Route calculation device, route calculation method, and program

Abstract

A route calculation device (10) according to the present invention includes: an acquisition unit (11) that acquires images captured by cameras installed at a plurality of observation points on a passageway; an identification unit (12) that identifies attributes of people included in the images; a trend calculation unit (13) that, on the basis of the attribute identification results, calculates trends regarding the attributes of the people traveling through each of the plurality of observation points; and a route calculation unit (14) that, on the basis of said trends, calculates a travel route suited to a person with a prescribed attribute.

Description

[Technical Field]

[0001] The present invention relates to a route calculation device, a route calculation method, and a recording medium. [Background technology]

[0002] Techniques related to the present invention are disclosed in Patent Documents 1 to 4 and Non-Patent Document 1.

[0003] Patent Document 1 discloses a technology that extracts barrier conditions that serve as criteria for determining whether a path can be passed or not based on the behavioral history of a wheelchair user, and calculates a recommended route for the wheelchair user based on the barrier conditions. The behavioral history data consists of image data of barriers taken by the wheelchair user and location information of the barriers. Barriers are irregularities or steps on the road surface, etc.

[0004] Patent Document 2 discloses a technology that registers how a user feels about multiple objects / events, such as places where many people are hanging out or places where people are making noise, analyzes real-time images to extract locations where the multiple objects / events are occurring, and calculates a route to present to the user based on the extraction results.

[0005] Patent Document 3 discloses a technology for searching for an evacuation route taking into consideration the location of the disaster and the situation of the disaster victims.

[0006] Patent Document 4 discloses a technology that calculates the feature values ​​of each of multiple key points of a human body contained in an image, searches for images containing human bodies with similar postures or movements based on the calculated feature values, and classifies images with similar postures or movements together.

[0007] Non-Patent Document 1 discloses a technique related to human skeleton estimation. [Prior art documents] [Patent documents]

[0008] [Patent Document 1] Japanese Patent Application Publication No. 2019-124587 [Patent Document 2] Japanese Patent Application Publication No. 2017-181353 [Patent Document 3] Japanese Patent Application Laid-Open No. 2015-129658 [Patent Document 4] International Publication No. 2021 / 084677 [Non-patent literature]

[0009] [Non-Patent Document 1] Zhe Cao, Tomas Simon, Shih-En Wei, Yaser Sheikh, "Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields", The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, P. 7291-7299 Summary of the Invention [Problem to be solved by the invention]

[0010] There is a demand for technology that does not simply calculate the shortest route, but also calculates routes suitable for people with various attributes, such as a safe route suitable for women, a route suitable for people using crutches, or a route suitable for people walking while pulling a suitcase.

[0011] The technology disclosed in Patent Document 1 can calculate a route suitable for a wheelchair user, but cannot calculate a route suitable for people with other attributes.

[0012] The technology disclosed in Patent Document 2 analyzes images to extract locations where multiple objects / events are currently occurring. This technology requires analyzing a huge amount of images taken of various locations in real time. As a result, the processing load on the computer becomes heavy.

[0013] The technology disclosed in Patent Document 3 can calculate a route suitable for people evacuating, but cannot calculate a route suitable for people with other attributes.

[0014] Patent Document 4 and Non-Patent Document 1 are technologies for estimating the posture and movement of a person, and are not technologies for calculating a travel route.

[0015] In view of the above-mentioned problems, an example of an object of the present invention is to provide a route calculation device, a route calculation method, and a recording medium that solve the problem of calculating travel routes suitable for people with various attributes. [Means for solving the problem]

[0016] According to one aspect of the present invention, an acquisition means for acquiring images captured by cameras installed at each of a plurality of observation points on the passage; An identification means for identifying attributes of passersby included in the image; a tendency calculation means for calculating a tendency of attributes of pedestrians passing through each of the plurality of observation points based on the result of the attribute identification; a route calculation means for calculating a travel route suitable for a pedestrian having the predetermined attribute based on the tendency; A route calculation device is provided, comprising:

[0017] According to one aspect of the present invention, The computer Images taken by cameras installed at multiple observation points along the corridor are acquired, Identifying attributes of passersby included in the image; Calculating the attribute trends of pedestrians passing through each of the plurality of observation points based on the attribute identification results; calculating a travel route suitable for a pedestrian having the predetermined attribute based on the tendency; A route calculation method is provided.

[0018] According to one aspect of the present invention, Computer, an acquisition means for acquiring images captured by cameras installed at each of a plurality of observation points on the passage; an identification means for identifying attributes of passersby included in the image; a tendency calculation means for calculating a tendency of attributes of pedestrians passing through each of the plurality of observation points based on the result of the identification of the attributes; a route calculation means for calculating a travel route suitable for a pedestrian having the predetermined attribute based on the tendency; A recording medium is provided on which a program for causing the device to function as the above is recorded. [Effects of the Invention]

[0019] According to one aspect of the present invention, a route calculation device, a route calculation method, and a recording medium are realized that solve the problem of calculating travel routes suitable for people with various attributes. [Brief explanation of the drawings]

[0020] The above and other objects, features and advantages are described below. Suitable This will become more apparent from the following embodiments and the accompanying drawings.

[0021] [Figure 1] FIG. 2 is a diagram illustrating an example of a functional block diagram of a route calculation device. [Figure 2] FIG. 2 is a diagram illustrating an example of a hardware configuration of a route calculation device. [Figure 3] FIG. 10 is a diagram for explaining the processing of an identification unit. [Figure 4] FIG. 2 is a diagram illustrating an example of information processed by the route calculation device. [Figure 5] FIG. 4 is a diagram illustrating an example of information output by the route calculation device. [Figure 6] 10 is a flowchart illustrating an example of a processing flow of the route calculation device. [Figure 7] 10 is a flowchart showing another example of the processing flow of the route calculation device. [Figure 8]FIG. 2 is a diagram illustrating an example of information processed by the route calculation device. [Figure 9] FIG. 2 is a diagram illustrating an example of a functional block diagram of a route calculation device. [Figure 10] FIG. 10 is a diagram illustrating an example of information that the route calculation device causes an output device to output. DETAILED DESCRIPTION OF THE INVENTION

[0022] Hereinafter, embodiments of the present invention will be described with reference to the drawings. In all the drawings, like components are designated by like reference numerals, and the description thereof will be omitted as appropriate.

[0023] First Embodiment 1 is a functional block diagram showing an overview of a route calculation device 10 according to the first embodiment. The route calculation device 10 includes an acquisition unit 11, an identification unit 12, a tendency calculation unit 13, a route calculation unit 14, and a storage unit 15. Note that the route calculation device 10 does not necessarily have to include the storage unit 15. In this case, an external device configured to be able to communicate with the route calculation device 10 includes the storage unit 15.

[0024] The acquisition unit 11 acquires images captured by cameras installed at each of a plurality of observation points on the passageway. The identification unit 12 identifies the attributes of pedestrians included in the images. The tendency calculation unit 13 calculates the attribute tendencies of pedestrians passing through each of the plurality of observation points based on the attribute identification results by the identification unit 12, and stores the calculated tendencies in the memory unit 15. The route calculation unit 14 calculates a travel route suitable for pedestrians with predetermined attributes based on the attribute tendencies of pedestrians passing through each of the plurality of observation points.

[0025] The route calculation device 10 having such a configuration solves the problem of calculating travel routes suitable for people with various attributes.

[0026] <Second embodiment> "overview" The route calculation device 10 of the second embodiment is a more specific embodiment of the route calculation device 10 of the first embodiment. The route calculation device 10 analyzes images taken by cameras installed at each of a plurality of observation points on a passageway and calculates the tendency of attributes of pedestrians passing through each of the plurality of observation points. The route calculation device 10 then calculates a travel route suitable for pedestrians of each attribute, which includes observation points frequently passed by pedestrians of each attribute, but does not include observation points not passed by or infrequently passed by pedestrians of each attribute. This will be described in detail below.

[0027] "Hardware Configuration" Next, an example of the hardware configuration of the route calculation device 10 will be described. Each functional unit of the route calculation device 10 is realized by any combination of hardware and software, centered around a CPU (Central Processing Unit) of any computer, memory, programs loaded into the memory, a storage unit such as a hard disk that stores the programs (this can store programs that are pre-loaded when the device is shipped, as well as programs downloaded from recording media such as CDs (Compact Discs) or servers on the Internet), and a network connection interface. Those skilled in the art will understand that there are many variations in the implementation methods and devices.

[0028] FIG. 2 is a block diagram illustrating an example of the hardware configuration of a route calculation device 10. As shown in FIG. 2, the route calculation device 10 has a processor 1A, a memory 2A, an input / output interface 3A, a peripheral circuit 4A, and a bus 5A. The peripheral circuit 4A includes various modules. The route calculation device 10 does not necessarily have to have the peripheral circuit 4A. Note that the route calculation device 10 may be composed of multiple devices that are physically and / or logically separated. In this case, each of the multiple devices can have the above hardware configuration.

[0029] The bus 5A is a data transmission path for the processor 1A, memory 2A, peripheral circuit 4A, and input / output interface 3A to transmit and receive data among them. The processor 1A is an arithmetic processing device such as a CPU or a GPU (Graphics Processing Unit). The memory 2A is a memory such as a RAM (Random Access Memory) or a ROM (Read Only Memory). The input / output interface 3A includes an interface for acquiring information from an input device, an external device, an external server, an external sensor, a camera, etc., and an interface for outputting information to an output device, an external device, an external server, etc. Examples of input devices include a keyboard, a mouse, a microphone, physical buttons, a touch panel, etc. Examples of output devices include a display, a speaker, a printer, a mailer, etc. The processor 1A can issue commands to each module and perform calculations based on the results of those calculations.

[0030] "Function Configuration" Next, the functional configuration of the route calculation device 10 according to the second embodiment will be described in detail. FIG. 1 shows an example of a functional block diagram of the route calculation device 10. As shown in the figure, the route calculation device 10 includes an acquisition unit 11, an identification unit 12, a tendency calculation unit 13, a route calculation unit 14, and a storage unit 15. As in the first embodiment, the route calculation device 10 does not necessarily have to include the storage unit 15. In this case, an external device configured to be able to communicate with the route calculation device 10 includes the storage unit 15.

[0031] The route calculation device 10 executes a data preparation step of preparing data necessary for route calculation, and a route calculation step of calculating a route based on the data prepared in the preparation step, as will be described in detail below.

[0032] -Data preparation process- The data preparation step is executed before the route calculation step, and is realized by the acquisition unit 11, the identification unit 12, and the tendency calculation unit 13.

[0033] The acquisition unit 11 acquires images captured by cameras installed at each of a plurality of observation points on the passage. The images include at least one of moving images and still images. The cameras are, for example, surveillance cameras. The cameras may capture moving images or still images at predetermined times (such as when a person is detected in the capture area). The cameras are installed in positions and orientations that allow them to capture images of pedestrians passing through the passage.

[0034] There are no particular restrictions on how to define the observation points. By setting up cameras at more observation points and calculating the trends in the attributes of passersby at more observation points, it becomes possible to accurately calculate travel routes suitable for passersby with specified attributes.

[0035] The acquisition unit 11 acquires images captured over a certain period of time (for example, several months, one year, or several years). The acquisition unit 11 may acquire images captured over a certain period of time all at once, or may acquire images sequentially as they are captured.

[0036] The route calculation device 10 may be configured to be able to communicate with a camera. The acquisition unit 11 may then acquire images transmitted by the camera. Alternatively, images captured by the camera may be stored in an arbitrary storage device. The acquisition unit 11 may then acquire the images stored in the storage device by any means.

[0037] In the embodiments, "acquisition" includes at least one of the following: a device going to retrieve data or information stored in another device or storage medium (active acquisition), and inputting data or information output from another device into the device (passive acquisition). Examples of active acquisition include making a request or inquiry to another device and receiving a reply, and accessing and reading information from another device or storage medium. An example of passive acquisition is receiving information that is distributed (or transmitted, pushed, etc.). Furthermore, "acquisition" may also mean selecting and acquiring data or information from received data or information, or selecting and receiving distributed data or information.

[0038] The identification unit 12 identifies attributes of pedestrians included in the images acquired by the acquisition unit 11. The identified attributes of pedestrians may affect the selection of a travel route. For example, the attributes of pedestrians include at least one of gender, age group, physique, belongings, means of transportation, and whether or not the pedestrian is accompanied.

[0039] The age groups may be regularly categorized at predetermined intervals, such as under 10 years old, teenagers, people in their 20s, and people in their 30s. Alternatively, the age groups may be categorized by ages that are considered to have similar travel routes. Examples of such categorized age groups include, but are not limited to, under 13 years old, 14 to under 18 years old, 18 to under 40 years old, 40 to under 65 years old, and 65 years old and over, taking into account factors such as physical strength.

[0040] Body type is defined to include classifications that can be identified through image analysis and that can affect route selection. For example, body type may be defined to include lean, normal, obese, etc.

[0041] The identified belongings include items that can be identified by image analysis and that may affect the selection of a travel route. For example, the identified belongings may include at least some of the following: clothing style (casual, formal), clothing type (jeans, short skirt, long skirt, etc.), shoe type (sandals, sneakers, boots, high heels, business shoes, etc.), accessories (hat, sunglasses, headphones, etc.), and objects held in hand (umbrella, bag, suitcase, stroller, crutches, white cane, etc.).

[0042] Transportation modes are defined to include classifications that can be identified through image analysis and that can influence route selection, such as walking, running, bicycling, stroller use, wheelchair use, white cane use, crutch use, etc.

[0043] The presence or absence of a companion may be determined based on the distance from another person, the length of time that the distance is maintained at or below a threshold, etc. For example, a person whose distance from another person is maintained at or below a threshold for a predetermined period of time or more may be determined to be the companion of that person. Note that the algorithm for determining the presence or absence of a companion through image analysis is not limited to this.

[0044] The identification unit 12 identifies attributes of passersby included in the image based on the results of analyzing the image. The image analysis is performed by a pre-prepared image analysis system 20. As shown in FIG. 3 , the identification unit 12 inputs the image to the image analysis system 20. The identification unit 12 then obtains the image analysis results from the image analysis system 20. The image analysis system 20 may be a part of the route calculation device 10, or may be an external device that is physically and / or logically independent from the route calculation device 10.

[0045] Here, we will explain the image analysis system 20. The image analysis system 20 has at least one of a face recognition function, a human shape recognition function, a posture recognition function, a movement recognition function, an appearance attribute recognition function, an image gradient feature detection function, an image color feature detection function, an object recognition function, a character recognition function, and a gaze detection function.

[0046] The face recognition function extracts facial features of a person. It may also compare and calculate the similarity between facial features (to determine whether they are the same person, etc.). It may also compare the extracted facial features with those of people pre-registered in a database to determine whether the person in the image is a person registered in the database.

[0047] The human recognition function extracts a person's physical features (e.g., overall features such as physique, height, clothing, etc.). Furthermore, the similarity between the physical features may be compared and calculated (e.g., to determine whether they are the same person). The extracted physical features may also be compared with those of people pre-registered in a database to determine whether the person in the image is a person registered in the database.

[0048] The posture recognition function and movement recognition function detect the joint points of a person and connect the joint points to create a stick figure model. Then, based on the stick figure model, the person is detected, the person's height is estimated, posture features are extracted, and movement is identified based on changes in posture. For example, the postures and movements of people moving by each of the above-mentioned means of transportation are defined in advance, and these postures and movements are detected. Furthermore, similarities between posture features and movement features may be compared and calculated (e.g., determining whether the postures or movements are the same). The estimated height may also be compared with the heights of people registered in a database in advance to identify whether the person in the image is registered in the database. The posture recognition function and movement recognition function may be realized by the technology disclosed in Patent Document 4 and Non-Patent Document 1 above.

[0049] The appearance attribute recognition function recognizes appearance attributes associated with a person (e.g., clothing style, type of clothing, type of shoes, accessories, objects held in hands, etc.). For example, the items to be identified as described above are defined in advance, and these items are detected through image analysis. Furthermore, the similarity of the recognized appearance attributes may be compared and calculated (making it possible to determine whether the attributes are the same). The recognized appearance attributes may also be compared with the appearance attributes of people pre-registered in a database to determine whether the person in the image is a person registered in the database.

[0050] Image gradient feature detection functions include SIFT, SURF, RIFF, ORB, BRISK, CARD, HOG, etc. According to these functions, the gradient features of each frame image are detected.

[0051] The image color feature detection function generates data indicating the color features of the image, such as a color histogram, and detects the color features of each frame image.

[0052] The object recognition function is realized using an engine such as YOLO (which can extract general objects (e.g., tools and equipment used in sports and other performances) and people). By using the object recognition function, various predefined objects can be detected from an image. Specifically, objects related to the above-mentioned possessions and means of transportation may be predefined and detected.

[0053] The character recognition function recognizes numbers, letters, etc.

[0054] The gaze detection function detects the gaze direction of people in the image.

[0055] Returning to FIG. 1, the tendency calculation unit 13 calculates the tendency of the attributes of pedestrians passing through each of the plurality of observation points, based on the attribute identification results by the identification unit 12.

[0056] For each observation point, the tendency calculation unit 13 identifies the attributes of pedestrians who frequently pass through the observation point. For example, the tendency calculation unit 13 may identify the attributes of pedestrians who have passed through the observation point at least once within a predetermined period as the attributes of pedestrians who frequently pass through the observation point.

[0057] Alternatively, the tendency calculation unit 13 may identify the attributes of pedestrians who have passed the observation point a reference number of times or more within a predetermined period as the attributes of pedestrians who frequently pass through the observation point. The absolute number of pedestrians for each attribute may differ. For example, the number of wheelchair users and white cane users is overwhelmingly smaller than that of pedestrians on foot. Therefore, the reference number may differ for each attribute. That is, in the case of wheelchair users, if they pass the observation point a first reference number of times (e.g., 5 times) or more within a predetermined period, it may be determined that wheelchair users frequently pass through the observation point. In the case of pedestrians, if they pass the observation point a second reference number of times (e.g., 100 times) or more within a predetermined period, it may be determined that pedestrians frequently pass through the observation point. Furthermore, the traffic volume differs for each observation point. Therefore, the reference number may differ for each observation point.

[0058] The tendency calculation unit 13 stores the calculation results in the storage unit 15. Fig. 4 shows a schematic example of the calculation results stored in the storage unit 15. In Fig. 4, the observation point identification information, the position information, and the passerby attribute tendency are linked to each other.

[0059] The observation point identification information is information for identifying a plurality of observation points from one another.

[0060] The location information indicates the location of each of the multiple observation points. Examples of the location information include latitude and longitude information and addresses.

[0061] The passerby attribute tendency indicates the calculation result of the tendency calculation unit 13. That is, the passerby attribute tendency indicates the attributes of passersby who are determined to frequently pass through each of the plurality of observation points.

[0062] -Route calculation process- The route calculation step is executed after the data preparation step and is realized by the route calculation unit 14.

[0063] The route calculation unit 14 calculates a travel route suitable for pedestrians with predetermined attributes based on the tendencies of the attributes of pedestrians passing through each of the plurality of observation points calculated by the tendency calculation unit 13.

[0064] The predetermined attribute may be specified by the user each time a search is performed. That is, the route calculation unit 14 may receive a user input specifying the attribute each time a search is performed. Then, the route calculation unit 14 may calculate a travel route suitable for a pedestrian having the attribute specified by the user input.

[0065] Alternatively, a predetermined attribute may be specified as a default in advance. Then, the route calculation unit 14 may calculate a travel route suitable for pedestrians with the attribute specified as a default in advance. In this case, multiple patterns may be specified as a default in advance. That is, a first attribute may be specified as a first pattern, and a second attribute may be specified as a second pattern. In this case, the route calculation unit 14 calculates, for each pattern, a travel route suitable for pedestrians with the attribute specified in each pattern. Note that the default values ​​may be editable (changed, added, deleted, etc.) by the user.

[0066] Furthermore, as the attribute of a pedestrian with a predetermined attribute, one attribute may be specified (e.g., male), or multiple attributes may be specified (e.g., male, in his 40s, walking). When multiple attributes are specified, the route calculation unit 14 may calculate a travel route suitable for a pedestrian who has all of the specified multiple attributes. In this case, for example, when the attributes "male," "in his 40s," and "walking" are specified, the route calculation unit 14 calculates a travel route suitable for a male in his 40s who travels on foot.

[0067] In addition, when multiple attributes are specified, the route calculation unit 14 may calculate travel routes suitable for multiple pedestrians each having one of the specified multiple attributes. In this case, for example, when the attributes "male," "forty-somethings," and "walking" are specified, the route calculation unit 14 calculates a travel route suitable for males, a travel route suitable for pedestrians in their forties, and a travel route suitable for pedestrians traveling on foot.

[0068] A route calculation unit 14 calculates a travel route suitable for a pedestrian with a predetermined attribute, which passes through observation points that are frequently passed by pedestrians with the attribute, and does not pass through observation points that are not frequently passed by pedestrians with the attribute.

[0069] Observation points that are frequently passed by pedestrians with a certain attribute and observation points that are not frequently passed by pedestrians with that attribute are identified based on the attribute trends of pedestrians passing through each of the multiple observation points (see FIG. 4) calculated by the trend calculation unit 13. An observation point that is frequently passed by pedestrians with a certain attribute is an observation point for which that attribute is registered in the pedestrian attribute trend column. An observation point that is not frequently passed by pedestrians with a certain attribute is an observation point for which that attribute is not registered in the pedestrian attribute trend column.

[0070] There are various algorithms for calculating a travel route that passes through observation points that are frequently passed by pedestrians with a certain attribute and does not pass through observation points that are not frequently passed by pedestrians with that attribute, and any method can be adopted. For example, the route calculation unit 14 may determine observation points that are not frequently passed by pedestrians with that attribute as impassable points. Then, the route calculation unit 14 may calculate a route from a start point specified by the user to a finish point specified by the user using any conventional algorithm, under the condition that the route does not pass through impassable points.

[0071] Furthermore, when multiple travel routes (e.g., a predetermined number M1 of travel routes in ascending order of distance) are calculated using the above method, the route calculation unit 14 may evaluate each of the multiple travel routes based on the extent to which the route includes observation points frequently traveled by pedestrians with that attribute. In this case, each of the multiple travel routes is evaluated using a method in which the more observation points frequently traveled by pedestrians with that attribute a travel route includes, the higher the evaluation value. Then, the route calculation unit 14 may calculate a predetermined number M2 (M1>M2) of travel routes in ascending order of evaluation value as travel routes suitable for pedestrians with that attribute.

[0072] The route calculation device 10 outputs the calculation results calculated by the route calculation unit 14. For example, the route calculation device 10 may be a server. The route calculation device 10 may then transmit the calculation results to a client terminal. Examples of the client terminal include, but are not limited to, a smartphone, a smartwatch, a tablet terminal, a mobile phone, and a personal computer.

[0073] 5 shows an example of the calculation results output by the route calculation device 10. In the example shown, a route suitable for male pedestrians and a route suitable for female pedestrians are displayed. The "S" mark in the figure indicates the start point, and the "G" mark indicates the finish point.

[0074] Next, an example of the processing flow of the route calculation device 10 will be described with reference to the flowcharts of FIGS.

[0075] 6 shows an example of the processing flow of the data preparation step. When the route calculation device 10 acquires images taken over a certain period of time by cameras installed at each of a plurality of observation points on the passage (S10), it identifies the attributes of pedestrians at each of the plurality of observation points by identifying the attributes of pedestrians included in the images (S11). Next, based on the identification results of S11, the route calculation device 10 calculates the tendency of the attributes of pedestrians passing through each of the plurality of observation points (S12), and registers the calculation results in the storage unit 15 (S13).

[0076] 7 shows an example of the processing flow of the route calculation step. The route calculation device 10 calculates a travel route suitable for pedestrians with predetermined attributes based on the tendency of attributes of pedestrians passing through each of the multiple observation points generated in the data preparation step (S20). Then, the route calculation device 10 outputs the calculated travel route (S21).

[0077] "Action and effect" The route calculation device 10 of the second embodiment analyzes images taken by cameras installed at each of a plurality of observation points on a passageway and calculates the tendency of attributes of pedestrians passing through each of the plurality of observation points. Then, the route calculation device 10 calculates a travel route that includes points frequently passed by pedestrians of each attribute and does not include points not passed by or infrequently passed by pedestrians of each attribute as a travel route suitable for pedestrians of each attribute.

[0078] According to such a route calculation device 10, it is possible to calculate travel routes suitable for people with various attributes.

[0079] Furthermore, when there are barriers (steps, stairs, unpaved roads, etc.) that can affect the selection of travel routes, whether or not pedestrians with each attribute will travel there is influenced by the local conditions, i.e., whether or not there are detours nearby, the condition of the barrier, etc. Technology that calculates travel routes that uniformly avoid barriers without reflecting the conditions of each local area may result in impractical results, such as calculating a very roundabout travel route.

[0080] Like the route calculation device 10, by calculating the tendency of attributes of pedestrians passing through each observation point based on the actual past travel records in each area, and calculating a travel route suitable for pedestrians with predetermined attributes based on the calculation results, it becomes possible to calculate an optimal travel route that reflects the circumstances of the area.

[0081] Furthermore, the route calculation device 10 can treat at least one of the following as attributes of a pedestrian: gender, age group, physique, belongings, means of transportation, and whether the pedestrian is accompanied. As a result, the route calculation device 10 can classify pedestrians based on various attributes that may affect the selection of a travel route, and calculate a travel route suitable for each pedestrian.

[0082] <Third embodiment> The route calculation device 10 of the third embodiment calculates the tendency of attributes of pedestrians passing through each of a plurality of observation points when each of a plurality of conditions defined based on time of day, day of the week, month, weather, season, etc. is satisfied. Then, the route calculation device 10 calculates a travel route suitable for pedestrians with predetermined attributes under predetermined conditions. This will be described in detail below.

[0083] The identification unit 12 identifies characteristics of the timing at which each image acquired by the acquisition unit 11 was captured. The characteristics may affect the selection of a travel route and may include, for example, at least one of time of day, day of the week, month, weather, and season. The image metadata may include the date and time of capture, day of the week, weather, etc. The identification unit 12 may then identify the characteristics based on the image metadata. Alternatively, the identification unit 12 may identify the date and time of capture by any means, and then access a server that stores past weather information to acquire the weather at the date and time of capture. The identification unit 12 may also identify the season at the date and time of capture based on information indicating the correspondence between the date and season that is registered in advance in the route calculation device 10.

[0084] The tendency calculation unit 13 calculates the tendency of attributes of pedestrians passing through each of the plurality of observation points when each of the plurality of conditions defined by the above characteristics is satisfied. The condition may be defined using one characteristic or multiple characteristics. Examples of conditions include, but are not limited to, "rain," "rain and 8 PM to 6 AM," etc.

[0085] The tendency calculation unit 13 identifies conditions that are satisfied by the timing at which each of the multiple images is captured, based on the characteristics of the timing at which each of the multiple images is captured.The tendency calculation unit 13 then calculates the tendency of the attributes of pedestrians passing through each of the multiple observation points when each condition is satisfied, based on the images captured when each condition is satisfied.The method for calculating the tendency is as described in the second embodiment.

[0086] Fig. 8 shows a schematic example of the calculation results calculated by the tendency calculation unit 13 and stored in the storage unit 15. In Fig. 8, observation point identification information, location information, and passerby attribute tendencies are linked to one another. The passerby attribute tendencies indicate, for each condition, the tendencies in the attributes of passersby who pass through each of the multiple observation points when each condition is satisfied.

[0087] The route calculation unit 14 calculates a travel route suitable for a pedestrian with a predetermined attribute under predetermined conditions.

[0088] The predetermined conditions may be specified by the user each time a search is performed. That is, the route calculation unit 14 may accept user input specifying the conditions each time a search is performed. Then, the route calculation unit 14 may calculate a travel route suitable for pedestrians with predetermined attributes under the conditions specified in the user input. Also, the route calculation unit 14 may accept user input specifying the attributes and conditions each time a search is performed. Then, the route calculation unit 14 may calculate a travel route suitable for pedestrians with the attributes specified in the user input under the conditions specified in the user input.

[0089] Alternatively, predetermined conditions may be specified as defaults in advance. Then, the route calculation unit 14 may calculate a travel route suitable for pedestrians with predetermined attributes under the conditions specified as defaults in advance. In this case, multiple patterns may be specified as defaults in advance. That is, a first condition may be specified as a first pattern, and a second condition may be specified as a second pattern. In this case, the route calculation unit 14 calculates a travel route suitable for pedestrians with predetermined attributes under the conditions of each pattern for each pattern. Note that the default values ​​may be editable (changed, added, deleted, etc.) by the user.

[0090] The predetermined attribute is specified by the method described in the second embodiment.

[0091] The route calculation unit 14 calculates a travel route suitable for a pedestrian with a predetermined attribute under predetermined conditions, which passes through observation points that are frequently passed by pedestrians with the attribute under predetermined conditions and does not pass through observation points that are not frequently passed by pedestrians with the attribute under predetermined conditions.

[0092] Observation points that are frequently passed by pedestrians with a certain attribute under predetermined conditions, and observation points that are not frequently passed by pedestrians with that attribute under predetermined conditions, are identified based on the attribute trends (see FIG. 8) of pedestrians who pass through each of the multiple observation points when each of the multiple conditions is satisfied, calculated by the trend calculation unit 13. For example, an observation point that is frequently passed by pedestrians with a certain attribute when condition 1 is satisfied is an observation point whose attribute is registered in the condition 1 column of the pedestrian attribute trend column. An observation point that is not frequently passed by pedestrians with a certain attribute when condition 1 is satisfied is an observation point whose attribute is not registered in the condition 1 column of the pedestrian attribute trend column.

[0093] The calculation of a travel route that passes through observation points that are frequently traveled by pedestrians with the specified attributes under specified conditions and does not pass through observation points that are not frequently traveled by pedestrians with the specified attributes under specified conditions is achieved using a method similar to that described in the second embodiment.

[0094] Other configurations of the route calculation device 10 of the third embodiment are similar to those of the route calculation device 10 of the first and second embodiments.

[0095] The route calculation device 10 of the third embodiment achieves the same effects as the route calculation device 10 of the first and second embodiments. In addition, characteristics such as time of day, day of the week, month, weather, and season can affect the selection of a travel route. For example, there are routes that are not often used at night but are often used in the morning or during the daytime while it is light. There are also routes that are often used on sunny days but become muddy on rainy days. reason There are passageways that are not used often.

[0096] The route calculation device 10 calculates the attribute tendencies of pedestrians passing through each observation point for each condition defined based on the characteristics described above, and calculates a travel route suitable for pedestrians with specified attributes under specified conditions based on the calculation results, thereby making it possible to calculate a travel route that is more suitable for each pedestrian.

[0097] <Fourth embodiment> The route calculation device 10 of the fourth embodiment has a function of causing an output device installed on a passage to output predetermined information based on the calculation results of the tendency of the attributes of pedestrians passing through observation points on each passage. This will be explained in detail below.

[0098] 9 shows an example of a functional block diagram of the route calculation device 10 according to the fourth embodiment. As shown in the figure, the route calculation device 10 differs from the route calculation devices 10 according to the first to third embodiments in that it includes an output control unit 16.

[0099] The output control unit 16 causes an output device installed on the passage to output information indicating whether the point where the output control unit 16 is installed is suitable for passage by pedestrians with a predetermined attribute. Pedestrians with attributes suitable for passage through the point are pedestrians with attributes who frequently pass through the point. Pedestrians with attributes unsuitable for passage through the point are pedestrians with attributes who do not frequently pass through the point. Pedestrians with attributes who frequently pass through the point and pedestrians with attributes who do not frequently pass through the point are identified based on the results calculated by the tendency calculation unit 13 (see Figures 4 and 8), as described in the second and third embodiments.

[0100] Examples of output devices include digital signage, displays, signs whose display content can be switched by computer control, speakers, etc. Information indicating the installation location of each of the multiple output devices is registered in advance in the route calculation device 10. Based on this information and the location information of each of the multiple observation points, the route calculation device 10 identifies observation points near the installation locations of each of the multiple output devices, and causes each of the multiple output devices to output predetermined information based on the calculation results of the attribute trends of passersby passing through the identified observation points.

[0101] The output control unit 16 may cause each output device to output information such as "Men, women, and pedestrians frequently use this passage," "Women rarely use this passage," or "Women rarely use this passage between 7 PM and 5 AM." Fig. 10 shows an example of information output to the output device 30, which is a display or digital signage.

[0102] As another example, the output control unit 16 may specify conditions satisfied at the current time (the timing at which the display is controlled). The conditions are as described in the third embodiment and are defined by characteristics including at least one of time period, day of the week, month, weather, and season. The output control unit 16 may acquire information indicating the characteristics at the current time from an external server, or may acquire information indicating the characteristics at the current time based on a clock function or a calendar function provided in the route calculation device 10. The output control unit 16 may then output information indicating whether passage through the point where the output device is installed is suitable for passage by pedestrians with predetermined attributes under the conditions satisfied at the current time. An example of such information is, "At the current time period, passage through this passage is not suitable for women."

[0103] In addition, when the attributes of passersby are diverse, the amount of information indicating whether the location where the device is installed is suitable for passersby with a predetermined attribute becomes enormous. Therefore, the output control unit 16 may be provided with the following means for narrowing down the information to be output.

[0104] For example, the output control unit 16 may acquire an image generated by a camera capturing an image of the vicinity of the output device in real time, analyze the image, and identify the attribute of a passerby who is currently present at the location where the output device is installed (hereinafter, "first attribute"). Identification of the attribute of a passerby is achieved in the same manner as the identification unit 12.

[0105] The output control unit 16 may then cause the output device to output information indicating whether passage through the installed point is suitable for passage by pedestrians with the first attribute. Examples of the information include "This passage is not suitable for women" and "This passage is suitable for wheelchair users."

[0106] As another example, the output control unit 16 may use the above method to identify conditions that must be met for the timing of controlling the display, and may also identify a first attribute of a passerby currently present at the location where the output device is installed. The output control unit 16 may then cause the output device to output information indicating whether or not passage through the location where the output device is installed is suitable for a passerby with the first attribute, under the conditions currently met. An example of such information is, "This passage is not suitable for women at the current time."

[0107] As another example, the output control unit 16 may acquire an image generated by a camera capturing an image of the vicinity of the output device, analyze the image, and identify passersby with attributes whose passing status at the location where the output device is installed within a recent predetermined time period satisfies a predetermined condition. An example of the predetermined condition is "the attribute that passed through most frequently." The output control unit 16 may then cause the output device to output information indicating whether the passage is suitable for passersby with the identified attribute. Examples of the information include "This passage is not suitable for women to pass through" and "This passage is not suitable for women to pass through at the current time."

[0108] As another example, the output control unit 16 may switch the attributes for which information is displayed at predetermined time intervals. That is, the display content may be switched in such a manner that, after displaying information indicating whether the location where the indicator is installed is suitable for pedestrians with a certain attribute, information indicating whether the location where the indicator is installed is suitable for pedestrians with another attribute is displayed, and then information indicating whether the location where the indicator is installed is suitable for pedestrians with yet another attribute is displayed.

[0109] Other configurations of the route calculation device 10 of the fourth embodiment are similar to those of the route calculation device 10 of the first to third embodiments.

[0110] According to the route calculation device 10 of the fourth embodiment, the first to third 3The route calculation device 10 of the fourth embodiment achieves the same effects as those of the route calculation device 10 of the previous embodiment. Furthermore, the route calculation device 10 of the fourth embodiment can cause an output device installed on a passage to output information indicating whether or not passage through the installed point is suitable for pedestrians with predetermined attributes. As a result, it becomes possible to give a sense of security to pedestrians, warn them, or call their attention.

[0111] Although the embodiments of the present invention have been described above with reference to the drawings, these are merely examples of the present invention, and various other configurations may be adopted. The configurations of the above-described embodiments may be combined with each other, or some of the configurations may be replaced with other configurations. Furthermore, various modifications may be made to the configurations of the above-described embodiments without departing from the spirit of the invention. Furthermore, the configurations and processes disclosed in the above-described embodiments and modified examples may be combined with each other.

[0112] In addition, although the flowcharts used in the above explanations show multiple steps (processes) in a sequential order, the order of steps executed in each embodiment is not limited to the order shown. In each embodiment, the order of the steps shown in the drawings can be changed as long as it does not cause any problems in terms of content. Furthermore, the above-mentioned embodiments can be combined as long as the content is not contradictory.

[0113] A part or all of the above-described embodiments can be described as, but not limited to, the following supplementary notes. 1. An acquisition means for acquiring images taken by cameras installed at each of a plurality of observation points on the passage; An identification means for identifying attributes of passersby included in the image; a tendency calculation means for calculating a tendency of attributes of pedestrians passing through each of the plurality of observation points based on the result of the attribute identification; a route calculation means for calculating a travel route suitable for a pedestrian having the predetermined attribute based on the tendency; A route calculation device having the following. 2. The route calculation means Accept user input specifying the attribute The route calculation device according to claim 1, wherein the route calculation device calculates a travel route suitable for a pedestrian having the attributes specified by the user input. 3. The route calculation device according to claim 1 or 2, wherein the attributes include at least one of gender, age group, physique, belongings, means of transportation, and whether or not the user is accompanied. 4. The identifying means further identifies a characteristic of the timing at which the image was captured; the tendency calculation means calculates a tendency of an attribute of a pedestrian passing through each of the plurality of observation points when each of the plurality of conditions defined by the characteristic is satisfied; 4. The route calculation device according to any one of 1 to 3, wherein the route calculation means calculates a travel route suitable for a pedestrian having a predetermined attribute under predetermined conditions. 5. The route calculation means Accepting user input specifying the attribute and the condition 5. The route calculation device according to 4, which calculates a travel route suitable for a pedestrian having the attributes specified by the user input under the conditions specified by the user input. 6. The route calculation device according to 4 or 5, wherein the characteristics include at least one of time of day, day of the week, month, weather, and season. 7. A route calculation device described in any one of 1 to 6, further comprising an output control means for causing an output device installed on the passage to output information indicating whether passage through the installed point is suitable for passage by pedestrians with the specified attributes. 8. The output control means Identifying a first attribute, which is the attribute of a passerby who is currently present at the location where the output device is installed, based on an image generated by a camera that photographs the vicinity of the output device; 8. The route calculation device according to 7, wherein the output device is caused to output information indicating whether passage through the installed point is suitable for passage by a pedestrian with the first attribute. 9. The computer Images taken by cameras installed at multiple observation points along the corridor are acquired, Identifying attributes of passersby included in the image; Calculating the attribute trends of pedestrians passing through each of the plurality of observation points based on the attribute identification results; calculating a travel route suitable for a pedestrian having the predetermined attribute based on the tendency; Route calculation method. 10. Computer an acquisition means for acquiring images captured by cameras installed at each of a plurality of observation points on the passage; an identification means for identifying attributes of passersby included in the image; a tendency calculation means for calculating a tendency of attributes of pedestrians passing through each of the plurality of observation points based on the result of the identification of the attributes; a route calculation means for calculating a travel route suitable for a pedestrian having the predetermined attribute based on the tendency; A recording medium on which a program that functions as a [Explanation of symbols]

[0114] 10 Route calculation device 11 Acquisition Department 12 Specific part 13 Trend calculation section 14 Route calculation section 15 Storage section 16 Output control section 20 Image analysis system 30 Output Devices 1A processor 2A Memory 3A input / output I / F 4A peripheral circuit 5A Bus

Claims

1. an acquisition means for acquiring images captured by cameras installed at each of a plurality of observation points on the passage; An identification means for identifying attributes of passersby included in the image; a tendency calculation means for calculating a tendency of attributes of pedestrians passing through each of the plurality of observation points based on the result of the attribute identification; a route calculation means for calculating a travel route suitable for a pedestrian having the predetermined attribute based on the tendency; A route calculation device having the following.

2. The route calculation means Accept user input specifying the attribute The route calculation device according to claim 1 , wherein a travel route suitable for a traveler having the attributes specified by the user input is calculated.

3. The route calculation means identifying a first observation point at which a pedestrian having the same attribute as a first pedestrian whose travel route is to be calculated passes with a frequency lower than a reference value; 3. The route calculation device according to claim 1, wherein a route that does not pass through the first observation point is calculated as the travel route suitable for the first traveler.

4. The identifying means further identifies a characteristic of the timing at which the image was captured; the tendency calculation means calculates a tendency of an attribute of a pedestrian passing through each of the plurality of observation points when each of the plurality of conditions defined by the characteristic is satisfied; 4. The route calculation device according to claim 1, wherein the route calculation means calculates a travel route suitable for a pedestrian having the predetermined attribute under predetermined conditions.

5. The route calculation means identifying a second observation point at which a pedestrian having the same attribute as the first pedestrian whose travel route is to be calculated passes with a frequency higher than a reference value; 3. The route calculation device according to claim 1, wherein a route passing through the second observation point is calculated as the travel route suitable for the first traveler.

6. The route calculation means Calculating an evaluation value of the travel route; The route calculation device according to claim 5 , wherein the route calculation device calculates a higher evaluation value as the number of second observation points passed through increases.

7. 7. The route calculation device according to claim 1, further comprising an output control means for causing an output device installed on the passage to output information indicating whether passage through the installed point is suitable for passage by a pedestrian with a predetermined attribute.

8. The output control means Identifying a first attribute, which is the attribute of a passerby who is currently present at the location where the output device is installed, based on an image generated by a camera that captures an image of the vicinity of the output device; The route calculation device according to claim 7 , wherein the output device outputs information indicating whether passage through the installed point is suitable for passage by a pedestrian with the first attribute.

9. The computer Images taken by cameras installed at multiple observation points along the corridor are acquired, Identifying attributes of passersby included in the image; Calculating the attribute trends of pedestrians passing through each of the plurality of observation points based on the attribute identification results; calculating a travel route suitable for a pedestrian having the predetermined attribute based on the tendency; Route calculation method.

10. Computer, an acquisition means for acquiring images captured by cameras installed at each of a plurality of observation points on the passage; an identification means for identifying attributes of passersby included in the image; a tendency calculation means for calculating a tendency of attributes of pedestrians passing through each of the plurality of observation points based on the result of the identification of the attributes; a route calculation means for calculating a travel route suitable for a pedestrian having the predetermined attribute based on the tendency; A program that functions as a

Images