Search apparatus, search method, and search system

Abstract

To provide a search apparatus, a search method, and a search system capable of finding a search object in a short time.SOLUTION: A search apparatus 20 includes: an information acquisition unit 21 which acquires search object information which is information on a search object; an image generation unit 22 which generates a simulated image of the search object from the search object information, as a generated image; a distribution unit 23 which distributes the search object information and the generated image to one or more imaging apparatuses 30. The search object information includes at least characteristic information which is information on characteristics of the search object. The image generation unit 22 may have a function of generating a generated image using an image dictionary trained by machine learning, from the characteristic information.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present disclosure relates to a search device, a search method, and a search system. [Background technology]

[0002] Patent Document 1 discloses a missing person search system that searches for images containing people who match or resemble the missing person when searching for a missing person, and uses these images as search information. The missing person search system of Patent Document 1 searches for images that have been arbitrarily uploaded to cloud storage by an unspecified number of users. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2021-149237 [Non-patent literature]

[0004] [Non-Patent Document 1] PC Koubou Next Mag, [Searched on January 29, 2024], Internet<https: / / www.pc-koubou.jp / magazine / 43855> Summary of the Invention [Problem to be solved by the invention]

[0005] Patent Document 1 discloses a technology for searching for missing persons that does not rely solely on information such as images provided by a search requester, but instead uses matching or similar images from images uploaded to the cloud. However, this technology is limited to selecting and using images stored in storage, and does not anticipate generating new images to use in a search. Therefore, if the appearance of the search target in past images has changed from the current appearance of the search target, it takes a long time to find the search target.

[0006] The present disclosure has been made to solve such problems, and aims to provide a search device, search method, and search system that can find a search target, such as a person being searched for, in a short period of time. [Means for solving the problem]

[0007] Therefore, the present disclosure provides a search device including an information acquisition unit that acquires search target information from information relating to a search target, an image generation unit that generates an image of the search target as a generated image from the search target information, and a distribution unit that distributes the search target information and the generated image to one or more imaging devices.

[0008] The present disclosure also provides a search method for a search device having an information acquisition unit that acquires search target information, which is information related to a search target; an image generation unit that generates an image of the search target from the search target information as a generated image; and a distribution unit that distributes the search target information and the generated image to one or more imaging devices, the search method comprising the steps of: causing the information acquisition unit to acquire the search target information; causing the image generation unit to generate the image of the search target from the search target information as the generated image; and causing the distribution unit to distribute the search target information and the generated image to one or more of the imaging devices.

[0009] Furthermore, the present disclosure provides a search system comprising a search device and one or more imaging devices, wherein the search device has an information acquisition unit that acquires search target information, which is information regarding a search target, an image generation unit that generates an image of the search target from the search target information as a generated image, and a first distribution unit that distributes the search target information and the generated image to one or more imaging devices, and the imaging device has a generated image receiving unit that receives the generated image from the search device, an imaging unit that images the imaging target, a search candidate extraction unit that extracts search candidates that are candidates for the search target from the images captured by the imaging unit, an image similarity determination unit that determines the similarity between the search candidate in the captured image and the search target in the generated image, and a second distribution unit that distributes the captured image of the search candidate, whose similarity is determined to be greater than or equal to a predetermined value by the image similarity determination unit, to the search device as a candidate image. [Effects of the Invention]

[0010] The present disclosure makes it possible to provide a search device, a search method, and a search system that can find a search target in a short time. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a configuration diagram illustrating a search system according to a first embodiment. [Figure 2] 1 is a block diagram illustrating a search device according to a first embodiment. [Figure 3] 1 is a block diagram illustrating an imaging device according to a first embodiment. [Figure 4] FIG. 3 is a flowchart illustrating a search method using the search device according to the first embodiment. [Figure 5] FIG. 10 is a flowchart illustrating a search method using the imaging device according to the second embodiment. [Figure 6] FIG. 10 is a block diagram illustrating a search device according to a second embodiment. [Figure 7] FIG. 10 is a flowchart illustrating a search method using the search device according to the second embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0012] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. In each drawing, the same or corresponding elements are designated by the same reference numerals, and for clarity of explanation, duplicate explanations are omitted as necessary. In addition, some reference numerals are omitted to avoid cluttering the drawings.

[0013] [Embodiment 1] <Search System> A search system according to a first embodiment will be described. FIG. 1 is a configuration diagram illustrating the search system 1 according to the first embodiment. As shown in FIG. 1, the search system 1 includes one or more terminals 10, a search device 20, and one or more imaging devices 30. Note that a plurality of terminals 10, including terminals 10a and 10b, or a single terminal 10, will be collectively referred to as terminals 10. A plurality of imaging devices 30, including imaging devices 30a and 30b, or a single imaging device 30, will be collectively referred to as imaging device 30. The terminals 10, the search device 20, and the imaging device 30 are interconnected via a network NW in a state in which information can be transmitted. The network NW includes communication lines such as the Internet, a local area network (LAN), and a wide area network (WAN), as well as communication networks such as base stations. The network NW may be wired or wireless, as long as the connections are capable of transmitting information.

[0014] The search system 1 includes, for example, a system for searching for a missing person such as a wandering elderly person. Note that the search target is not limited to people such as missing persons and children, but may also include missing pets such as dogs and cats.

[0015] <terminal> The terminal 10 includes communication devices such as a smartphone, tablet, or PC (Personal Computer) owned by the person requesting the search. The terminal 10 may also include communication devices such as a telephone or fax. The person requesting the search, such as a family member searching for the search target, sends a search request and information about the search target from the terminal 10 to the search device 20.

[0016] <Search target information> Information about the search target is called "search target information." Search target information includes attribute information and characteristic information. Search target information includes at least characteristic information. "Attribute information" includes information such as the contact information of the person requesting the search, the date of the search request, and the relationship with the search target. "Characteristic information" includes information about the characteristics of the search target. Characteristic information includes at least one of "text information" including the search target's age, gender, facial and body characteristics, and clothing on the day, and "image information" of previous images of the search target. Furthermore, characteristic information includes at least one of "text information" including the type of pet, age, facial and body characteristics, and clothing on the day, and "image information" of previous images of the pet. Images previously captured of the search target are called "search target images." Therefore, search target information includes at least one of text information and a search target image. Search target information may include at least a search target image. The search requester transmits attribute information and characteristic information from the terminal 10 to the search device 20 as search target information.

[0017] <Searching Device> The search device 20 generates various images of the search target based on the search target information acquired from the terminal 10, and distributes them to an imaging device 30, which may include a vehicle's drive recorder, a security camera, or the like. The search device 20 may be installed in the police, a government agency, or the like. For example, the search device 20 may function as a missing person search system for the police, a government agency, or the like. The search device 20 includes an information processing device such as a server, a PC, a tablet, or a smartphone. The search device 20 may also include an information processing device owned by an individual, such as a personal smartphone, tablet, or PC. In this case, the search device 20 may include a form in which the search device requests a few acquaintances to search for the search target.

[0018] 2 is a block diagram illustrating a search device 20 according to embodiment 1. As shown in FIG. 2, the search device 20 includes an information acquisition unit 21, an image generation unit 22, and a distribution unit 23. The search device 20 may further include a control unit 26, a storage unit 27, and a communication unit 28.

[0019] <Information acquisition section> The information acquisition unit 21 acquires search target information. For example, the information acquisition unit 21 acquires characteristic information of a missing wandering elderly person or the like. A person requesting a search, such as a family member, transmits the search target information along with a search request from the terminal 10. In response to this, the information acquisition unit 21 receives the search request from the person requesting the search and the search target information from the terminal 10. It is preferable that the information acquisition unit 21 be provided with an image of the search target that is close to the "current appearance" of the search target by the person requesting the search, such as a family member.

[0020] The search requester may also send characteristic information of the search target along with the search request from a terminal 10 that includes a communication device such as a telephone or fax. In response to this, the information acquisition unit 21 acquires the characteristic information of the search target by automatically converting voice into text data. The information acquisition unit 21 acquires the image of the search target as image information by automatically scanning a photograph sent by fax. Note that an operator who has acquired the characteristic information of the search target via telephone, fax, or the like may manually input the text data or image information from a terminal 10 managed by the operator or a terminal 10 provided in the search device 20.

[0021] <Image Generation Section> Image generation unit 22 generates a generated image based on the search target information, which is an image of the search target. For example, image generation unit 22 generates a generated image for search based on feature information acquired from terminal 10. Image generation unit 22 may generate variations of generated images, such as images of the search target with changes in clothing, facial expression, posture, etc., or images of the search target walking and from behind, based on images of the search target provided by a family member. Image generation unit 22 may also generate a generated image that matches the current age of the search target from an image of the search target, such as a photograph taken several years ago. In this way, image generation unit 22 generates a generated image that resembles the "current appearance" of the search target based on images of the search target provided by a search requester, such as a family member.

[0022] Furthermore, as described above, it is expected that the clothing, facial expression, posture, walking style, back view, etc., of the search target image provided together with the search request will differ from the current appearance of the person wandering around. The search target image may be an old photograph of the search target, or a front-facing ID photo. Therefore, the image generation unit 22 generates generated images including many variations of clothing, facial expression, posture, walking style, back view, etc., from the front-facing ID photo. In this way, the search device 20 can easily find the search target by using generated images that are close to the generated "current appearance" in the search.

[0023] The image generation unit 22 may generate a generated image using an image generation AI. For example, the image generation unit 22 may generate a generated image using an image generation AI from text data such as "a 60-year-old man wearing a blue shirt" and "wearing glasses, short hair, thick eyebrows, and a slightly hunched back," or from an image of the search target provided by a family member. The image generation unit 22 may use a general-purpose image generation AI tool provided by various companies, or may create and use an original image generation AI.

[0024] The image generation unit 22 may have a function of generating a generated image from the feature information using an image dictionary learned by machine learning. In this case, the feature information preferably includes at least information about the age and clothing of the search target. When machine learning is performed on the image dictionary in the image generation AI, the information provided to the image generation AI may include text data such as "60-year-old man wearing a blue shirt" and "wearing glasses, short hair, thick eyebrows, slightly hunched back." The information provided to the image generation AI may also include image information.

[0025] The image generation unit 22 may generate a generated image by combining an image of the clothes worn on the day with the provided image of the search target, like a montage photo, without using an image generation AI. The image generation unit 22 may generate a generated image by combining an image of the hairstyle or glasses worn on the day with the provided image of the search target. In other words, the image generation unit 22 may generate an image by replacing part of the facial photograph of the search target provided.

[0026] <Distribution Department> The distribution unit 23 distributes the search target information and generated images to one or more imaging devices 30. Specifically, the distribution unit 23 distributes the search target information and generated images to imaging devices 30 such as drive recorders in each vehicle, security cameras, and live cameras in various locations. The imaging devices 30 to which the information is distributed are preferably selected in accordance with the wishes of the family, taking privacy into consideration. Furthermore, the distribution unit 23 may be limited to distributing only to vehicles under the jurisdiction of management companies such as taxi companies and route bus companies, excluding general vehicles. Such companies may distribute only to their own vehicles.

[0027] <Control unit, memory unit, communication unit> The search device 20 further includes a control unit 26, a storage unit 27, and a communication unit 28. The control unit 26 includes a processing unit 26a such as a processor, a ROM (Read Only Memory) 26b, and a RAM (Random Access Memory) 26c. The processing unit 26a loads programs stored in the storage unit 27, the ROM 26b, a storage medium, etc. into a work memory area on the RAM 26c and executes them. The processing unit 26a also drives and controls each unit connected via a bus 29.

[0028] The ROM 26b stores control programs and various parameters for controlling the search device 20. The RAM 26c temporarily holds loaded programs and various data, and also provides a work area used by the control unit 26 to perform various processes. The search device 20 includes the above-mentioned information acquisition unit 21, image generation unit 22, and distribution unit 23 as functional blocks realized by programs processed by the processing unit 26a. For convenience, these components are shown as being connected via a bus 29 so as to be able to communicate with each other as appropriate.

[0029] The storage unit 27 is, for example, a storage device such as a hard disk drive, a solid state drive, or a flash memory. The storage unit 27 may store programs executed by the control unit 26, data required for program execution, an operating system, etc. These programs, etc. are read out by the control unit 26 as needed, transferred to the RAM 26c, and then read out and executed by the processing unit 26a.

[0030] The communication unit 28 controls communication with the network NW, etc. The bus 29 is a path for transmitting and receiving control signals, data signals, etc. between the various units.

[0031] <Imaging device> The imaging device 30 includes a drive recorder mounted on a vehicle, a security camera in a store, and a live camera or a monitoring camera installed on a road or in various facilities. The imaging device 30 may be mounted on each of a plurality of vehicles. The imaging device 30 may be mounted on each of a plurality of stores. The imaging device 30 may be mounted on each of a plurality of locations on a road or in various facilities.

[0032] 3 is a block diagram illustrating an imaging device 30 according to embodiment 1. As shown in FIG. 3, the imaging device 30 includes a generated image receiving unit 31, an imaging unit 32, a search candidate extraction unit 33, an image similarity determination unit 34, and a distribution unit 35. The generated image receiving unit 31 receives a generated image from the search device 20.

[0033] The imaging unit 32 captures an image of an object to be imaged. If the imaging device 30 is a drive recorder, the imaging unit 32 captures an image of the area around the vehicle as the object to be imaged. If the imaging device 30 is a security camera, the imaging unit 32 captures an image of the inside of a store as the object to be imaged. If the imaging device 30 is a live camera, the imaging unit 32 captures an image of the road or the inside and outside of various facilities as the object to be imaged. The image captured by the imaging unit 32 is called a "captured image."

[0034] The search candidate extraction unit 33 extracts candidates for the search target from the captured image. A candidate for the search target captured in the captured image is called a "search candidate." For example, if the search target is a person, the search candidate extraction unit 33 extracts the person from the captured image as a search candidate. If the search target is a pet, the search candidate extraction unit 33 extracts the pet from the captured image as a search candidate. The search candidate extraction unit 33 may include, for example, face recognition software and motion recognition software.

[0035] The image similarity determination unit 34 determines the similarity between a search candidate in the captured image and the search target in the generated image. If the similarity is equal to or greater than a predetermined value, the image similarity determination unit 34 determines the captured image in which the search candidate is captured as a "candidate image." Note that the image similarity determination unit 34 in the imaging device 30 may have a similarity determination function equivalent to that of the image similarity determination unit 25 in the search device 20a described below, or may have a similarity determination function simpler than that of the image similarity determination unit 25.

[0036] The distribution unit 35 distributes, as candidate images, captured images of search candidates whose similarity is determined to be equal to or greater than a predetermined value by the image similarity determination unit 34 to the search device 20. In this way, when the imaging device 30 recognizes a person or pet in the captured video that resembles the search target in the generated image, it automatically transmits candidate images of the similar person or pet to the search device 20 together with location information of the shooting location.

[0037] The search device 20 may compare and determine the search candidate in the candidate image with the search target in the search target image. When the search device 20 determines that the search candidate in the candidate image received from the imaging device 30 is the search target, it transmits the candidate image and location information to the search requester and the administrative agency.

[0038] The imaging device 30 further includes a control unit 36, a storage unit 37, a communication unit 38, and a bus 39. The control unit 36 ​​includes a processing unit 36a, a ROM 36b, and a RAM 36c. The functions of the control unit 36, the storage unit 37, the communication unit 38, the bus 39, the processing unit 36a, the ROM 36b, and the RAM 36c in the imaging device 30 are similar to the functions of the control unit 26, the storage unit 27, the communication unit 28, the bus 29, the processing unit 26a, the ROM 26b, and the RAM 26c.

[0039] Although the imaging device 30 is described as determining the similarity between a search candidate extracted from a captured image and the search target in the generated image and transmitting candidate images having a similarity equal to or greater than a predetermined value to the search device 20, this is not limited thereto. For example, the imaging device 30 may not include the generated image receiving unit 31, the search candidate extraction unit 33, and the image similarity determination unit 34. The imaging device 30 may transmit the captured images captured by the imaging unit 32 to the search device 20 as real-time video or recorded video. Then, the search device 20 may extract a search candidate from the captured image and determine the similarity between the search candidate and the search target in the generated image.

[0040] Furthermore, imaging device 30 does not necessarily have to include search candidate extraction unit 33 and image similarity determination unit 34. Imaging device 30 may display the generated images received by generated image receiving unit 31 on the display unit of a car navigation system, the display of a PC in a security room, etc. Then, the driver and security guard may identify captured images of people and pets similar to the search target in the generated images as candidate images, and cause distribution unit 35 to distribute the identified candidate images.

[0041] <Search method> Next, a search method using the search device 20 of this embodiment will be described. Fig. 4 is a flowchart illustrating a search method using the search device 20 according to embodiment 1. As shown in step S11 of Fig. 4, the control unit 26 causes the information acquisition unit 21 to acquire search target information.

[0042] For example, when a missing person such as a wandering elderly person occurs, a search requester such as a family member, guardian, or facility official contacts an administrative agency such as a city hall, ward office, or police from the terminal 10. Patterns in which the search requester contacts from the terminal 10 may include, for example, the following two contact patterns.

[0043] <Contact Pattern A> The age, physical characteristics, clothing characteristics, and facial photograph of the person being searched for <Contact Pattern B> The age, physical characteristics, and clothing of the person being searched for

[0044] The information acquisition unit 21 acquires this search target information as feature information.

[0045] Next, as shown in step S12, the control unit 26 causes the image generation unit 22 to generate an image of the anticipated search target as a generated image from the acquired search target information. For example, the image generation unit 22 may generate the following three image patterns from the above two contact patterns:

[0046] <Image pattern A1> A generative image is generated by a generative AI based on the facial, body, clothing, hairstyle, and other features of the face photo. In this case, images of the body, clothing, hairstyle, etc. may also be generated. <Image pattern A2> A generated image is generated by combining the face in the face photograph with images of the body features, clothing, hairstyle, etc. on the day in a montage-like manner. In this case, images of the body, clothing, hairstyle, etc. may also be generated. <Image pattern B> A generative image is generated using AI based on the search target's age, facial and physique, clothing, hairstyle, and other characteristics.

[0047] The image generating unit 22 may have a function of generating a generated image from the feature information acquired by the information acquiring unit 21 using an image dictionary learned by machine learning.

[0048] Next, as shown in step S13, the control unit 26 causes the distribution unit 23 to distribute the search target information and the generated image to one or more imaging devices 30. The imaging devices 30, such as drive recorders, security cameras, and live cameras registered with administrative agencies, are connected to the search device 20 in a state where information can be transmitted. Thus, the imaging devices 30 receive the search target information and the generated image from the search device 20.

[0049] 5 is a flowchart illustrating a search method using the imaging device 30 according to the embodiment. As shown in step S21 of FIG. 5, the control unit 36 ​​causes the generated image receiving unit 31 to receive the generated image.

[0050] Next, as shown in step S22, the control unit 36 ​​causes the imaging unit 32 to capture an image. Note that, although step S21 is performed before step S22 in the figure, this is not limiting, and step S22 may be performed before step S21, or steps S21 and S22 may be performed in parallel. Next, as shown in step S23, the control unit 36 ​​causes the search candidate extraction unit 33 to extract search candidates from the captured image.

[0051] Next, as shown in step S24, the control unit 36 ​​causes the image similarity determination unit 34 to determine whether the similarity between the search candidate extracted from the captured image and the search target in the generated image is equal to or greater than a predetermined value. If the similarity is equal to or greater than the predetermined value (YES) in step S24, the control unit 36 ​​causes the image similarity determination unit 34 to determine that the captured image is a candidate image. Then, as shown in step S25, the control unit 36 ​​causes the distribution unit 35 to distribute the candidate image to the search device 20. Then, the process proceeds to step S26. On the other hand, if the similarity is less than the predetermined value (NO) in step S24, the process proceeds to step S26.

[0052] Next, as shown in step S26, the control unit 36 ​​determines whether to end the process. If the answer is NO in step S26, that is, not to end the process, the process returns to step S22 and steps S22 to S25 are repeated. On the other hand, if the answer is YES in step S26, that is, to end the process, the process ends.

[0053] Next, the effects of this embodiment will be described. The search device 20 of this embodiment generates a generated image based on the acquired search target information and delivers it to the imaging device 30. The generated image may be generated using image generation AI, taking into consideration the age, gender, facial and body features, and the clothing worn on the day, or may be created by changing the clothing of the search target in a montage-like manner on a provided facial photograph. In this way, the search device 20 delivers generated images in various variations to the imaging device 30, such as a drive recorder, of each vehicle, allowing the search target to be found in a short time.

[0054] Previous related technologies involved government agencies such as city halls using speakers in the neighborhood to notify people of the characteristics of wandering elderly people, and then people visually identify them. However, while the characteristics of wandering elderly people are input into the human brain, memories fade over time, reducing the likelihood of identifying them. Furthermore, because the detection is done by human eyes, there is a chance of misidentification. Thus, previous related technologies were limited to the range of information that could be recognized by listening to information about wandering elderly people broadcast from speakers in the neighborhood, and only people outdoors encountered wandering elderly people.

[0055] In contrast, in this embodiment, an image of the person being searched for is created using generative AI technology and transmitted to the image capture device 30 installed in many vehicles equipped with dashcams and those moving at high speeds, allowing the search to be carried out. In other words, the number of eyes available to find the person being searched for can be increased, and the search area can be expanded. Therefore, the time required to find the person being searched for can be shortened. This, for example, can minimize the risk to the life of the person being searched for. Furthermore, by installing the search device 20 in an administrative agency such as the police and registering the image capture device 30, such as a dashcam, with the administrative agency, the administrative agency and the images captured by the image capture device 30 can be connected via a communication line, allowing recorded images from many vehicles to be collected at the administrative agency. This can improve the efficiency of searches by administrative agencies.

[0056] In the search device 20 of this embodiment, the image generation unit 22 can generate generated images that resemble the current appearance of the search target, including numerous variations in facial expressions, postures, etc., from old photographs of the search target or front-facing ID photographs. This makes it easier to find the search target. Furthermore, the image generation unit 22 can generate generated images using text data by using an image dictionary trained by machine learning on text data. Furthermore, the image generation unit 22 can generate generated images by combining a provided image of the search target with an image of the clothing the person was wearing on the day, like a photomontage. This allows the search device 20 to generate generated images that include numerous variations. In this way, it is easier to find the search target.

[0057] [Embodiment 2] Next, a search device according to a second embodiment will be described. In this embodiment, when information on a candidate image is provided from an imaging device 30 such as a drive recorder, the search device uses image recognition technology to calculate the similarity between the search target image provided by the family of the search target or the like and the candidate image provided by the imaging device 30. If the calculated similarity is equal to or greater than a predetermined value, the search device inputs the provided candidate image as new additional information into the image generation AI and regenerates a generated image. This allows the search device to generate a generated image that is even closer to the search target. The search device also repeats this information update.

[0058] Fig. 6 is a block diagram illustrating a search device 20a according to embodiment 2. As shown in Fig. 6, the search device 20a includes a candidate image receiving unit 24 and an image similarity determining unit 25 in addition to the configuration of the search device 20. In the search device 20a of this embodiment, the search target information includes at least a search target image.

[0059] The candidate image receiving unit 24 receives, from the imaging device 30, a candidate image obtained by capturing an image of a search candidate that is a candidate for a search target.

[0060] The image similarity determination unit 25 determines the similarity between the search target in the search target image and the search candidate in the candidate image. For example, the image similarity determination unit 25 calculates the similarity by comparing the search target in a search target image provided by a family member, which is a photograph taken in the past, with the search candidate in the candidate image transmitted from each drive recorder, security camera, or live camera. The image similarity determination unit 25 may determine the similarity between the search target in the search target image and the search candidate in the candidate image by using a known image recognition technique such as comparing feature points, as disclosed in Non-Patent Document 1.

[0061] When the similarity is equal to or greater than a predetermined value, the image generation unit 22 generates a new generated image based on the search target image and the candidate image. The distribution unit 23 distributes the new generated image generated by the image generation unit 22.

[0062] Next, a search method using the search device 20a according to this embodiment will be described. Fig. 7 is a flowchart illustrating a search method using the search device 20a according to embodiment 2. Steps S11 to S13 in Fig. 7 are the same as steps S11 to S13 in embodiment 1 described above.

[0063] Next, as shown in step S14, it is determined whether to terminate. If the answer is YES, that is, to terminate, the processing is terminated. On the other hand, if the answer is NO, that is, to not terminate, in step S14, the candidate image receiving unit 24 determines whether a candidate image capturing a search candidate that is a candidate for the search target has been received from the imaging device 30, as shown in step S15. If the imaging device 30 extracts a search candidate similar to the search target in the generated image from the captured image, it automatically transmits the captured image to the search device 20a as a candidate image.

[0064] If the candidate image receiving unit 24 has not received the candidate image (NO) in step S15, the process returns to step S14. On the other hand, if the candidate image receiving unit 24 has received the candidate image (YES) in step S15, the image similarity determination unit 25 determines whether the similarity between the search target in the search target image and the search candidate in the candidate image is equal to or greater than a threshold value, as shown in step S16.

[0065] If step S16 determines that the similarity between the search target in the search target image and the search candidate in the candidate image is less than a predetermined value (NO), the process returns to step S14. On the other hand, if step S16 determines that the similarity between the search target in the search target image and the search candidate in the candidate image is greater than or equal to a predetermined value (YES), the image generation unit 22 generates a new generated image based on the search target image and the candidate image, as shown in step S17. Then, the distribution unit 23 distributes the newly generated generated image to the imaging device 30, as shown in step S18. The process then returns to step S14. In this manner, the search target can be searched for.

[0066] According to this embodiment, the search device 20a is equipped with a candidate image receiving unit 24 and an image similarity determining unit 25, and is therefore able to determine whether a search candidate in a candidate image transmitted from an image capturing device 30 is similar to the search target. This allows the search to be narrowed down to information from image capturing devices 30 that have transmitted candidate images with a similarity equal to or greater than a predetermined value, thereby narrowing down to information with a high degree of certainty. This reduces the search effort.

[0067] If the similarity between a candidate image transmitted from imaging device 30 and the search target image is equal to or greater than a predetermined value, this indicates that the candidate image is likely to be an image of the "current appearance" of the missing person or other person being searched for. Therefore, image generation unit 22 generates a new generated image based on the search target image and the candidate image with a similarity equal to or greater than a predetermined value, and distribution unit 23 distributes the newly generated generated image to imaging device 30. This allows the generated image to be closer to the "current appearance," making it easier to find the search target. Other configurations and effects are included in the description of embodiment 1.

[0068] The present disclosure is not limited to the descriptions of the above-mentioned first and second embodiments, and can be modified as appropriate without departing from the spirit of the present disclosure. For example, a combination of any of the configurations of the first and second embodiments is also within the scope of the technical idea of ​​the present embodiments. Furthermore, a search system including a search device and an imaging device, and a search program that causes a computer to execute a search method, as shown below, are also within the scope of the technical idea of ​​the embodiments.

[0069] (Appendix 1) Search equipment, one or more imaging devices; Equipped with The search device is an information acquisition unit that acquires search target information from information related to a search target; an image generation unit that generates an image of the search target as a generated image from the search target information; a distribution unit that distributes the search target information and the generated image to one or more imaging devices; and The imaging device is a generated image receiving unit that receives the generated image from the search device; an imaging unit that captures an image of an object; a search candidate extraction unit that extracts search candidates that are candidates for the search target from the captured image captured by the imaging unit; an image similarity determination unit that determines a similarity between the search candidate in the captured image and the search target in the generated image; a distribution unit that distributes, to the search device, a captured image of the search candidate whose similarity is determined to be equal to or greater than a predetermined value by the image similarity determination unit, as a candidate image; having Search system. (Appendix 2) a person information acquisition unit that acquires search target information of information relating to a search target; an image generation unit that generates an image of the search target as a generated image from the search target information; a distribution unit that distributes the search target information and the generated image to one or more imaging devices; A search program using a search device having causing the information acquisition unit to acquire the search target information; a step of causing the image generation unit to generate the image of the anticipated search target as the generated image from the search target information; causing the distribution unit to distribute the search target information and the generated image to one or more of the imaging devices; A search program that causes a computer to execute the following.

[0070] Furthermore, the above-described search program includes a set of instructions (or software code) that, when loaded into a computer, causes the computer to perform one or more functions described in the embodiments. The program may be stored in a non-transitory computer-readable medium or a tangible storage medium. By way of example and not limitation, computer-readable media or tangible storage media include random-access memory (RAM), read-only memory (ROM), flash memory, solid-state drive (SSD) or other memory technologies, CD-ROM, digital versatile disc (DVD), Blu-ray (registered trademark) disc or other optical disk storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage device. The program may also be transmitted on a transitory computer-readable medium or communication medium. By way of example and not limitation, transitory computer-readable media or communication media include electrical, optical, acoustic, or other forms of propagated signals. [Explanation of symbols]

[0071] 1. Search System 10, 10a, 10b terminals 20, 20a Search Device 21 Information Acquisition Department 22 Image generation unit 23 Distribution Department 24 Candidate image receiving unit 25 Image similarity determination unit 26 Control Unit 26a Processing section 26b ROM 26c RAM 27 Memory section 28 Communications Department 29 Bus 30, 30a, 30b Imaging device 31 Generated image receiving unit 32 Imaging unit 33 Search candidate extraction unit 34 Image similarity determination unit 35 Distribution Department 36 Control Unit 36a Processing section 36b ROM 36c RAM 37 Memory section 38 Communications Department 39 Bus NW Network

Claims

1. an information acquisition unit that acquires search target information from information related to a search target; an image generation unit that generates an image of the search target as a generated image from the search target information; a distribution unit that distributes the search target information and the generated image to one or more imaging devices; Equipped with Search equipment.

2. the search target information includes at least characteristic information of information regarding characteristics of the search target; The image generation unit has a function of generating the generated image from the feature information using an image dictionary learned by machine learning. The search device according to claim 1 .

3. The search target information includes at least a search target image obtained by capturing an image of the search target, a candidate image receiving unit that receives, from the imaging device, a candidate image of a search candidate that is a candidate for the search target; an image similarity determination unit that determines a similarity between the search target in the search target image and the search candidate in the candidate image; Furthermore, the image generation unit generates a new generated image based on the search target image and the candidate image when the similarity is equal to or greater than a predetermined value; The search device according to claim 2.

4. an information acquisition unit that acquires search target information from information related to a search target; an image generation unit that generates an image of the search target as a generated image from the search target information; a distribution unit that distributes the search target information and the generated image to one or more imaging devices; a step of causing the information acquisition unit to acquire the search target information in a search device having the above-mentioned a step of causing the image generation unit to generate the image of the anticipated search target as the generated image from the search target information; causing the distribution unit to distribute the search target information and the generated image to one or more of the imaging devices; A search method comprising:

5. Search equipment, one or more imaging devices; Equipped with The search device is an information acquisition unit that acquires search target information from information related to a search target; an image generation unit that generates an image of the search target as a generated image from the search target information; a first distribution unit that distributes the search target information and the generated image to one or more imaging devices; and The imaging device is a generated image receiving unit that receives the generated image from the search device; an imaging unit that captures an image of an object; a search candidate extraction unit that extracts search candidates that are candidates for the search target from the captured image captured by the imaging unit; an image similarity determination unit that determines a similarity between the search candidate in the captured image and the search target in the generated image; a second distribution unit that distributes, to the search device, the captured image of the search candidate whose similarity is determined to be equal to or greater than a predetermined value by the image similarity determination unit, as a candidate image; having Search system.

Images