JP6645166B2 - Travel time storage system, travel time storage method, and travel time storage program - Google Patents

Description

  The present invention relates to storage of a person's travel time.

  In general, it is possible to calculate the time required to move between two points based on the time when the same person exists at two points. By accumulating the traveling time, the traveling time when the person moves between the two points can be calculated thereafter.

  Patent Literature 1 discloses a method of automatically measuring the time taken for a person to move between a certain point and another point. In this method, a device carried by a person transmits information unique to the person, and devices arranged at each measurement point receive the information. With this method, it is possible to acquire the time during which a specific person has been present at each measurement point. The travel time of the person is obtained by calculating the difference between the existence times at each measurement point. The acquired travel time is stored together with the measurement point and the measurement time.

JP 2003-279364 A

  A first problem in the related art is that the type and characteristics of each person are not taken into account in the storage of travel time.

  The reason is that information on travel time is collected based on the device carried by the person, and information on the type and characteristics of the person is not collected.

  An object of the present invention is to provide a system that stores information on the type and characteristics of a person and the travel time of the person in association with each other.

  According to one embodiment of the present invention, a moving time storage system includes a plurality of image capturing points including at least one person captured by a plurality of cameras, a capturing time of the image, face data of the person, and a type of the person. A feature extraction unit that extracts a moving time, a moving time calculating unit that calculates a moving time of the person based on the image capturing point, the image capturing time, and the face data, and associating the moving time with the type. Storage means for storing.

  A first effect of the present invention is that information on the type and characteristics of a person and the travel time of the person can be stored in association with each other.

FIG. 1 is a block diagram showing a configuration of a travel time storage system 1 according to the first embodiment of the present invention. FIG. 2 is a flowchart showing the operation of the travel time storage system 1 according to the first embodiment of the present invention. FIG. 3 is a block diagram showing a configuration of the travel time storage system 2 according to the second embodiment of the present invention. FIG. 4 is a flowchart showing the operation of the travel time storage system 2 according to the second embodiment of the present invention. FIG. 5 is a block diagram showing a configuration of the travel time storage system 3 according to the third embodiment of the present invention. FIG. 6 is a flowchart showing the operation of the travel time storage system 3 according to the third embodiment of the present invention. FIG. 7 is a flowchart showing the operation of the travel time storage system 3 according to the third embodiment of the present invention. FIG. 8 is a block diagram showing a configuration of the travel time storage system 4 according to the fourth embodiment of the present invention. FIG. 9 is a schematic diagram of a movement time and its appearance frequency histogram according to the fourth embodiment of the present invention. FIG. 10 is a block diagram showing the configuration of the travel time storage system 5 according to the first embodiment of the present invention. FIG. 11 is a block diagram showing the configuration of the travel time storage system 6 according to the second embodiment of the present invention. FIG. 12 is a conceptual diagram showing general face recognition characteristics.

  Next, embodiments for carrying out the present invention will be described in detail with reference to the drawings. In each of the drawings and the embodiments described in the specification, components having the same functions are denoted by the same reference numerals.

  Here, matters generally known about the characteristics of face recognition using images will be described.

  In general, there are two types of error rates for face image recognition: a false rejection rate (also called FalseNegativeRate) and a false acceptance rate (also called FalsePositiveRate). These error rates are in a trade-off relationship with each other. In operation, these error rates are set by adjusting a threshold value related to a face recognition score as a face image recognition result.

  In general, in face image recognition, the main factors that deteriorate the recognition rate are aging of the face due to the passage of time since the face image was registered, differences in the appearance of the face due to lighting conditions, and the camera installation angle. This is due to a difference in face direction (posture) caused by the difference. Therefore, the face image recognition process is generally executed after the face image is converted into data called a feature amount. However, a reduction in the recognition rate still occurs. In order to increase the recognition rate of a face image, it is important to register the latest face image of a person and to register a plurality of face images to correspond to various shooting conditions.

  FIG. 12 is a conceptual diagram showing the face recognition characteristics described above.

  FIG. 12 illustrates a feature amount of a face image as a two-dimensional vector and illustrates the feature amount on a plane. The feature amounts of the faces of the persons A and B are each registered as three points. The feature amount of the latest face image of the registered person usually becomes a high face recognition score equal to or larger than the threshold T, and falls in the area of the threshold T shown in FIG. However, for example, when there is aging which is one of the above-mentioned deterioration factors, the feature amount of the latest face image of the registered person A has a low face recognition score. Therefore, as shown by the black circle in the figure, the feature amount of the latest face image of the person A does not enter the area of the threshold T but enters the area of the threshold Y which is a lower threshold. In the area of the threshold Y, the feature amount of the face of another person B can also be included. Therefore, when the point indicated by the black circle is additionally registered as the person A, the person B may be erroneously registered as the person A. However, when the point indicated by the black circle is the feature amount of the latest face image of the person A who has aged, this point must be additionally registered as the person A in order to correctly operate the subsequent face recognition. . It is desirable that whether the point indicated by the black circle is actually the person A is confirmed by using other information.

[First Embodiment]
FIG. 1 is a block diagram showing a configuration of a travel time storage system 1 according to the first embodiment of the present invention.

  Referring to FIG. 1, the moving time storage system 1 according to the first embodiment of the present invention includes a feature extracting unit 11, a moving time calculating unit 12, and a storing unit 13.

  The moving time storage system 1 according to the first embodiment includes an image capturing point of a plurality of images including at least one person imaged by a plurality of cameras, an image capturing time of the image, face data of the person, and the person Is extracted. Further, the travel time storage system 1 calculates the travel time of the person based on the imaging point, the imaging time, and the face data. Then, the travel time storage system 1 stores the travel time and the type of the person in association with each other.

  That is, the travel time storage system 1 in the first embodiment calculates the travel time of the person based on the image of the person, not the device carried by the person. Therefore, since the traveling time storage system 1 can acquire the type of the person based on the image of the person, it is possible to store the type of the person and the traveling time in association with each other.

  Hereinafter, each component included in the travel time storage system 1 according to the first embodiment will be described.

=== Feature extraction means 11 ===
The feature extracting unit 11 receives a plurality of images captured by a plurality of cameras, each including at least one person. The feature extracting unit 11 extracts the face data of the person, the type of the person, the image capturing point of the image, and the image capturing time of the image based on the received image.

  The face data is data to be used for the face recognition described above. For example, data obtained by converting a face image called a face image or a feature amount into a numerical sequence corresponds to this.

  The type of a person means all appearance features of the person that can be determined from the image. For example, gender, physique (height, etc.), age, clothing (type of clothing, color, belongings) and the like.

  The imaging point is information on the position where the camera is installed (for example, a point ID or the like), and means that the person in the image exists at that point.

  The imaging time is the time when the camera captured the image, and means that the person in the image was present at that time.

=== Moving time calculation means 12 ===
When the feature extracting unit 11 newly extracts face data of the same person as the registered person, the moving time calculating unit 12 calculates the image capturing time of the image from which the new face data was extracted and the image stored immediately before that person. Is calculated with respect to the imaging time. The time calculated here is the moving time between the imaging points of the two images for which the difference between the imaging times has been calculated.

=== Storage means 13 ===
The storage unit 13 associates information (section ID and the like) indicating the section in which the person has moved, the travel time of the section calculated by the travel time calculation unit 12, and the type of the person extracted by the feature extraction unit 11. Attach and remember. The storage unit 13 may further store a person ID specifying a person in association with the person ID.

  The person ID is a number unique to the person. Information to which the same person ID is assigned means information of the same person.

  In the present embodiment, when the feature extraction unit 11 extracts face data of the same person as a registered person, the storage unit 13 acquires the type of the registered person. However, the timing for acquiring the type of the person is not limited to this.

  Table 1 shows an example of the association between the moving section, the moving time, and the type of the person stored in the storage unit 13.

  Table 1 is a table in which the travel time of the same section 1-2 is associated with the type of person. However, information of a plurality of sections may be put together in the same table. It is preferable that several types of persons are stored as shown in Table 1, but it is sufficient if at least one type is stored.

  FIG. 2 is a flowchart showing the operation of the travel time storage system 1 according to the first embodiment of the present invention.

  The moving time storage system 1 extracts the face data of the person, the type of the person, the image capturing point of the image, and the image capturing time of the image based on the image including the person (step S1).

  The moving time storage system 1 checks whether the person in this image is a registered person based on the extracted face data (step S2).

  When it is determined that the person in the image is a registered person (Y in step S3), the moving time storage system 1 captures the image capturing point and the capturing time of the extracted image, and captures the image in which the person was stored most recently. The travel time between the two points is calculated based on the point and the imaging time (step S4).

  The travel time storage system 1 stores the calculated travel time in association with the extracted type of person (step S5).

  When the moving time storage system 1 determines that the person in the image is a new person (N in step S3), the characteristics of the person extracted in step S1 (face data, type of person, imaging point, and imaging time) Is stored in association with the new person ID, that is, the extracted person is newly registered as a person (step S6). The travel time storage system 1 according to the first embodiment calculates a travel time of a person based on an image of the person. The travel time storage system 1 can acquire the type of the person by using the image of the person, so that the type of the person and the travel time can be stored in association with each other.

[Second embodiment]
FIG. 3 is a block diagram showing a configuration of the travel time storage system 2 according to the second embodiment of the present invention.

  Referring to FIG. 3, the movement time storage system 2 according to the second embodiment of the present invention includes a data processing device 210 and a storage device 220.

  The data processing device 210 includes a face registration unit 211, a face matching unit 212, a same person determination unit 213, a feature extraction unit 11, and a movement time calculation unit 12.

  The storage device 220 includes a registration data storage unit 221, a person presence data storage unit 222, and the storage unit 13.

  The moving time storage system 2 of the present embodiment further includes a face registration unit 211, a face matching unit 212, the same person determination unit 213, a registration data storage unit 221, and a person presence data storage unit 222. This is different from the moving time storage system 1 of the first embodiment.

  Hereinafter, differences between the present embodiment and the first embodiment will be described.

=== Registered Data Storage Means 221 ===
The registration data storage unit 221 stores the person ID and the face data extracted by the feature extraction unit 11 in association with each other. At this time, the registration data storage unit 221 may further store the type of the person in association with the person.

  Basically, new person IDs are sequentially assigned to the person IDs. However, when the same person determination unit 213 described later determines that the registered person is the same person, the person ID of the registered person is assigned.

=== Person Presence Data Storage Means 222 ===
The person presence data storage unit 222 stores the person ID, the image capturing point of the image extracted by the feature extracting unit 11, and the image capturing time of the image in association with each other. Here, the person ID, the image capturing point of the image extracted by the feature extracting unit 11, and the image capturing time of the image are collectively called person presence information.

=== Face Matching Means 212 ===
The face matching unit 212 checks the face data extracted by the feature extracting unit 11 with the face data stored in the registered data storage unit 221. The face matching unit 212 calculates a face recognition score indicating the degree of similarity between the extracted face data and the face data of a registered person.

=== Same person determination means 213 ===
The same person determination unit 213 compares the face recognition score calculated by the face matching unit 212 with a predetermined threshold, and determines whether the extracted person is the same as a registered person. If the face recognition score t exceeds a very high threshold T, the same person determination means 213 determines that the extracted person is the same person as the registered person.

  FIG. 4 is a flowchart showing a process in which the moving time storage system 2 registers a feature of a person extracted based on an image in association with a person ID.

  First, the moving time storage system 2 compares the face data extracted based on the image captured by the camera with the face data stored in the registration data storage unit 221 (step S211). Here, a face recognition score t representing the degree of similarity between the extracted face data and the registered face data is calculated. In the initial state, the face data is not registered in the registration data storage unit 221, so that this processing is practically omitted.

  Next, the travel time storage system 2 compares the face recognition score with a predetermined threshold value, and determines whether the extracted person is the same person as the registered person (Step S212). When the face recognition score t satisfies the following equation, the moving time storage system 2 determines that the extracted person is the same person as the registered person.

  The moving time storage system 2 determines that the extracted person is the same as the registered person P when the face recognition score t is higher than a predetermined threshold T as in Expression (1). Normally, a plurality of face data of a plurality of persons are registered in the registration data storage unit 221. However, the travel time storage system 2 may execute the above processing for the person P having the first face recognition score.

  Next, the moving time storage system 2 stores the extracted face data, the image capturing point, and the image capturing time in association with the person ID, that is, stores the extracted characteristics of the person as characteristics of a new person or a registered person. Register as a feature. When it is determined that the extracted person is the same person as the registered person P (Y in step S212), the moving time storage system 2 stores the extracted face data, the image capturing point, and the image capturing time of the person P. The information is stored in association with the person ID (step S213). When it is determined that the extracted person is different from the registered person P (N in step S212), the moving time storage system 2 sets the extracted face data, the imaging location, and the imaging time to a new person. The ID is stored in association with the ID (step S214).

  Here, the face data and the person ID are stored in the registered data storage unit 221. The imaging point, the imaging time, and the person ID are stored in the person presence data storage unit 222.

  Note that the face data and the person ID may be registered in advance before the system starts operating. It works without any problem even if there is no person presence information.

  Further, the registered data storage unit 221 and the person presence data storage unit 222 may be stored in one data format. For example, they may be stored together in tabular data (relational database).

  Also, the camera used for face recognition need not be a fixed camera. The camera itself may perform Pan-Tilt-zoom turning, or the camera may be mounted on a person or a car and moved. That is, the effect of the present invention can be obtained as long as the point where the person at the time of capturing the face is present can be obtained. In this case, since an imaging point appears at an arbitrary granularity, a point where no measurement history exists may occur. However, if the map is divided into meshes and one point ID is assigned to one mesh range, the present invention is realized. Applicable as it is.

  Further, in order to increase the information stored in the person presence data storage means 222, the person presence information obtained by means other than face recognition may be additionally used. For example, the owner recognition result using the personal mobile terminal may be added to the person presence data storage unit 222.

  The travel time storage system 2 according to the second embodiment calculates the travel time of a person based on the image of the person. The travel time storage system 2 can acquire the type of the person by using the image of the person, so that the type of the person and the travel time can be stored in association with each other.

[Third Embodiment]
Next, a third embodiment of the present invention will be described.

  The travel time storage system 3 according to the third embodiment differs from the travel time storage system 2 according to the second embodiment in the following two points. The first point is that the travel time calculation means 32 calculates the time (travel time) during which the same person travels between two locations, and, based on travel time calculated in the past, the shortest travel time between the two locations. The point is that the time (hereinafter, referred to as the shortest travel time) and the travel time obtained by combining a plurality of routes (hereinafter, referred to as a combined route travel time) are also calculated. The shortest travel time is stored in the shortest travel time storage unit 323 in association with the travel section. The detailed operation of the movement time calculation means 32 will be described later with reference to FIG. The second point is that, based on the face recognition score calculated by the face matching unit 212, the predetermined threshold value, and the shortest moving time, the same person determination unit 313 determines that the person extracted from the image is the same as the registered person. Is to determine whether The other configuration shown in FIG. 5 is the same as that of the travel time storage system 2 according to the second embodiment, and thus detailed description is omitted. The detailed operation of the same person determination means 313 will be described below.

=== Same person determination means 313 ===
The same person determination unit 213 compares the face recognition score calculated by the face matching unit 212 with a predetermined threshold, and determines whether the extracted person is the same as a registered person. If the face recognition score t exceeds a very high threshold value T or is equal to or greater than a predetermined threshold value Y (<T), and the movement between the extracted person's imaging point and the registered person's imaging point If the time exceeds the shortest movement time S stored in the storage unit 23, the same person determination unit 313 determines that the extracted person is the same person as the registered person.

  FIG. 6 is a flowchart showing a process in which the travel time storage system 3 calculates the shortest travel time between two points.

  First, the travel time storage system 3 searches for the shortest value of the difference between the times at which the same person was present at two locations from the information on the person ID, the imaging location, and the imaging time stored in the human presence data storage unit 222. Then, the travel time is directly calculated (step S321). For example, the travel time storage system 3 calculates the difference between the times when the same person exists between the two points 1-2, and sets the minimum value thereof as the shortest travel time. When the information shown in Table 2 is stored in the person presence data storage unit 222, the provisional shortest travel time shown in Table 3 is obtained.

  Next, the travel time storage system 3 calculates a travel time when a plurality of routes are combined (step S322). This corresponds to the travel time when it is considered that there is one or more waypoints. For example, when there is one waypoint, the total time of the difference between the imaging times at the point 1 and another point 3 and the difference between the imaging times at the point 3 and the point 2 is determined as the movement between the points 1-3-2. Time. Table 3 shows the shortest travel time obtained by calculating the travel time when a plurality of routes are combined as described above, and taking the minimum of them and the previously calculated direct travel time.

When multiple routes are combined and considered, not only the travel time is calculated based on only one person's presence information, but also the travel time is calculated based on the person's presence information of another person, and they are added. You may. In this way, the shortest travel time of the travel time that can be predicted from the person presence information is set as the shortest travel time. The rightmost column of Table 4 simply shows the calculation procedure. C ₂ (P) −C ₁ (P) appearing in Expression 2 described later is simply expressed as m (1,2).

  When the amount of the person presence information is not sufficient, there are many sections in which the movement time is unknown as in the latter half of Table 4. In that case, the moving time can be calculated as shown in Table 5 by setting the difference time of the reverse route as the shortest time candidate.

  It should be noted that the travel time of a person between points changes depending on traffic conditions and changes in transportation. For this reason, by narrowing down the person presence information used for calculating the shortest travel time to the latest fixed period (eg, one day, one week, etc.), it is possible to calculate the shortest travel time more accurately.

  Further, the travel time between points often fluctuates periodically depending on the date and time, the day of the week, and the season. For this reason, by narrowing down the person presence information used to calculate the shortest travel time by date and time, day of the week, and season, more accurate calculation of the shortest travel time can be performed.

  Furthermore, there is a lower limit to the travel time between points due to physical travel speed restrictions. For this reason, a predetermined lower limit may be set manually for some of the shortest movement times.

  FIG. 7 is a flowchart showing a process in which the moving time storage system 3 registers a feature of a person extracted based on an image in association with a person ID.

  The difference from the second embodiment is step S312 in FIG. In step S312, based on the face recognition score, the predetermined threshold value, and the shortest movement time stored in the shortest movement time storage means 323, it is determined whether or not the person extracted by the feature extraction means 11 is the same as the registered person. Is determined (step S312). When the face recognition score satisfies the above equation (1) or the following equation (2), the moving time storage system 3 determines that the extracted person and the registered person are the same.

It is assumed that the point ID of the point (imaging point) where the camera currently being processed is located is 1, and the point ID of the point where the latest existence of a certain person (person ID = P) is stored is 2. The time when the face was photographed by the camera currently being processed (imaging time) is C ₁ (P), the time when the latest existence of the person is stored is C ₂ (P), Let the shortest moving time be M (1,2).

  Equation 2 represents the condition for the same person determination. The time at which the extracted face recognition score t (P, 1, 2) of the person is higher than the threshold Y lower than the threshold T, and the time at which the latest existence of the person P is stored is determined by the camera currently being processed. When the time difference (moving time) from the photographing time is longer than the shortest moving time in the same section 1-2, the same person determination unit 313 determines that the extracted person is the same person as the person P.

  Note that in the storage unit 23, only the equation (1) is applied to a section where the shortest travel time is unknown, that is, a section in which one or both of two points have no person presence information of the same person.

  Next, effects of the present embodiment will be described.

  In the present embodiment, the face data and the person ID are registered after the face matching means is provided, and the same person determination means determines whether or not the person is the same in consideration of both the moving time and the face recognition score. Since the face registration unit is provided, the movement time can be calculated with higher accuracy.

  Hereinafter, the reason why this effect is obtained will be described.

  In the present embodiment, the ID of the registered person is determined only when the face recognition score t is larger than the very high threshold T or when both the conditions of the face recognition score threshold Y and the moving time are satisfied. Registered at. For this reason, with regard to the person ID of the registered face data, a plurality of IDs may be assigned to one face data, but it is extremely rare that one ID is assigned to the faces of a plurality of persons. . That is, since it is unlikely that another person is erroneously recognized and accepted, the shortest traveling time is erroneously underestimated. In particular, since the threshold Y slightly lower than the very high threshold T can be used, even when the registered face data and the latest face data are slightly different, the latest face data can be registered, and the face recognition accuracy can be improved. Will be higher. From this, when the system of the present invention is operated, the face recognition accuracy and the movement time calculation accuracy are mutually improved.

Furthermore, in the calculation of the shortest travel time, not only the direct travel time between two points is memorized, but also a plurality of sections are calculated by combining the times required by different persons, so that the existence (movement) between the points The travel time can be calculated even in a section where the result of ()) is small, and the shortest travel time can be calculated with higher accuracy.
Further, by narrowing the data used for calculating the travel time to periodic information and the latest information, higher calculation accuracy can be obtained.

  In the case of the second embodiment, if the threshold value T of the face recognition score is adjusted to be too high without any darkness, the false acceptance rate approaches 0% without limit, but the false rejection rate increases. May not be able to recognize most of the person in front of the vehicle, and may not be able to measure the travel time sufficiently. However, in the present embodiment, since the threshold Y lower than the threshold T is also used, this problem is reduced.

[Fourth Embodiment]
Next, a fourth embodiment of the present invention will be described.

  The travel time storage system 4 in the fourth embodiment differs from the travel time storage system 3 in the third embodiment in the following two points. The first point is that the storage means 43 stores the appearance frequency index in association with the moving section, the moving time, and the type of the person. The appearance frequency is an index indicating the frequency at which a certain travel time has appeared in the past. The storage means 43 also stores the non-shortest travel time. That is, the appearance frequency index is a frequency histogram related to the travel time (see FIG. 9). The second point is that, based on the face recognition score calculated by the face matching unit 212, the predetermined threshold value, the shortest movement time, and the appearance frequency index, the same person determination unit 413 registers the person extracted from the image. The point is to determine whether the person is the same person. The other configuration shown in FIG. 8 is the same as that of the travel time storage system 2 of the third embodiment, and thus a detailed description is omitted. The detailed operation of the same person determination means 413 will be described below.

=== Same person determination means 413 ===
The same person determination unit 413 determines whether or not the registered person is the same person using not only the expressions (1) and (2) but also the appearance frequency q of the moving time. When the appearance frequency q is equal to or more than a certain threshold Q, the likelihood that the relevant person exists in the place with the moving time is high. FIG. 9 is a schematic diagram of the movement time and its appearance frequency histogram. In the example of this schematic diagram, it is assumed that there are two types of travel routes / means of a person (for example, a train and a walk), and most past people have travel times closer to one of the two peak travel times. It turns out that it was. For this reason, in the present embodiment, the same person determination means 413 uses the following equation (3) as the same person determination condition. Specifically, when Expression (1), Expression (2), or Expression (3) is satisfied, the same person determination unit 413 determines that the extracted person is the same person as the registered person. .

  Y ′ in Expression (3) is a threshold value of the face recognition score. Y 'may be a fixed value, but can be set lower as the appearance frequency q increases. Therefore, Y ′ may be calculated as in equation (4) using a monotonous function f (·). For example, Y 'may be calculated by a sigmoid function in which parameters are adjusted so that f (·) becomes a monotonous phenomenon.

  Next, effects of the present embodiment will be described.

  By using the present embodiment, it becomes easy to recognize the same person by face recognition on a route where a person frequently moves, and more information can be used when calculating the movement time table. As a result, the moving time calculation accuracy becomes higher.

<Example 1>
Next, a first embodiment of the second aspect of the present invention will be described with reference to FIG. The present embodiment is divided into two types: processing by a camera at each point and processing over a plurality of points. Therefore, as shown in FIG. 10, the PC 510 may be assigned to processing at each point, and the server 520 may be assigned to processing spanning a plurality of points. The processing at each point includes a face matching unit 212, a same person determination unit 213, and a feature extraction unit 11, which are configured as software. On the other hand, the processing over a plurality of points includes a registration data storage unit 221, a person presence data storage unit 222, a travel time calculation unit 12, and a storage unit 13, which are configured as data on a memory and software. As for each storage unit, a storage unit may be prepared on the PC 510 side to temporarily store data, and then transmitted to the server 520 side for storage. In the case of this embodiment, the PC 510 always processes one or a plurality of camera images, and the server 520 only needs to update the storage unit at the timing when the moving time calculating unit calculates the moving time. The calculation of the travel time may be performed periodically.

<Example 2>
Next, a second embodiment of the third aspect of the present invention will be described with reference to FIG. The present embodiment is an example in which the shortest travel time is calculated, and then the travel route of a specific person is specified. The difference from FIG. 5 is that a moving path specifying unit 600 is provided, and the moving path of the designated person ID is specified. Specifically, the data of the corresponding person ID is referred to and extracted from the person presence data storage unit 222. According to the present invention, since the person data storage unit 222 stores valid data regarding the face recognition result and the shortest travel time, it is possible to grasp the transition of the location of the person.

  As described above, the present invention has been described with reference to the respective embodiments and examples, but the present invention is not limited to the above embodiments. Various changes understandable to those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention.

  In addition, each component in each embodiment of the present invention can be realized by a computer and a program, as well as by realizing the function in hardware. The program is provided by being recorded on a computer-readable recording medium such as a magnetic disk or a semiconductor memory, and is read by the computer when the computer starts up. By controlling the operation of the computer, the read program causes the computer to function as a component in each of the above-described embodiments.

  Some or all of the above embodiments and examples can be described as in the following supplementary notes, but are not limited thereto.

(Appendix 1)
A feature extraction unit that extracts an imaging point of a plurality of images including at least one person imaged by a plurality of cameras, an imaging time of the image, face data of the person, and a type of the person;
Based on the imaging point, the imaging time, and the face data, travel time calculation means for calculating the travel time of the person,
A moving time storage system, comprising: a storage unit configured to store the moving time and the type in association with each other.

(Appendix 2)
The shortest travel time, which is the shortest travel time, of the sum of the travel time between the two points and the travel time between the two points via another point where another person was present, 2. The travel time storage system according to claim 1, wherein

(Appendix 3)
3. The travel time storage system according to claim 2, wherein the shortest travel time is calculated based on the image capturing point and the image capturing time during the latest fixed time.

(Appendix 4)
3. The travel time storage system according to claim 2, wherein the shortest travel time is calculated based on the past regular imaging point and the imaging time.

[Industrial applicability]
ADVANTAGE OF THE INVENTION According to this invention, it can apply to the use which specifies a moving route of a person with higher precision, and analyzes an action. For example, it can be used for analyzing the behavior of a visitor in a large-scale store.

1-6 Moving time storage system 11 Feature extraction means 12, 32 Moving time calculation means 13, 43 Storage means 210 Data processing device 212 Face matching means 213, 313, 413 Same person determination means 220 Storage device 221 Registered data storage means 222 People Presence data storage means 323 Shortest travel time storage means 510 PC
520 server 600 travel route specifying means

Claims (8)

A feature extraction unit that extracts an imaging point of a plurality of images including at least one person imaged by a plurality of cameras, an imaging time of the image, face data of the person, and a type of the person;
Based on the imaging point, the imaging time, and the face data, travel time calculation means for calculating the travel time of the person,
Storage means for storing the travel time and the type in association with each other,
Face matching means for calculating a face recognition score representing a degree of similarity between the face data of the person and face data of a registered person,
Comparing the face recognition score with a threshold, and when the face recognition score exceeds the threshold, the same person determination unit that determines that the person extracted from the image is the same person as the registered person. Prepared,
The same person determination unit refers to a relationship between the movement time of the registered person and the frequency of appearance at the movement time, and regarding the movement time of the person extracted from the image, the appearance of the registered person. The moving time storage system sets the threshold value lower as the frequency increases .   The travel time storage system according to claim 1, wherein the storage unit stores the travel time and the plurality of types in association with each other. The shortest travel time, which is the shortest travel time, of the sum of the travel time between two locations and the travel time between the two locations via another location where another person was present, The travel time storage system according to claim 1 , wherein the travel time storage system is calculated. The moving time storage according to claim 3 , wherein the same person determination unit determines whether a person extracted from the image is the same as the registered person based on the shortest moving time and the face recognition score. system. The same person determination means determines whether a person extracted from the image is the same person as the registered person when the face recognition score exceeds the threshold and the movement time exceeds the shortest movement time. 4. The travel time storage system according to claim 3 , wherein: The travel time storage system according to claim 3 , wherein the shortest travel time is calculated based on information on a reverse route between two points. Extracting the image capturing points of a plurality of images including at least one person imaged by a plurality of cameras, the image capturing time of the images, the face data of the person, and the type of the person,
Based on the imaging point, the imaging time, and the face data, calculate the travel time of the person,
The travel time and the type are stored in association with each other,
Calculating a face recognition score representing the degree of similarity between the face data of the person and the face data of a registered person,
Comparing the face recognition score and a threshold, if the face recognition score exceeds the threshold, determine that the person extracted from the image is the same person as the registered person,
Referring to the relationship between the travel time of the registered person and the frequency of appearance at that travel time, the travel time of the person extracted from the image, the higher the frequency of appearance of the registered person, the higher the threshold Travel time storage method to set low . On the computer,
A process of extracting an image capturing point of a plurality of images including at least one person imaged by a plurality of cameras, an image capturing time of the image, face data of the person, and a type of the person;
A process of calculating the moving time of the person based on the imaging point, the imaging time, and the face data;
A process of storing the travel time and the type in association with each other;
A process of calculating a face recognition score representing a degree of similarity between the face data of the person and face data of a registered person;
Comparing the face recognition score and a threshold, when the face recognition score exceeds the threshold, performing a process of determining that the person extracted from the image is the same person as the registered person,
In the process of determining the same person, the relationship between the travel time of the registered person and the frequency of appearance at the travel time is referred to, and the travel time of the person extracted from the image, A moving time storage program for setting the threshold value lower as the appearance frequency increases .

Images