooting directions,
When there are a plurality of the searched images, based on the similarity between images obtained by shooting the object corresponding to each of the searched images from a shooting direction different from the shooting direction of the image by a predetermined angle, Outputting a shooting instruction of the object from a shooting direction different from the shooting direction of the target image by the predetermined angle in the predetermined direction;
An imaging instruction program for executing a process.

(Supplementary note 2)
For each of a plurality of angles, calculate the similarity between images obtained by shooting the object corresponding to each of the searched images from the shooting direction of the image from the shooting direction different by the angle from the shooting direction of the image,
Selecting one of the plurality of angles based on the calculated similarity;
Let the process run,
The output process is as follows:
Outputting a shooting instruction of the object from a shooting direction different from the shooting direction of the target image by the angle selected in the predetermined direction;
The shooting instruction program according to appendix 1, characterized in that:

(Supplementary note 3) The calculation process is as follows.
For a combination of each direction of a plurality of directions and each angle of a plurality of angles, an object corresponding to each of the searched images was photographed from a photographing direction different from the photographing direction of the image by the angle. Calculate the similarity between images,
The process to select is
Based on the calculated similarity, select one of the plurality of directions and one of the plurality of angles,
The output process is as follows:
Outputting a shooting instruction of the object from a shooting direction different from the angle selected in the direction selected from the shooting direction of the target image;
The shooting instruction program according to appendix 2, characterized in that:

(Supplementary note 4)
If the searched image is one, information on the object corresponding to the searched image is output.
The imaging instruction program according to any one of appendices 1 to 3, wherein the program is executed.

(Supplementary note 5)
When there are a plurality of searched images and the objects corresponding to the searched images are the same, information on the objects is output.
The imaging instruction program according to any one of appendices 1 to 4, wherein the program is executed.

(Appendix 6)
Obtaining the photographing direction of the target image obtained by photographing the target object obtained after outputting the photographing instruction;
Determining whether the acquired shooting direction matches the shooting direction instructed by the shooting instruction;
Let the process run,
The output process is as follows:
If it is determined that they do not match, the shooting instruction is output.
The imaging instruction program according to any one of appendices 1 to 5, characterized in that:

(Appendix 7) Acquiring a target image of the target object,
For each of a plurality of objects, search for an image similar to the acquired target image from images obtained by capturing the object from each of a plurality of shooting directions,
When there are a plurality of the searched images, based on the similarity between images obtained by shooting the object corresponding to each of the searched images from a shooting direction different from the shooting direction of the image by a predetermined angle, Outputting a shooting instruction of the object from a shooting direction different from the shooting direction of the target image by the predetermined angle in the predetermined direction;
An imaging instruction apparatus having a control unit.

(Appendix 8) The computer
Get the target image of the target,
For each of a plurality of objects, search for an image similar to the acquired target image from images obtained by capturing the object from each of a plurality of shooting directions,
When there are a plurality of the searched images, based on the similarity between images obtained by shooting the object corresponding to each of the searched images from a shooting direction different from the shooting direction of the image by a predetermined angle, Outputting a shooting instruction of the object from a shooting direction different from the shooting direction of the target image by the predetermined angle in the predetermined direction;
An imaging instruction method characterized by executing processing.

S photographing instruction device C photographing device 200 object identification system 210 network 300, 400 bus 301, 401 CPU
302, 402 Memory 303, 403 I / F
304, 404 Disc drive 305, 405 Disc 406 Display 407 Camera 408 Compass 500 Image information 600 Object information table 700 Shooting information table 800 Similarity information table 900 Comparison information table 1000 Shooting instruction 1101 Acquisition unit 1102 Search unit 1103 Selection unit 1104 Output unit 1105 determination unit

Claims (7)

On the computer,
Get the target image of the target,
For each of a plurality of objects, search for an image similar to the acquired target image from images obtained by capturing the object from each of a plurality of shooting directions,
When there are a plurality of the searched images, based on the similarity between images obtained by shooting the object corresponding to each of the searched images from a shooting direction different from the shooting direction of the image by a predetermined angle, Outputting a shooting instruction of the object from a shooting direction different from the shooting direction of the target image by the predetermined angle in the predetermined direction;
An imaging instruction program for executing a process. In the computer,
For each of a plurality of angles, calculate the similarity between images obtained by shooting the object corresponding to each of the searched images from the shooting direction of the image from the shooting direction different by the angle from the shooting direction of the image,
Selecting one of the plurality of angles based on the calculated similarity;
Let the process run,
The output process is as follows:
Outputting a shooting instruction of the object from a shooting direction different from the shooting direction of the target image by the angle selected in the predetermined direction;
The imaging instruction program according to claim 1, wherein: The calculation process is as follows:
For a combination of each direction of a plurality of directions and each angle of a plurality of angles, an object corresponding to each of the searched images was photographed from a photographing direction different from the photographing direction of the image by the angle. Calculate the similarity between images,
The process to select is
Based on the calculated similarity, select one of the plurality of directions and one of the plurality of angles,
The output process is as follows:
Outputting a shooting instruction of the object from a shooting direction different from the angle selected in the direction selected from the shooting direction of the target image;
The imaging instruction program according to claim 2, wherein: In the computer,
If the searched image is one, information on the object corresponding to the searched image is output.
The imaging instruction program according to claim 1, wherein the process is executed. In the computer,
When there are a plurality of searched images and the objects corresponding to the searched images are the same, information on the objects is output.
The photographing instruction program according to any one of claims 1 to 4, wherein the processing is executed. Get the target image of the target,
For each of a plurality of objects, search for an image similar to the acquired target image from images obtained by capturing the object from each of a plurality of shooting directions,
When there are a plurality of the searched images, based on the similarity between images obtained by shooting the object corresponding to each of the searched images from a shooting direction different from the shooting direction of the image by a predetermined angle, Outputting a shooting instruction of the object from a shooting direction different from the shooting direction of the target image by the predetermined angle in the predetermined direction;
An imaging instruction apparatus having a control unit. Computer
Get the target image of the target,
For each of a plurality of objects, search for an image similar to the acquired target image from images obtained by capturing the object from each of a plurality of shooting directions,
When there are a plurality of the searched images, based on the similarity between images obtained by shooting the object corresponding to each of the searched images from a shooting direction different from the shooting direction of the image by a predetermined angle, Outputting a shooting instruction of the object from a shooting direction different from the shooting direction of the target image by the predetermined angle in the predetermined direction;
An imaging instruction method characterized by executing processing.

Images