Target distribution estimation program, device, system and method for performing expanded estimation based on address and location survey results

Abstract

To provide an object distribution estimation program that estimates a high-resolution distribution with high accuracy not based upon correct answer data on an actual distribution.SOLUTION: In a communication facility apparatus 1 which executes an object distribution estimation program, an existence number determination part 111 determines an existence number distribution of objects from positioning results of terminals 2. An address / location expansion coefficient determination part 112 determines, for each area, an address / location expansion coefficient of objects in an address / location area based upon the number of existing terminals in a time section at an address / location and an address / location area investigation result. An address / location area determination part 113 determines, for each spot, an address / location area of an object to which a terminal at the spot is added from history information on a registration information / positioning result. An expansion coefficient estimation part 114 estimates, for each spot, an expansion coefficient of the spot from a ratio calculated regarding the area of the object to which a specified terminal is added as an address / location area and the address / location expansion coefficient. An object existence distribution determination part 115 determines the existence number of objects at respective spots from the existence number distribution and the estimated enlargement coefficient.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a technique for estimating a distribution of objects, for example, a population distribution, in a given area. [Background technology]

[0002] Understanding the population distribution (or people flow, including distribution that changes over time) in a given area as high-resolution data is extremely important for marketing, efficient operation of transportation networks (avoiding congestion, etc.), urban planning including the development of public facilities, and even infectious disease countermeasures.

[0003] Here, increasing the resolution of population distribution (people flow) data means decreasing the size of the regional mesh (hereafter abbreviated as mesh), which is the unit for tabulating population (people flow). In this regard, the positioning data groups from mobile devices (owned by users) that have traditionally been used as the source data for population distribution are, for example, data groups related to points in latitude and longitude coordinates, and in that sense have the highest resolution. However, it is natural that such positioning data groups alone cannot provide information on the positions of all people in the target area.

[0004] Therefore, in practice, population distribution data must be generated by performing an expansion estimate to the total population from the mobile device positioning data. However, if an expansion estimate is simply performed from only the mobile device positioning data, for example, for meshes where no positioning data exists, the error in the estimate due to the expansion estimate will be large (because no matter how many times you multiply 0 people, it still remains 0 people). Furthermore, particularly in urban areas, there is a good chance that GPS (Global Positioning System) positioning data originally contains large errors due to the influence of densely packed buildings, etc.

[0005] As a result, in order to generate population distribution (people flow) data that accurately reflects the overall distribution trends of all people, it has become necessary to set the mesh size large, i.e., low resolution, and to avoid situations where each mesh does not contain positioning data or does not (does) have positioning data that it should (should not) have. Figure 6 shows a schematic diagram of how the accuracy (precision) of a population distribution heat map generated using a smaller mesh (high-resolution mesh) from sparse terminal location information (positioning data) actually drops significantly.

[0006] As an example of a technology for achieving high resolution of population distribution (people flow), which has been difficult to achieve, Non-Patent Document 1 discloses a technology that models the spatial correlation of people flow and the effects of external factors such as weather, and uses AI (artificial intelligence) to estimate a high-resolution people flow map while using the total population as a constraint.

[0007] In addition, Non-Patent Document 2 discloses a technology for generating high-resolution people flow data by combining an AI model that achieves super-resolution in the spatial direction and an AI model that achieves super-resolution in the temporal direction.

[0008] Furthermore, Non-Patent Document 3 discloses a technology for generating detailed people flow data by combining (a) smartphone location information and (b) beacon location information. In this technology, (a) smartphone location information is positioned as "data with low resolution that can tell us the volume of people flow (population) in a certain wide area," and (b) beacon location information is positioned as "data for only a portion of users, but from which detailed location information can be obtained," and the data of (b) beacon location information is expanded and estimated so that the total population is the number of (a) smartphone location information. [Prior art documents] [Non-patent literature]

[0009] [Non-Patent Document 1] Fan Zhou, Liang Li, Ting Zhong, Goce Trajcevski, Kunpeng Zhang, Jiahao Wang, “Enhancing Urban Flow Maps via Neural ODEs”, Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence (IJCAI2020) Main track. Pages 1295-1302, 2020 [Non-Patent Document 2] Zong, Z., Feng, J., Liu, K., Shi, H., & Li, Y., “DeepDPM: Dynamic Population Mapping via Deep Neural Network”, Proceedings of the AAAI Conference on Artificial Intelligence, 33(01), pp.1294-1301. 2019 [Non-Patent Document 3] "Tomorrow's Congestion Forecast Portal", [online], [Retrieved March 10, 2022], Internet<URL: https: / / prtimes.jp / main / html / rd / p / 000000004.000090522.html> Summary of the Invention [Problem to be solved by the invention]

[0010] However, the conventional techniques disclosed in the above-mentioned Non-Patent Documents 1 to 3 still have problems with increasing the resolution of population distribution (people flow) data.

[0011] For example, the techniques disclosed in Non-Patent Documents 1 and 2 are both premised on the acquisition of high-resolution correct data of people flow in the estimation target area. In other words, high-resolution correct data is required for training an AI model. However, in reality, such correct data is rarely acquired, and therefore this premise is often unrealistic. For example, a group of GPS positioning data from a smartphone provides only the location information of "users who have a contract with the target telecommunications carrier, own a GPS-equipped model, and agree to the provision of GPS positioning data" out of the location information of the entire population that should be the correct answer, and it is clearly not possible to use this as correct data for people flow (population distribution).

[0012] In addition, the technology disclosed in Non-Patent Document 3 uses interpolation processing to make up for the location information (location data) from beacons (b), which is only collected spot-wise, and there are concerns that the accuracy of the estimated location may decrease. Furthermore, the smartphone location information (a) is considered to be "data that indicates the volume of people flow (population)," and for example, this information is merely subjected to a simple, uniform, expanded estimate. As a result, there are concerns that the accuracy of the estimated number of people may decrease.

[0013] Therefore, the present invention aims to provide a target distribution estimation program, device, system, and method that can estimate the distribution of a high-resolution target with higher precision (accuracy) without relying on ground-truth data related to the actual distribution of the target. [Means for solving the problem]

[0014] According to the present invention, there is provided a target distribution estimation program for causing a computer to function to estimate the distribution of a plurality of targets in a predetermined area. 、 For each area within the designated region, Regarding the device attached to the subjectan address / location expansion coefficient determination means for determining the number of terminals staying in the area during a time period related to the address or location based on historical information of positioning results or registration information about the terminals, and for determining an address / location expansion coefficient as an expansion coefficient for a target whose address or location area is the area based on the number of terminals and the acquired survey results of the address or location of the target; For each point within the specified area, Regarding the device attached to the subject an address / location area determination means for identifying a terminal related to the location from the positioning result, and determining an address / location area as the address or location of a target to which the identified terminal is assigned from registration information on the identified terminal or history information on the positioning result; an expansion coefficient estimation means for calculating, for each said point, a proportion of said areas in the target to which said specified terminal is attached that is an address / location area, and estimating an expansion coefficient of said point based on said proportion and said address / location expansion coefficient; a presence distribution determining means for determining a presence distribution as a distribution of the number of objects present in a specified area from a positioning result obtained during a specified period for a terminal attached to the object, the presence distribution relating to an estimation time point of the distribution of a plurality of objects; an object presence distribution determining means for determining the number of objects present at each of a plurality of points within the predetermined area based on the presence distribution and the estimated expansion coefficient; A target distribution estimation program is provided that causes a computer to function as the

[0015] In this object distribution estimation program according to the present invention, it is also preferable that the expansion coefficient estimation means weights the address / location expansion coefficient for each point by the relevant ratio and then sums the coefficients across the address / location area, and determines the result of the summation as the expansion coefficient for the point.

[0016] In one embodiment of the object distribution estimation program according to the present invention, the presence count distribution determination means determines an estimation time presence count distribution as a distribution of the numbers present at or immediately before the estimation time of the distribution from the positioning results accumulated during a first predetermined period and the positioning results acquired during a second predetermined period that is shorter than the first predetermined period and immediately before the estimation time of the distribution, It is also preferable that the target presence distribution determining means determines the number of targets present at each of a plurality of points within the specified area based on the estimated presence number distribution and the estimated expansion coefficient.

[0017] Furthermore, in the above embodiment, it is also preferable that the presence distribution determination means sets, as the first specified period, a period corresponding to a period type related to a preset month, day, day of the week, weekday / holiday, and / or time zone, which period type includes the estimated time point of the distribution.

[0018] Furthermore, in the above embodiment, it is also preferable that the presence distribution determination means determines the presence distribution at the time of estimation by assigning a greater weight to the positioning result the more recent the positioning time point related to the positioning result is within the first specified period and the second specified period.

[0019] In addition, in the above embodiment, it is also preferable that the presence distribution determination means, for each positioning result obtained in the first specified period and the second specified period, sets at least one presence point corresponding to the positioning result within an area having a size that is a monotonically increasing function of the positioning error included in the positioning result or determined from the positioning result, the presence point being weighted so that the sum of the weighted points is 1, and determines the estimation-time presence distribution from the presence point.

[0020] Further, in the above embodiment, the abundance distribution determining means determining a subperiod presence number distribution as a distribution of the number of devices present at a representative time point of each subperiod from the positioning result acquired in the subperiod for each subperiod among a plurality of subperiods included in the first predetermined period and the second predetermined period; The subperiod abundance distribution at the representative time point is extrapolated to the abundance distribution at the estimation time point, and the abundance distribution obtained by the extrapolation is determined as the abundance distribution at the estimation time point. It is also preferable.

[0021] According to the present invention, there is also provided an object distribution estimation device for estimating the distribution of a plurality of objects in a predetermined area, 、 For each area within the designated region, Regarding the device attached to the subject an address / location expansion coefficient determination means for determining the number of terminals staying in the area during a time period related to the address or location based on historical information of positioning results or registration information about the terminals, and for determining an address / location expansion coefficient as an expansion coefficient for a target whose address or location area is the area based on the number of terminals and the acquired survey results of the address or location of the target; For each point within the specified area, Regarding the device attached to the subject an address / location area determination means for identifying a terminal related to the location from the positioning result, and determining an address / location area as the address or location of a target to which the identified terminal is assigned from registration information on the identified terminal or history information on the positioning result; an expansion coefficient estimation means for calculating, for each said point, a proportion of said areas in the target to which said specified terminal is attached that is an address / location area, and estimating an expansion coefficient of said point based on said proportion and said address / location expansion coefficient; a presence distribution determining means for determining a presence distribution as a distribution of the number of objects present in a specified area from a positioning result obtained during a specified period for a terminal attached to the object, the presence distribution relating to an estimation time point of the distribution of a plurality of objects; an object presence distribution determining means for determining the number of objects present at each of a plurality of points within the predetermined area based on the presence distribution and the estimated expansion coefficient; An object distribution estimator is provided, comprising:

[0022] According to the present invention, there is further provided an object distribution estimation system for estimating a distribution of a plurality of objects in a predetermined area, 、 For each area within the designated region, Regarding the device attached to the subject an address / location expansion coefficient determination means for determining the number of terminals staying in the area during a time period related to the address or location based on historical information of positioning results or registration information about the terminals, and for determining an address / location expansion coefficient as an expansion coefficient for a target whose address or location area is the area based on the number of terminals and the acquired survey results of the address or location of the target; For each point within the specified area, Regarding the device attached to the subjectan address / location area determination means for identifying a terminal related to the location from the positioning result, and determining an address / location area as the address or location of a target to which the identified terminal is assigned from registration information on the identified terminal or history information on the positioning result; an expansion coefficient estimation means for calculating, for each said point, a proportion of said areas in the target to which said specified terminal is attached that is an address / location area, and estimating an expansion coefficient of said point based on said proportion and said address / location expansion coefficient; a presence distribution determining means for determining a presence distribution as a distribution of the number of objects present in a specified area from a positioning result obtained during a specified period for a terminal attached to the object, the presence distribution relating to an estimation time point of the distribution of a plurality of objects; an object presence distribution determining means for determining the number of objects present at each of a plurality of points within the predetermined area based on the presence distribution and the estimated expansion coefficient; An object distribution estimation system is provided having:

[0023] According to the present invention, there is further provided a method for estimating a distribution of a plurality of objects in a predetermined area, the method being implemented by a computer. 、 For each area within the designated region, Regarding the device attached to the subject Based on the positioning result history information or the registration information of the terminal, the number of the terminals staying in the area during the time period related to the address or location is determined, and based on the number of the terminals and the survey results of the address or location of the target obtained, an address / location expansion coefficient is determined as an expansion coefficient for the target whose address or location area is the area, and for each point within the specified area, Regarding the device attached to the subject A step of identifying a terminal related to the location from the positioning result, and determining an address or location area as an address or location of a target to which the identified terminal is assigned from registration information on the identified terminal or history information on the positioning result; A step of calculating, for each said location, a proportion of the target to which the specified terminal is attached that is an address / location area of ​​each of said areas, and estimating an expansion coefficient of said location based on said proportion and said address / location expansion coefficient; A step of determining a presence distribution as a distribution of the number of objects present in the specified area from the positioning results obtained during a specified period of time for the terminal attached to the object, the presence distribution relating to an estimation time point of the distribution of a plurality of objects; determining a number of the objects at each of a plurality of locations within the predetermined area based on the number distribution and the estimated expansion factor; A method for estimating a target distribution is provided, having the following: Effect of the Invention

[0024] According to the object distribution estimation program, device, system, and method of the present invention, the distribution of an object can be estimated with high resolution and higher precision (accuracy) without relying on ground-truth data regarding the actual distribution of the object. [Brief description of the drawings]

[0025] [Figure 1] 1 is a functional block diagram showing a functional configuration of an embodiment of a target distribution estimation device according to the present invention. [Diagram 2] FIG. 13 is a schematic diagram for explaining another embodiment of the existence count distribution determination process according to the present invention. [Diagram 3] FIG. 13 is a schematic diagram for explaining still another embodiment of the abundance distribution determination process according to the present invention. [Figure 4] FIG. 13 is a schematic diagram for explaining another embodiment of the existence count distribution determination process according to the present invention. [Diagram 5] 1 is a schematic diagram showing an overview of an embodiment of a target distribution estimation method according to the present invention; [Figure 6] FIG. 1 is a schematic diagram for explaining a conventional problem in generating a population distribution heat map using a high-resolution mesh. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0026] Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.

[0027] [Target distribution estimation device] FIG. 1 is a functional block diagram showing a functional configuration of an embodiment of an object distribution estimation device according to the present invention.

[0028] The communication equipment 1 shown in Fig. 1 is of course an equipment that performs the function of communication equipment (e.g., a relay function), but it is also an embodiment of an object distribution estimation device according to the present invention. That is, the communication equipment 1 is an equipment that estimates the distribution of a plurality of objects in a predetermined area (e.g., the entire country of 47 prefectures), in this embodiment, the distribution of a large number of people (i.e., population distribution), and outputs population distribution data (e.g., a population distribution graph or a population heat map) as the estimation result.

[0029] Specifically, the communication equipment 1 receives a “terminal positioning result” which is a positioning result for a terminal 2 (e.g., a smartphone carried by a person) attached to a human (target), for example, (a) GPS positioning data from the GPS (Global Positioning System) function installed in terminal 2; (b) base station positioning data by a base station positioning method using base station 3, and (c) Positioning data obtained by a wireless LAN access point when the terminal 2 is equipped with a wireless LAN (Local Area Network) communication function such as Wi-Fi (registered trademark) At least one of the above is obtained, and the population distribution in a specified area (nationwide) is estimated based on this data.

[0030] In order to perform such estimation, the communication equipment 1 (A) A presence count distribution determination unit 111 that determines a "presence count distribution" (in this embodiment, an estimated presence count distribution described later) as a distribution of the number of objects present in a predetermined area (for example, nationwide) from the "terminal positioning results" acquired during a predetermined period; (B) an address and location expansion coefficient determination unit 112 that determines, for each of a plurality of zones within a predetermined region (nationwide), the number of terminals 2 that are present within the zone during a set time period related to the address or location (for example, the nighttime period from 1:00 to 5:00) based on history information of the "terminal positioning result" or registration information on the terminals 2, and determines an "address and location expansion coefficient" as an expansion coefficient related to the objects (humans) whose address or location area is the zone based on the number of terminals 2 and the "address and location investigation result" which is an investigation result of the address or location of the object (human) that has been acquired; (C) an address / location area determination unit 113 that identifies a terminal 2 that exists within a predetermined range including each point among a plurality of points within a predetermined region (nationwide) from the "terminal location result" and determines an "address / location area" as the address or location of the object (human) to which the identified terminal 2 is attached from history information of the "terminal location result" for the identified terminal 2 or from registered information of the object (human) to which the identified terminal 2 is attached; (D) an expansion coefficient estimation unit 114 that calculates the ratio of each of a plurality of areas of the target (person) to which the identified terminal 2 is attached as an "address / location area" for each point, and estimates the "expansion coefficient" of the point based on this ratio and the "address / location expansion coefficient"; (E) an object presence distribution determination unit 115 that determines the number of objects (people) present (the number of people present) at each of a plurality of points within a predetermined region (nationwide) based on the "presence distribution" (presence distribution at the time of estimation) and the estimated "expansion coefficient"; The present invention is characterized by having the following.

[0031] Here, in this embodiment, the target presence distribution determination unit 115 (E) above generates and outputs population distribution data (for example, a population distribution graph or a population heat map) as an estimation result from the determined number of people present at each point.

[0032] In addition, in this embodiment, the "address and location investigation result" in (B) above can be, for example, publicly available census data. Of course, data other than census data that shows the addresses and locations of all subjects (everyone) in a specific area, or data that shows the number of all subjects (everyone) at each location can be used as the "address and location investigation result".

[0033] The communications equipment 1 uses such "address and location survey results" to calculate an "address and location expansion coefficient," and then uses this "address and location expansion coefficient" and the "address and location area" of targets (people) attached with terminals 2 that are present within a specified range that includes each point to estimate the "expansion coefficient" for each point, thereby determining the number of targets present (number of people present) at each point.

[0034] Here, even if the correct number of people at each location (each area) of the target (human) is not obtained, the "Address / Location Expansion Coefficient" is calculated using the "Address / Location Survey Results", so the "Expansion Coefficient" for each location can be obtained. Furthermore, this "expansion coefficient" is different from the conventional simple and uniform one such as (total population) / (total number of users permitted to provide location information), and is more precise (accurate) as it reflects the characteristics (that affect the actual number of people) at that location.

[0035] Furthermore, a sufficient number of points within the predetermined region (nationwide) in (E) above can be set. As a result, it is possible to generate high-resolution object distribution data (population distribution data). As described above, the communication equipment 1 makes it possible to estimate the distribution of objects (people) with high resolution and with higher precision (accuracy) without relying on correct answer data related to the actual distribution of objects (people).

[0036] In this embodiment, the target for which distribution is sought is "people", and the following description will also assume that the target is "people" to estimate the population distribution (people flow), but the target according to the present invention is not limited to "people". In other words, various targets can be adopted as targets according to the present invention, as long as they include those equipped with a terminal 2 that can obtain "terminal positioning results", and as long as location survey results and survey data showing the number of targets per area are available. For example, "automobiles" can be adopted as targets according to the present invention, as long as at least some of the automobiles are equipped with a terminal 2, and as long as survey results of the garage location (location) where the automobiles are registered and survey data of the number of automobiles per area are available.

[0037] [Device functional configuration, target distribution estimation system, program, and method] Also according to the functional block diagram of FIG. 1, a communication equipment device 1 as one embodiment of a target distribution estimation device according to the present invention has a communication interface unit 101, a positioning result storage unit 102, an address / location investigation result storage unit 103, a user interface unit 104, and a processor / memory (a calculation processing system with a memory function).

[0038] Here, the processor memory stores one embodiment of the target distribution estimation program according to the present invention, has a computer function, and executes the target distribution estimation program to perform the target distribution estimation process. Therefore, the communication equipment device 1 may be a device dedicated to the target distribution estimation process other than communication equipment, and may further be a cloud server, a non-cloud server, a personal computer (PC), a notebook or tablet computer, or even a mobile terminal such as a smartphone, equipped with the target distribution estimation program according to the present invention.

[0039] Furthermore, the processor memory includes a presence count distribution determination unit 111, an address / location expansion coefficient determination unit 112 including a nighttime staying terminal determination unit 112a, an address / location area determination unit 113, an expansion coefficient estimation unit 114, an object presence distribution determination unit 115, and a communication control unit 121. The functional components described above can be considered as functions realized by executing an object distribution estimation program stored in the processor memory. Also, the process flow shown by connecting the functional components of the communication facility device 1 in FIG. 1 with arrows can be understood as one embodiment of the object distribution estimation method according to the present invention.

[0040] Incidentally, at least one of the presence count distribution determination unit 111, the address / location expansion coefficient determination unit 112, the address / location area determination unit 113, the expansion coefficient estimation unit 114, and the object presence distribution determination unit 115 may be a functional component of another device, and the object distribution estimation function may be performed by a plurality of devices (e.g., a plurality of servers) as a whole. In this case, the plurality of devices as a whole constitutes the object distribution estimation system according to the present invention.

[0041] Hereinafter, a description of the communication control function (e.g., relay function) in the communication facility equipment 1 will be omitted, and only the target distribution estimation processing function according to the present invention will be described. Also in Fig. 1, the positioning result storage unit 102 acquires the terminal positioning result (GPS positioning data, etc.) of the above-mentioned terminal 2 from the terminal 2 via the communication interface unit 101 and the communication control unit 121, and stores and manages it. Here, this terminal positioning result is positioning data only for users who have contracted with the communication carrier and have consented to the provision of terminal position information, and is data only for a portion of people (users) out of the total population to be the target of the distribution to be estimated.

[0042] Furthermore, the address and location investigation result storage unit 103 acquires the address and location investigation result from the address and location investigation result management server 4 via the communication interface unit 101 and the communication control unit 121, and stores and manages it. Here, in this embodiment, the address and location investigation result is census data. The census data is not data from which it is possible to know, for example, the number of all people (the number of people present) in a certain mesh (area). However, according to the census data, it is possible to derive the number of all people whose address (or location) is within the mesh (area). This makes it possible to determine the "address and location expansion coefficient" described later.

[0043] <Existence distribution determining means> Similarly, in the functional block diagram of FIG. 1, in this embodiment, the presence distribution determination unit 111 receives the following from the positioning result storage unit 102: (a) terminal positioning results (positioning data) accumulated during a first predetermined period (e.g., one month); (b) A terminal positioning result (positioning data) acquired during a second predetermined period that is shorter than the first predetermined period and immediately precedes the distribution estimation time (for example, the last hour when the current time is the distribution estimation time), From this, the "distribution of numbers present at the time of estimation" is determined as the distribution of numbers present at the time of distribution estimation or immediately before the time of distribution estimation.

[0044] This estimated presence distribution is generated and determined from the positioning data of only a portion of users, not the entire population, but is close to the presence distribution at the time of distribution estimation (for example, the current time), and is the presence distribution to which the magnification coefficient should be multiplied after this. Here, since the positioning data around the time of distribution estimation (the current time) is too few as sample points, a certain period (for example, the most recent hour) is set as the second predetermined period according to the positioning interval and data amount.

[0045] In addition, the terminal positioning results accumulated over a long first predetermined period (for example, one month) are also used in determining the estimated presence distribution because, as described above, the terminal positioning results are not the entire population but only the positioning data of some users, and in particular in sparsely populated suburbs and rural areas, positioning data cannot be obtained at many points, resulting in the population at those points being zero. In other words, by setting a long first predetermined period (for example, one month), if it is possible to measure even one person per month, for example, it is possible to set the number of people per day to a finite value exceeding zero, such as about 0.03 (≒1 / 30). As a result, it is possible to more accurately grasp, for example, the shading of the number of people at each point. Incidentally, this can be said to be a similar method to trying to obtain a camera image with more accurate shading by lengthening the exposure time (light accumulation time) when taking a camera picture in the dark, and taking in more light into the camera sensor.

[0046] Here, in this embodiment, the presence distribution determination unit 111 regards the positioning data group accumulated in the first predetermined period and the positioning data group acquired in the second predetermined period as a point group in the presence distribution space, and uses Kernel Density Estimation (KDE), a known method for estimating the entire distribution from a finite number of sample points, to determine the estimated presence distribution C E Determine (x).

[0047] More specifically, the existence distribution determination unit 111 determines W A and W B The "W" is set appropriately in advance. A +W B = 1" and T A and T B are the length of the first predetermined period (e.g., 720 hours (= 24 hours / day × 30 days)) and the length of the second predetermined period (e.g., 1 hour), (a) A weight W is assigned to each piece of positioning data (=(latitude, longitude)) of the positioning data group (point cloud) accumulated during a first predetermined period. A / T A Multiply it by (b) A weight W is assigned to each point (positioning data (=(latitude, longitude))) of the positioning data group (point cloud) acquired during the second predetermined period. B / T B Multiply it by The weighted positioning data set (x1, x2, . . . , x n ) as sample points, and (1) C E (x)=P E (x)×V E (x) Here, P E (x)=(nh) -1 Σ i=1 n K((xx i ) / h) The estimated population distribution C E Calculate (x), where h is the bandwidth (smoothing parameter) and K is the kernel function, which can be a standard (zero mean, unit variance) Gaussian function as shown in Equation (2) below. (2) K(x')=(2π) -0.5 exp(-x' 2 / 2)

[0048] In addition, in the above formula (1), P E (x) is the probability density distribution of the presence of a human (target) in the first and second predetermined periods. E (x) is a volume determined using the number of positioning data groups acquired in a second predetermined period, and has a probability density value (P E This is the "conversion rate" when converting the value of (x) into the number of people (number of subjects). In other words, the distribution of the number of people present at the time of estimation C E (x) is the existence probability density distribution P E (x) in this volume V E This is multiplied by (x) and can be regarded as the distribution of the number of people (number of subjects) (related to the positioning data).

[0049] As a result, the estimated population distribution C EThe value of (x) (the number of people present at the time of estimation) is interpreted as the expected number of people (related to the positioning data) at point x. In other words, the distribution of the number of people present at the time of estimation C E According to (x), it is possible to derive the estimated number of people present at any point. E Since (x) is used to estimate the number of people (number of subjects) at the time of distribution estimation, in this embodiment, V (as a volume close to the time of distribution estimation) E It is calculated by multiplying by (x).

[0050] Incidentally, KDE is implemented as a standard function model in many numerical analysis and programming platforms such as MATLAB (registered trademark), which are used by many engineers and researchers in this field. Of course, if a function allows the estimation of the entire distribution from a finite number of sample points, that function may be used instead of KDE to determine the population distribution at the time of estimation.

[0051] In addition, it is also preferable that the presence distribution determination unit 111 sets, as the first predetermined period, a period type related to at least one of a preset month, date, day of the week, weekday / holiday, and time period, which corresponds to the period type to which the population distribution estimation time belongs. For example, if the population distribution estimation time is "between 8:00 on Wednesday", the first predetermined period may be set to "between 8:00 on Wednesday in the past M months (M is a positive integer)". This allows the presence distribution at the time of estimation to reflect circumstances specific to the period type, such as the number of people commuting to work or school being the largest in a certain area in the "between 8:00 on Wednesday" in the estimation time presence distribution, thereby making it possible to perform a more accurate (precise) population distribution estimation. Note that, for each required period type, a presence distribution at the time of estimation related to the period type may be prepared in advance.

[0052] Furthermore, the presence distribution determination unit 111 uses the positioning data x i The more recent the positioning time point of (i=1, 2, . . . , n) is within the first predetermined period, the more weight is assigned to this positioning data x i , and the estimated population distribution C EIt is also preferable to determine (x). Hereinafter, this embodiment will be described with reference to FIG. 2 (a schematic diagram for explaining another embodiment of the existence count distribution determination process).

[0053] As shown in FIG. 2, the presence distribution determination unit 111 determines the presence distribution C E When determining (x), it is also preferable to assign a larger weight to the positioning data as the positioning time (positioning time) of the positioning data becomes newer (as the distribution estimation time approaches). A / T A (multiplied by weight W B / T B The positioning data multiplied by is input to the KDE model to estimate the population distribution C E When generating and determining (x), the following equation is used for these positioning data: (3) W T =(tm-t0) / (t1-t0) The weight W calculated by T You may also input data that has been multiplied by this.

[0054] This makes it possible to obtain a distribution of the number of people present at the time of estimation that is closer to the distribution estimation time, and to improve the accuracy of the final population distribution estimation. Note that the weights are not limited to those in the above formula (3), and various other weights can be used as long as they are monotonically increasing functions of the positioning time (positioning time point) tm.

[0055] Furthermore, it is also preferable that, for each of the positioning data (terminal positioning result) accumulated in the first specified period and the positioning data (terminal positioning result) acquired in the second specified period, the presence count distribution determination unit 111 sets W presence points (sample points) corresponding to the positioning data, W (W is a positive integer) with "weights" assigned to the W presence points, the sum of which is 1, within an area (for example, a circular area with the positioning error as a radius) having a size that is a monotonically increasing function of the positioning error included in or determined from the positioning data, and determines the estimation-time presence count distribution from the W presence points (sample points).

[0056] Here, the positioning error may be the one that is usually given to the acquired positioning data. Alternatively, an appropriate positioning error may be determined in advance depending on whether the acquired positioning data is GPS positioning data, base station positioning data, or Wi-Fi (registered trademark) positioning data. Hereinafter, this embodiment will be described with reference to FIG. 3 (a schematic diagram for explaining yet another embodiment of the presence distribution determination process).

[0057] As shown in Fig. 3, in this embodiment, the presence count distribution determination unit 111 sets a circular area for one piece of positioning data, with the position (latitude, longitude) of the data as the center and the positioning error as the radius, and generates and arranges W presence points (sample points) each assigned with a weight of 1 / W within this circular area (for example, generates and arranges 100 presence points with a weight of 0.01). Here, it is also preferable to arrange the set W presence points (sample points) so that the closer they are to the center of the circular area, the higher the density. Furthermore, it is also preferable to assign weights such that the values ​​of each weight are not uniform, but that the closer to the center of the circular area, the higher the weight is assigned (the sum is 1).

[0058] In any case, when the positioning data group is subsequently input into the KDE model to determine the estimated presence distribution, it is possible to increase the precision (accuracy) of the determined estimated presence distribution by increasing the number of highly reliable sample points. Here, the smaller the positioning error, the smaller the correct sample points are considered to be distributed within a smaller circular area (further closer to the center), so by increasing the number of sample points according to the above-mentioned method, it is expected that the precision (accuracy) of the estimated presence distribution will be improved. Note that the presence points (sample points) within the circular area can be set by known sampling methods other than those mentioned above.

[0059] In addition, the existence number distribution determination unit 111 (a) For each of a plurality of sub-periods (e.g., three one-hour periods in the most recent three hours) included in the second predetermined period (e.g., the most recent three hours), from the positioning data (terminal positioning results) obtained in the sub-period, determine the "sub-period occupancy distribution" as the occupancy distribution at the representative time point (e.g., the midpoint of the sub-period) of the sub-period. (b) Extrapolate the "sub-period occupancy distribution" at the representative time point to the occupancy distribution at the distribution estimation time point, and determine the estimated occupancy distribution as the occupancy distribution obtained by the extrapolation. This is also preferable. In the above (a), it is also possible to set the plurality of sub-periods from among the first predetermined period (e.g., one month) and the second predetermined period. Hereinafter, this embodiment will be described with reference to FIG. 4 (a schematic diagram for explaining another embodiment of the occupancy distribution determination process).

[0060] In the example shown in FIG. 4, a first sub-period including the distribution estimation time point t(0) and having a representative time point (the midpoint of the sub-period in FIG. 4) of t(-1) (<t(0)) and a second sub-period having a representative time point of t(-2) (<t(-1)) are set. Here, the occupancy distribution CE -2 (x) in the second sub-period is generally data at time point t(-2), while the occupancy distribution CE -1 (x) in the first sub-period is considered to be generally data at time point t(-1). Therefore, the occupancy distribution at the distribution estimation time point t(0) is approximately calculated from the occupancy distributions at these two time points using the extrapolation process, and this is taken as the estimated occupancy distribution C E (x).

[0061] Specifically, for example, using the following equation (4) C E (x) = CE -1 (x) + (t(0) - t(-1)) × ((CE -1 (x) - CE -2 (x)) / (t(-1) - t(-2))) to determine the estimated occupancy distribution C EOf course, even if there are three or more sub-periods, the estimated population distribution C can be generated in the same manner by known extrapolation processing. E (x) is generated. It is also possible to use a higher order extrapolation process (such as spline extrapolation) rather than a linear extrapolation process. In any case, this extrapolation process produces a distribution of the number of people present at the time of estimation C that is closer to the distribution estimation time t(0). E (x) can be obtained, which can improve the accuracy of the final population distribution estimate.

[0062] Here, the above-described existence count distribution determination process in existence count distribution determination unit 111 of this embodiment will be summarized. Fig. 5 is a schematic diagram showing an outline of one embodiment of the object distribution estimation method according to the present invention.

[0063] As shown in FIG. 5, in this embodiment, the presence distribution determination unit 111 uses KDE to determine the presence distribution C E (x) is determined. Here, the distribution of the number of people present at the time of estimation C E (x) is then used in the object presence distribution determining unit 115 to generate object presence distribution data (eg, a high-resolution population distribution graph or a high-resolution population heat map).

[0064] <Method of determining address / location expansion coefficient> Returning to the functional block diagram of Figure 1, in this embodiment, the night-time staying terminal determination unit 112a of the address / location expansion coefficient determination unit 112 determines, for each mesh into which a specific area (e.g. the entire country) is divided, "the number of terminals 2 staying within the mesh" during a set time period related to the address or location, which in this embodiment is "at night" from 1:00 to 5:00, based on historical information of positioning data (terminal positioning results).

[0065] Here, with regard to "the number of terminals 2 staying within the mesh," for example, a publicly known movement / stay determination method (*) can be used to generate time series data of user movement and stay from historical information of positioning data, and for each user, the mesh in which the user has spent the longest time staying (the mesh with the highest probability of staying) during the set "nighttime" is identified, and the number of terminals 2 for which the identified mesh is the mesh is found, and this number can be regarded as "the number of terminals 2 staying within the mesh." (*) For example, the method disclosed in non-patent document: Nao Kobayashi, Hiroki Ishizuka, Atsunori Minamikawa, Shigeki Muramatsu, Tomohiro Ono, “Proposal of a method for estimating movement and stay status suitable for mobile phone communication history,” IPSJ Transactions on Information Processing Society of Japan, Database, vol. 10, no. 1, pp.13-23, 2017, can be used.

[0066] As a modification, the nighttime staying terminal determination unit 112a can obtain registered information (e.g., contract information in a contract with a telecommunications carrier) of the person (terminal user) to whom the terminal 2 is attached, and determine the "number of terminals 2 that will (will) be staying in the mesh (at night)" based on this. For example, the number of terminals 2 whose registered addresses are within the mesh may be determined as this number.

[0067] Based on the number of terminals 2 determined in this manner and the acquired census data (address and location survey results), the address / location expansion coefficient determination unit 112 determines the address / location expansion coefficient as an expansion coefficient for people (targets) whose address or location area (address / location mesh) is the mesh.

[0068] Specifically, for each mesh (representative point (e.g., center point) of the mesh) x, the number of terminals 2 determined as S(x) and the number of people (survey result number of people) N(x) whose address or location is within the mesh x determined from the census data are used to calculate the following equation: (5) r(x)=N(x) / S(x) The address / location expansion coefficient r(x) for mesh x can be calculated by the above. In other words, the ratio of the number of people N(x) in the survey obtained from census data that can grasp the number of people (population) in mesh x to the number of terminals 2 during "nighttime" S(x) calculated from positioning data that can only grasp the positions of some users is calculated, and this is used as the address / location expansion coefficient r(x).

[0069] Here, the meshes are set to a high granularity that can be set using census data, for example, a rectangle of 250m x 250m, with a view to ultimately generating high-resolution population distribution data. When a given area (nationwide) is divided into n meshes (x1, x2, . . . , xn) of this high granularity, the address / location expansion coefficient r of all meshes can be expressed as r = (r(x1), r(x2), . . . , r(xn)).

[0070] Above, we have explained the address / location expansion coefficient, which is, so to speak, an expansion coefficient specific to an area (mesh). In general, the expansion coefficient is a value related to the ratio of "users who have contracted with a telecommunications carrier and have consented to the provision of their terminal location information" to the total population. Whether or not to "contract with a telecommunications carrier" and "consent to the provision of terminal location information" is essentially up to each individual's way of thinking and judgment, but it is thought that there is a certain degree of regionality (tendencies depending on the area where a person lives, for example, so-called prefectural characteristics).

[0071] For example, at the prefecture level, it may be possible that residents of prefecture A have a high approval rate, but residents of prefecture B have a low approval rate. In this case, the address / location expansion coefficient determination unit 112 seeks a coefficient with higher precision (accuracy), and determines, for example, the address / location expansion coefficient for each prefecture = (expansion coefficient for prefecture A, expansion coefficient for prefecture B, . . .) (Note that the prefecture level is merely an example, and in this embodiment, as described above, a mesh with a higher granularity is set).

[0072] <Method for determining address / location area> Also in the functional block diagram of FIG. 1, the address / location area determination unit 113 identifies terminals 2 that are present within a predetermined range including a plurality of points within a predetermined area (nationwide), which in this embodiment is a large number of points to accommodate high resolution, from the positioning data (terminal positioning results), (for example, a circular area with a predetermined radius centered on the point), and (a) from history information of the positioning data (terminal positioning result) for the identified terminal 2, or (b) from registered information of the person (terminal user) to whom the identified terminal 2 is attached, The "address / location mesh" (address / location area) is determined as the address or location of the person (subject) to whom the identified terminal 2 is attached.

[0073] Here, the address / location mesh can be determined from the history information in (a) above in the same manner as the nighttime staying terminal determination process in the nighttime staying terminal determination unit 112a described above. That is, time-series data on movement and stay for the identified terminal 2 (located within a predetermined range including the point) may be generated, and the mesh with the longest stay time (highest stay probability) during the set "nighttime" may be determined as the address / location mesh.

[0074] On the other hand, the determination of the "address / location area" from the registered information in (b) above can be done, for example, by acquiring the subscriber information (in the contract with the telecommunications carrier) of the user of the identified terminal 2, and determining the address / location mesh based on this. For example, the mesh including the registered address can be set as the address / location mesh.

[0075] For convenience in estimating the expansion coefficient later, the mesh group from which the address / location mesh is determined is set in this embodiment to be the same as the mesh group set by the address / location expansion coefficient determination unit 112 described above (for example, a rectangular mesh group of 250 m x 250 m).

[0076] <Means for determining magnification coefficient> Also in the functional block diagram of Figure 1, the expansion coefficient estimation unit 114 calculates, based on the address / location mesh information determined by the address / location area determination unit 113, for each of the multiple (many) set locations, the "proportion of each of the multiple (many) meshes as an address / location mesh" for the person (terminal user) to whom the identified terminal 2 is attached, and estimates the "expansion coefficient" of the location based on this proportion and the determined address / location expansion coefficient.

[0077] Specifically, in this embodiment, when n meshes (x1, x2, . . . , xn) are set that divide a specific region (nationwide) at a high granularity, the expansion coefficient estimation unit 114 calculates the following: (a) Calculate the proportion h(xi) (i=1, 2, . . . , n) of users of terminal 2 identified at point x (those existing within a specified range including point x) whose addresses / locations (mesh) are xi (i=1, 2, . . . , n), (b) Determine the distribution of addresses or locations (breakdown of addresses or locations) H(x) = (h(x1), h(x2), ..., h(xn)) of users of terminal 2 identified at point x (located within a specified range including point x). Here, the distribution H(x) in (b) above is, so to speak, information about where (in what mesh) a user of terminal 2 near point x lives (is located).

[0078] Next, the expansion coefficient estimation unit 114 weights the address / location expansion coefficient r(xi) (i=1, 2, . . . , n) for each point x in the multiple (large number) points set, using a ratio h(xi) (i=1, 2, . . . , n), and sums them across the (n) address / location meshes, and determines the result of the summation as the expansion coefficient for point x. That is, (c) For each point x, the magnification coefficient R(x) for that point x is calculated using the following formula: (6) R(x)=H(x)*r =h(x1)×r(x1)+h(x2)×r(x2)+···+h(xn)×r(xn) It is calculated by:

[0079] Here, the proportion h(xi) is, more specifically, the proportion of people at point x who have an address (place of residence) of xi. The address / location expansion coefficient r(xi) indicates that if there is one person at point x who has an address (place of residence), then there will actually be r(xi) people there. As a result, R(x) in the above formula (8) can be considered a more accurate expansion coefficient that takes into account the address (place of residence) of each person (each terminal 2) at point x.

[0080] In this embodiment, the magnification coefficient estimation unit 114 comprehensively estimates the magnification coefficient (magnification coefficient map) R(X) for all points X=(x1, x2, . . . , xm) in a specified region (nationwide) that are to be made high-resolution. This makes it possible to generate highly accurate population distribution data, for example, a high-resolution population heat map, as will be described later. Note that all points X=(x1, x2, . . . , xm) can also be a set of representative points (for example, central points) of each mesh in a high-resolution population heat map (in that case, m=n (the number of meshes)).

[0081] <Method of determining target presence distribution> Similarly, in the functional block diagram of FIG. 1, the object presence distribution determination unit 115 determines the presence distribution C E Based on (x) and the expansion coefficient R(x) estimated by the expansion coefficient estimation unit 114, the number of people (objects) present at each of multiple points within a specified area (e.g., the entire country), i.e., the number of people (population), is determined.

[0082] Specifically, in this embodiment, the target presence distribution determination unit 115 calculates the number of people C(x) at each point x among the multiple (large number) points set, using the following formula: (7) C(x)=C E (x) × R(x) In this way, the target presence distribution determination unit 115 calculates the presence number distribution C at the time of estimation, which is a number of people determined from the positioning data of some users but is capable of calculating the number of people at any point x.E The number of people C(x) at any point x can be determined by multiplying (x) by the magnification factor R(x) at that point x, which represents the ratio between the total number of people and the number of users for which positioning data is obtained.

[0083] The target presence distribution determination unit 115 can also calculate and determine the number of people distribution C(x) for any (multiple) points x in a specific region (for example, the whole country) to generate, for example, a high-resolution population distribution graph. It is also possible to calculate and determine the number of people distribution C(X) for all points to be highly resolved, for example, the representative point (for example, the center point) X=(x1, x2, . . . , xn) of each mesh to generate, for example, a high-resolution population heat map.

[0084] The population distribution data (number of people distribution C(X)), high-resolution population distribution graph, and high-resolution population heat map generated by the target presence distribution determination unit 115 may be displayed on the user interface unit 104 having a display unit and presented to an operator / manager of the device 1. Also, they may be transmitted to an external information processing device via the communication control unit 121 and the communication interface unit 101 and used therein.

[0085] Here, we briefly summarize the processing in the address / location expansion coefficient determination unit 112, address / location area determination unit 113, expansion coefficient estimation unit 114, and target presence distribution determination unit 115 of this embodiment described above. As shown in Fig. 5, the address / location expansion coefficient determination unit 112 derives an address / location expansion coefficient for each mesh using positioning data (terminal position information) and census data (address / location survey results).

[0086] Next, as also shown in FIG. 5, the address / location area determination unit 113 determines address / location mesh information for terminals 2 present at each point (a predetermined range including each point) from the history information of the positioning data (terminal location information). Next, the expansion coefficient estimation unit 114 estimates an expansion coefficient R(x) (expansion coefficient map R(X)) at each point using the "proportion of each mesh as an address / location mesh" calculated based on the address / location mesh information and the address / location expansion coefficient. Finally, the target presence distribution determination unit 115 estimates the expansion coefficient R(x) (expansion coefficient map R(X)) at each point based on the expansion coefficient R(x) and the presence number distribution C at estimation time determined by the presence number distribution determination unit 111. E Using (x), the number of people C(x) at any (many) points x is calculated and determined, and from this, a high-resolution population distribution graph or a high-resolution population heat map is generated.

[0087] As explained in detail above, in the present invention, an address / location expansion coefficient is calculated using the results of an address / location survey, and this address / location expansion coefficient and the address / location area of ​​targets attached to terminals 2 that exist within a specified range including each point are used to estimate the expansion coefficient for each point, thereby determining the number of targets present at each point.

[0088] Here, even if the correct number of people at each target point (each area) is not obtained, the address / location expansion coefficient is calculated using the address / location survey results, so the expansion coefficient for each point can be obtained. In addition, this expansion coefficient has a higher precision (accuracy) reflecting the characteristics (that affect the actual number of targets) at that point. As a result, according to the present invention, it is possible to estimate the distribution of targets with high resolution with higher precision (accuracy) without relying on correct data related to the actual distribution of targets (people), and even without relying on real number measurement means such as smart poles scattered around the country or machine learning models.

[0089] Furthermore, for example, by applying the object distribution estimation method according to the present invention, accurate and detailed prediction of people flow and even traffic behavior in urban areas can be performed, and resilient traffic and transportation infrastructure can be developed, and efficient urban planning can be implemented to solve the problems caused by urbanization. In other words, according to the present invention, it is possible to contribute to Goal 9 "Develop resilient infrastructure, promote sustainable industrialization and foster innovation" and Goal 11 "Make cities inclusive, safe, resilient and sustainable" of the Sustainable Development Goals (SDGs) led by the United Nations.

[0090] With respect to the various embodiments of the present invention described above, various changes, modifications, and omissions within the scope of the technical ideas and viewpoints of the present invention can be easily made by those skilled in the art. The above description is merely an example of the embodiments, and does not impose unnecessary restrictions. The present invention is limited only by the scope of the claims and their equivalents. [Explanation of symbols]

[0091] 1. Communication equipment (target distribution estimation device) 101 Communication interface section 102 Positioning result storage unit 103 Address / Location Investigation Results Storage Department 104 User Interface Section 111 Existence distribution determination part 112 Address and location expansion coefficient determination unit 113 Address / location area determination department 114 Magnification factor estimation unit 115 Target Presence Distribution Determination Unit 121 Communication control section 2. Terminal 3 base station 4 Address / location search result management server

Claims

1. A target distribution estimation program for causing a computer to function to estimate a distribution of a plurality of targets in a predetermined area, comprising: an address / location expansion coefficient determination means for determining, for each zone within the specified region, the number of terminals staying in the zone during a time period related to an address or location based on history information of positioning results for terminals attached to the target or registration information for the terminal, and for determining an address / location expansion coefficient as an expansion coefficient for the target whose address or location zone is the zone based on the number of terminals and the acquired survey results for the target's address or location; an address / location area determination means for determining, for each point within the specified area, a terminal related to the point from the positioning result for the terminal attached to the target, and determining, from registration information for the identified terminal or history information of the positioning result, an address / location area as the address or location of the target to which the identified terminal is attached; an expansion coefficient estimation means for calculating, for each said point, a proportion of said area in the target to which said specified terminal is attached that is an address / location area, and estimating an expansion coefficient of said point based on said proportion and said address / location expansion coefficient; a presence distribution determining means for determining a presence distribution as a distribution of the number of objects present in the predetermined area from a positioning result obtained during a predetermined period for a terminal attached to the object, the presence distribution relating to an estimation time point of the distribution of the plurality of objects; an object presence distribution determining means for determining the number of objects present at each of a plurality of points within the predetermined area based on the presence distribution and the estimated expansion coefficient; A target distribution estimation program that causes a computer to function as a

2. The target distribution estimation program described in claim 1, characterized in that the expansion coefficient estimation means weights the address / location expansion coefficient for each point by the relevant ratio, then adds up the weights across the address / location area, and determines the result of the addition as the expansion coefficient for the point.

3. the presence number distribution determination means determines an estimation time presence number distribution as a distribution of the numbers present at or immediately before the estimation time point of the distribution from the positioning results accumulated during a first predetermined period and the positioning results acquired during a second predetermined period that is shorter than the first predetermined period and immediately before the estimation time point of the distribution; The target presence distribution determining means determines the number of targets present at each of a plurality of points within the specified area based on the estimated presence distribution and the estimated expansion coefficient.

3. The target distribution estimation program according to claim 1 or 2.

4. The object distribution estimation program according to claim 3, characterized in that the presence distribution determination means sets, as the first predetermined period, a period type related to a preset month, day, day of the week, weekday / holiday, and / or time zone, and a period corresponding to a period type to which the estimation time point of the distribution belongs.

5. The object distribution estimation program according to claim 3 or 4, characterized in that the presence distribution determination means determines the presence distribution at the time of estimation by assigning a greater weight to the positioning result the more recent the positioning time point related to the positioning result is within the first specified period and the second specified period.

6. The object distribution estimation program according to any one of claims 3 to 5, characterized in that, for each positioning result acquired during the first specified period and the second specified period, the presence distribution determination means sets at least one presence point corresponding to the positioning result, within an area having a size that is a monotonically increasing function of the positioning error included in the positioning result or determined from the positioning result, the presence point being weighted so that the sum of the weighted points is 1, and determines the presence distribution at the time of estimation from the presence point.

7. The abundance distribution determining means determining a subperiod presence number distribution as a distribution of the number of objects present at a representative time point of each subperiod from the positioning result acquired in the subperiod for each subperiod among a plurality of subperiods included in the first predetermined period and the second predetermined period; The subperiod abundance distribution at the representative time is extrapolated to the abundance distribution at the estimation time, and the abundance distribution obtained by the extrapolation is determined as the abundance distribution at the estimation time.

7. The target distribution estimation program according to claim 3,

8. An object distribution estimation device that estimates a distribution of a plurality of objects in a predetermined area, comprising: an address / location expansion coefficient determination means for determining, for each zone within the specified region, the number of terminals staying in the zone during a time period related to an address or location based on history information of positioning results for terminals attached to the target or registration information for the terminal, and for determining an address / location expansion coefficient as an expansion coefficient for the target whose address or location zone is the zone based on the number of terminals and the acquired survey results for the target's address or location; an address / location area determination means for determining, for each point within the specified area, a terminal related to the point from the positioning result for the terminal attached to the target, and determining, from registration information for the identified terminal or history information of the positioning result, an address / location area as the address or location of the target to which the identified terminal is attached; an expansion coefficient estimation means for calculating, for each said point, a proportion of said area in the target to which said specified terminal is attached that is an address / location area, and estimating an expansion coefficient of said point based on said proportion and said address / location expansion coefficient; a presence distribution determining means for determining a presence distribution as a distribution of the number of objects present in the predetermined area from a positioning result obtained during a predetermined period for a terminal attached to the object, the presence distribution relating to an estimation time point of the distribution of the plurality of objects; an object presence distribution determining means for determining the number of objects present at each of a plurality of points within the predetermined area based on the presence distribution and the estimated expansion coefficient; A target distribution estimation device comprising:

9. An object distribution estimation system for estimating a distribution of a plurality of objects in a predetermined area, comprising: an address / location expansion coefficient determination means for determining, for each zone within the specified region, the number of terminals staying in the zone during a time period related to an address or location based on history information of positioning results for terminals attached to the target or registration information for the terminal, and for determining an address / location expansion coefficient as an expansion coefficient for the target whose address or location zone is the zone based on the number of terminals and the acquired survey results for the target's address or location; an address / location area determination means for determining, for each point within the specified area, a terminal related to the point from the positioning result for the terminal attached to the target, and determining, from registration information for the identified terminal or history information of the positioning result, an address / location area as the address or location of the target to which the identified terminal is attached; an expansion coefficient estimation means for calculating, for each said point, a proportion of said area in the target to which said specified terminal is attached that is an address / location area, and estimating an expansion coefficient of said point based on said proportion and said address / location expansion coefficient; a presence distribution determining means for determining a presence distribution as a distribution of the number of objects present in the predetermined area from a positioning result obtained during a predetermined period for a terminal attached to the object, the presence distribution relating to an estimation time point of the distribution of the plurality of objects; an object presence distribution determining means for determining the number of objects present at each of a plurality of points within the predetermined area based on the presence distribution and the estimated expansion coefficient; A target distribution estimation system comprising:

10. 1. An object distribution estimation method implemented by a computer for estimating a distribution of a plurality of objects in a predetermined area, comprising: For each area within the specified area, determining the number of terminals staying in the area during a time period related to an address or location based on historical information of the positioning results for the terminal attached to the target or registration information for the terminal, and determining an address / location expansion coefficient as an expansion coefficient for the target whose address or location area is the area based on the number of terminals and the acquired survey results of the address or location of the target, and also identifying a terminal related to each point within the specified area from the positioning results for the terminal attached to the target, and determining an address / location area as the address or location of the target to which the identified terminal is attached based on the registration information or historical information of the positioning results for the identified terminal; A step of calculating, for each said location, a proportion of the target to which the specified terminal is attached that is an address / location area of ​​each of said areas, and estimating an expansion coefficient of said location based on said proportion and said address / location expansion coefficient; A step of determining a presence distribution as a distribution of the number of objects present in the predetermined area from the positioning results obtained during a predetermined period of time for the terminal attached to the object, the presence distribution relating to an estimation time point of the distribution of the plurality of objects; determining a number of the objects at each of a plurality of locations within the predetermined area based on the number distribution and the estimated expansion factor; A method for estimating a distribution of an object, comprising:

Images