CN110888979A

CN110888979A - Interest region extraction method and device and computer storage medium

Info

Publication number: CN110888979A
Application number: CN201811052608.3A
Authority: CN
Inventors: 曾瑞; 邵波; 孙芳杰
Original assignee: China Mobile Communications Group Co Ltd; China Mobile Group Heilongjiang Co Ltd
Current assignee: China Mobile Communications Group Co Ltd; China Mobile Group Heilongjiang Co Ltd
Priority date: 2018-09-10
Filing date: 2018-09-10
Publication date: 2020-03-17

Abstract

The embodiment of the invention discloses a method for extracting an interest area, which comprises the following steps: determining a quasi-staying point according to an acquisition point of the movement track information for representing the target object; according to the track characteristic parameters of the acquisition points corresponding to the quasi-staying points, determining purification parameters corresponding to the quasi-staying points, and purifying the quasi-staying points according to the relationship between the purification parameters and the set conditions to obtain the staying points; and clustering the stop points, and obtaining the interest area of the target object according to the clustering result. Meanwhile, the embodiment of the invention also discloses an interest region extraction device and a computer storage medium.

Description

Interest region extraction method and device and computer storage medium

Technical Field

The invention relates to the technical field of big data, in particular to a method and a device for extracting an interest region and a computer storage medium.

Background

With the development of mobile internet and big data technology, data of operators are gradually complicated, user requirements are diversified, and traditional operation ideas cannot meet competitive requirements. Therefore, the business-centered service is shifted to the user-centered service, a mature and complete user label system is constructed, user characteristics are accurately described, all-round analysis and understanding of user requirements are realized, differentiated information services are provided for the user, and the user insight value is improved. In the big data era, operators are under business transformation pressure, and are gradually changed from a single communication service provider to a diversified information service provider. Therefore, the industry chain and the value chain of the operators also need to be richer, and users become the core in the new value chain. In the process of building a user tag system, besides basic feature tags such as consumption capacity, social attributes, social behaviors, terminal features, position features, internet behaviors, temperament and the like of a user need to be described, the space-time position track information of the user actually records objective activity behaviors of the user, important information such as intention, activity rules and potential interests and hobbies of the user can be continuously mined from the behavior activities, and the method has important significance in aspect of behavior insights and value change of the user. However, in the related art, there is a problem that the accuracy is not high for the extraction of the region of interest.

Disclosure of Invention

In view of this, embodiments of the present invention provide a method and an apparatus for extracting an interest region, and a computer storage medium, which can improve accuracy.

In order to achieve the above purpose, the technical solution of the embodiment of the present invention is realized as follows:

in a first aspect, an embodiment of the present invention provides a method for extracting a region of interest, where the method includes:

determining a quasi-staying point according to an acquisition point of the movement track information for representing the target object;

according to the track characteristic parameters of the acquisition points corresponding to the quasi-staying points, determining purification parameters corresponding to the quasi-staying points, and purifying the quasi-staying points according to the relationship between the purification parameters and the set conditions to obtain the staying points;

and clustering the stop points, and obtaining the interest area of the target object according to the clustering result.

In a second aspect, an embodiment of the present invention provides an apparatus for extracting a region of interest, including:

the extraction module is used for determining a quasi-stop point according to an acquisition point of the movement track information for representing the target object;

the purification module is used for determining purification parameters corresponding to the quasi-staying points according to the track characteristic parameters of the collection points corresponding to the quasi-staying points, and purifying the quasi-staying points according to the relationship between the purification parameters and the set conditions to obtain the staying points;

and the clustering module is used for clustering the stop points and obtaining the interest area of the target object according to the clustering result.

In a third aspect, an embodiment of the present invention provides an interest region extraction apparatus, including: a processor and a memory for storing a computer program operable on the processor, wherein the processor is configured to implement the region of interest extraction method of the first aspect when the computer program is executed.

In a fourth aspect, an embodiment of the present invention provides a computer storage medium, which stores a computer program, and when the computer program is executed by a processor, the method for extracting a region of interest according to the first aspect is implemented.

According to the interest region extraction method, device and computer storage medium provided by the embodiment of the invention, after the quasi-staying point is determined according to the acquisition point for representing the moving track information of the target object, the quasi-staying point is purified based on the relation between the purification parameter corresponding to the quasi-staying point and the setting condition, the purified staying point is clustered, and the interest region of the target object is obtained according to the clustering result. Therefore, the quasi-staying points determined based on the acquisition points representing the movement track information of the target object are purified, so that the interfered staying point data can be reduced, the abnormal staying points influencing the accuracy of the interest area are deleted, the staying points obtained after purification are clustered, and the accuracy of the obtained interest area of the target object is improved.

Drawings

FIG. 1 is a schematic flow chart illustrating a method for extracting a region of interest according to an embodiment of the present invention;

FIG. 2 is a schematic structural diagram of an interesting region extracting apparatus according to an embodiment of the present invention;

FIG. 3 is a schematic structural diagram of an interesting region extracting apparatus according to another embodiment of the present invention;

FIG. 4 is a schematic view of an acquisition site in accordance with an embodiment of the present invention;

FIG. 5 is a schematic diagram of extracting a stop point based on an acquisition point according to an embodiment of the present invention;

FIG. 6 is a schematic diagram of extracting quasi-stop points based on acquisition points in an embodiment of the present invention;

FIG. 7 is a schematic diagram of the extraction of quasi-stop points based on acquisition points in another embodiment of the present invention;

fig. 8 is a schematic diagram of stay rates corresponding to different travel modes in an embodiment of the present invention;

fig. 9 is a schematic diagram of an angle corresponding to the acquisition point in an embodiment of the present invention.

Detailed Description

The present invention will be described in further detail with reference to the accompanying drawings and specific embodiments.

Example one

Referring to fig. 1, a method for extracting a region of interest provided in an embodiment of the present invention includes the following steps:

step S101: determining a quasi-staying point according to an acquisition point of the movement track information for representing the target object;

here, the target object may be an object for characterizing movement of a target person, such as a mobile phone that may be the target person, or the like directly. The moving track information of the target object represents the track generated by the target object in the moving process, and by monitoring or collecting the moving process of the target object, the moving track of the target object can be described based on the track characteristic parameters of the collecting points, wherein the track characteristic parameters comprise time, geographical position and the like. The time is the time when the moving process of the target object is collected, namely the time when a collection point is obtained; the geographic position is the position of the target object when the moving process of the target object is collected. The geographic position can be identified by the existing longitude and latitude coordinate mode, and can also be identified by the set region position. Therefore, the geographical position of the target object which arrives or stops in the historical time can be known through the acquisition point. The acquisition points can be acquired in real time, that is, the moving process of the target object can be monitored or acquired in real time. Because the movement track of the target object is temporal, that is, the target object usually moves from one geographical location to another geographical location within a period of time, the acquisition points are sequentially arranged according to the chronological order, and can be used for representing the movement track information of the target object. It should be noted that there are a plurality of acquisition points for characterizing the movement trajectory information of the target object, and the acquisition points can be represented by an acquisition point set.

In an optional embodiment, the determining a quasi-stop point according to an acquisition point of movement trajectory information for characterizing a target object includes:

determining an initial acquisition point;

taking the acquisition point which is more than a set distance threshold value, more than a set time interval threshold value and closest to the initial acquisition point in relative distance as a target acquisition point;

and determining corresponding quasi-staying points according to the initial acquisition points and the acquisition points between the initial acquisition points and the target acquisition points.

Here, for each quasi-staying point, there are two or more acquisition points corresponding to the quasi-staying point, so it can also be said that the quasi-staying point corresponds to one acquisition point set, and the acquisition point set includes a plurality of acquisition points sequentially arranged according to a chronological order, and the quasi-staying point is determined according to the acquisition point set. The initial acquisition point is determined as a first acquisition point corresponding to each quasi-stop point, so that the quasi-stop point is acquired by taking the first acquisition point as a starting point. Taking the acquisition of a first quasi-staying point as an example, taking a first acquisition point arranged in time sequence in all acquisition points as an initial acquisition point, sequentially calculating the distance and the time interval between the first acquisition point and the acquisition points such as a second acquisition point and a third acquisition point, taking the acquisition point which is more than a set distance threshold value, more than a set time interval threshold value and closest to the first acquisition point in relative distance as a target acquisition point, determining the first quasi-staying point according to the first acquisition point and the acquisition point between the first acquisition point and the target acquisition point, and taking the target acquisition point as the first acquisition point corresponding to the second quasi-staying point. Because the distance between the acquisition points corresponding to each quasi-staying point is smaller than the set distance threshold, the acquisition points corresponding to each quasi-staying point are considered to be close to each other in the geographic position, and each quasi-staying point can be represented by the acquisition point corresponding to the quasi-staying point. For example, when the trajectory characteristic parameters of the acquisition points include time and geographic position, and the geographic position is latitude and longitude, the average latitude and average longitude of all acquisition points corresponding to the quasi-stop point may be used as the geographic position of the quasi-stop point, the difference between the time of the last acquisition point and the time of the first acquisition point in all acquisition points corresponding to the quasi-stop point may be used as the duration span of the quasi-stop point, the time of the first acquisition point may be used as the start time of the quasi-stop point, and the time of the last acquisition point may be used as the end time of the quasi-stop point. It should be noted that all the acquisition points corresponding to each quasi-staying point are sequentially arranged according to a time sequence, the first acquisition point is the acquisition point with the most advanced time among all the acquisition points corresponding to the quasi-staying point, that is, the acquisition point with the farthest time interval from the current time among all the acquisition points corresponding to the quasi-staying point, and the last acquisition point is the acquisition point with the most advanced time among all the acquisition points corresponding to the quasi-staying point, that is, the acquisition point with the closest time interval from the current time among all the acquisition points corresponding to the quasi-staying point. The target acquisition point may be an acquisition point that is adjacent to the initial acquisition point or may be an acquisition point that is non-adjacent to the initial acquisition point. The distance between any two acquisition points can be obtained according to the geographic positions of the two acquisition points, and the time interval between any two acquisition points can be obtained according to the time of the two acquisition points. The distance threshold may be set according to actual requirements, for example, set to 5 meters or 7 meters. The time interval threshold can be set according to actual needs, such as 4 minutes or 10 minutes. Since there may be a plurality of acquisition points having a distance from the initial acquisition point greater than a set distance threshold and a time interval greater than a set time interval threshold, an acquisition point satisfying the above condition and having the closest relative distance from the initial acquisition point is taken as a target acquisition point. For example, if, of five acquisition points a, b, c, d, and e, if a is taken as an initial acquisition point, the distance between a and b and the distance between a and c are both smaller than a set distance threshold, the distances between a, d, and e are respectively greater than the set distance threshold, and the distance between a and d is smaller than the distance between a and e, then d is taken as a target acquisition point, a corresponding quasi-stop point is determined according to the three acquisition points a, b, and c, and d is taken as an updated initial acquisition point to calculate the next quasi-stop point.

Therefore, the acquisition points of the movement track information for representing the target object are preprocessed according to the distance threshold and the time interval threshold, so that the quasi-stop points with the quantity less than that of the acquisition points are extracted, the processing speed is increased, and the accuracy of determining the stop points according to the quasi-stop points in the follow-up process is improved.

In an optional embodiment, after determining the corresponding quasi-stagnation point according to the initial acquisition point and the acquisition point between the initial acquisition point and the target acquisition point, the method further includes:

taking the target acquisition point as an updated initial acquisition point;

taking the acquisition point which is more than a set distance threshold value, more than a set time interval threshold value and closest to the initial acquisition point in relative distance as an updated target acquisition point;

and determining a corresponding quasi-staying point according to the initial acquisition point and the acquisition point between the initial acquisition point and the target acquisition point.

Here, based on the five acquisition points a, b, c, d, and e in the above embodiment as an example, if a is taken as an initial acquisition point and a corresponding target acquisition point is d, d is taken as an updated initial acquisition point, and thus, a quasi-stop point corresponding to d as a first acquisition point is continuously acquired. Therefore, the speed of acquiring the quasi-staying point is improved by updating the initial acquisition point in time.

Step S102: according to the track characteristic parameters of the acquisition points corresponding to the quasi-staying points, determining purification parameters corresponding to the quasi-staying points, and purifying the quasi-staying points according to the relationship between the purification parameters and the set conditions to obtain the staying points;

here, the trajectory characteristic parameters include time and geographic location, and the refining parameters include at least one of: average speed, stay rate, speed change rate and angle change rate. The average speed refers to the average moving speed of the target object in the acquisition points corresponding to the quasi-staying points, and the staying rate is the ratio of the number of the acquisition points of which the speed in the acquisition points corresponding to the quasi-staying points is less than a set speed threshold value to the number of the acquisition points corresponding to the quasi-staying points, or the ratio of the number of the acquisition points of which the speed in the acquisition points corresponding to the quasi-staying points is less than the set speed threshold value to the moving distance corresponding to the quasi-staying points; the speed change rate is the ratio of the number of the acquisition points, corresponding to the quasi-stop point, of which the speed change rate is greater than a set first speed change rate threshold value, in the acquisition points, to the number of the acquisition points, corresponding to the quasi-stop point, or the ratio of the number of the acquisition points, corresponding to the quasi-stop point, of which the speed change rate is greater than a set first speed change rate threshold value, in the acquisition points, to the moving distance, corresponding to the quasi-stop point; the angle conversion rate is the ratio of the number of the acquisition points of which the inner angles are larger than a set angle threshold value and the number of the acquisition points except the first and the last acquisition points. And the moving distance corresponding to the quasi-staying point is the distance accumulation sum between the acquisition points corresponding to the quasi-staying point. For example, assuming that the acquisition points corresponding to a quasi-stop point are a, b, and c, the moving distance corresponding to the quasi-stop point is the distance between a and b plus the distance between b and c. The setting conditions are respectively set according to different purification parameters, when the purification parameters corresponding to the quasi-staying points meet the corresponding setting conditions, the quasi-staying points are used as the staying points, and when the purification parameters corresponding to the quasi-staying points do not meet the corresponding setting conditions, the quasi-staying points are not used as the staying points, so that the purification of the aligned staying points is realized.

In an alternative embodiment, the purification parameter comprises an average velocity; the method comprises the following steps of determining purification parameters corresponding to the quasi-staying points according to track characteristic parameters of the collection points corresponding to the quasi-staying points, purifying the quasi-staying points according to the relation between the purification parameters and set conditions, and acquiring the staying points, and comprises the following steps:

determining the staying time and the moving distance corresponding to the quasi staying point according to the time and the geographic position of the acquisition point corresponding to the quasi staying point;

obtaining the average speed corresponding to the quasi-staying point according to the staying time and the moving distance;

and when the average speed corresponding to the quasi-staying point is determined to be smaller than a set average speed threshold value, taking the quasi-staying point as a staying point.

Here, for each quasi-stop point, there are two or more acquisition points corresponding to the quasi-stop point. The determining the stopping time and the moving distance corresponding to the quasi-stopping point according to the time and the geographic position of the acquisition point corresponding to the quasi-stopping point comprises the following steps: determining the dwell time corresponding to the quasi-dwell point according to the time of a first acquisition point and the time of a last acquisition point corresponding to the quasi-dwell point, wherein the dwell time corresponding to the quasi-dwell point is obtained by subtracting the time of the first acquisition point from the time of the last acquisition point; and determining the moving distance corresponding to the quasi-staying point according to the geographical position of each acquisition point corresponding to the quasi-staying point, wherein the moving distance corresponding to the quasi-staying point is the sum of the distances between the acquisition points corresponding to the quasi-staying point. Correspondingly, the step of obtaining the average speed corresponding to the quasi-staying point according to the staying time and the moving distance comprises dividing the moving distance by the staying time, and taking the obtained value as the average speed corresponding to the quasi-staying point. For example, assuming that the acquisition points corresponding to a quasi-stop point are a, b, c, d, e, the average speed corresponding to the quasi-stop point is the distance between a and b, the distance between b and c, the distance between c and d, and the sum of the distances between d and e divided by the time interval between a and e. The average speed threshold value can be set according to actual requirements, for example, the average speed threshold value can be set to 1km/h or 2 km/h. When the average speed corresponding to the quasi-staying point is smaller than the set average speed threshold, the target object moves slowly between the acquisition points corresponding to the quasi-staying point, and the target object may stay or wander in an area or place of interest, so that the quasi-staying point is used as the staying point. Therefore, the quasi-stop point with the average speed smaller than the set average speed threshold value is used as the stop point, the accuracy of stop point acquisition is improved, and the accuracy of the subsequently acquired interest area is further improved.

In an alternative embodiment, the purification parameters include residence rate; the method comprises the following steps of determining purification parameters corresponding to the quasi-staying points according to track characteristic parameters of the collection points corresponding to the quasi-staying points, purifying the quasi-staying points according to the relation between the purification parameters and set conditions, and acquiring the staying points, and comprises the following steps:

acquiring the speed corresponding to each acquisition point in the acquisition points corresponding to the quasi-staying point according to the time and the geographic position of the acquisition points corresponding to the quasi-staying point;

acquiring the staying rate corresponding to the quasi staying point according to the number of the acquisition points corresponding to the quasi staying point and the number of the acquisition points of which the speed is less than a set speed threshold value in the acquisition points corresponding to the quasi staying point;

and when the stopping rate corresponding to the quasi stopping point is determined to be larger than a set stopping rate threshold value, taking the quasi stopping point as a stopping point.

Specifically, the speed corresponding to each acquisition point in the acquisition points corresponding to the quasi-staying point is obtained according to the time and the geographic position of the acquisition points corresponding to the quasi-staying point; determining the number of the acquisition points with the speed smaller than a set speed threshold in the acquisition points corresponding to the quasi-staying point according to the speed corresponding to each acquisition point; acquiring the stopping rate corresponding to the quasi stopping point according to the number of the acquisition points with the speed less than a set speed threshold and the number of the acquisition points corresponding to the quasi stopping point; and when the stopping rate corresponding to the quasi stopping point is determined to be larger than a set stopping rate threshold value, taking the quasi stopping point as a stopping point.

Here, for each quasi-stop point, there are two or more acquisition points corresponding to the quasi-stop point. The acquiring the speed corresponding to each acquisition point in the acquisition points corresponding to the quasi-staying point according to the time and the geographic position of the acquisition points corresponding to the quasi-staying point comprises the following steps: and determining the speed corresponding to each acquisition point according to the time and the geographic position of two adjacent acquisition points in the acquisition points corresponding to the quasi-staying point. It is understood that the velocity corresponding to any one acquisition point is the distance between the acquisition point and the next adjacent acquisition point divided by the time interval between the acquisition point and the next adjacent acquisition point, said next adjacent acquisition point being the next adjacent acquisition point after and adjacent to the time of the acquisition point. Taking the speed corresponding to the first acquisition point corresponding to the quasi-staying point as an example, the speed corresponding to the first acquisition point is the distance between the second acquisition point and the first acquisition point divided by the time interval between the second acquisition point and the first acquisition point. For example, assuming that the acquisition points corresponding to one quasi-staying point are a, b, c, d, e, the speed corresponding to the acquisition point a is obtained by dividing the distance between a and b by the time interval between a and b; and obtaining the speed corresponding to the acquisition point b according to the distance between b and c divided by the time interval between b and c, and so on. The acquiring the stopping rate corresponding to the quasi stopping point according to the number of the acquiring points with the speed less than the set speed threshold and the number of the acquiring points corresponding to the quasi stopping point comprises the following steps: and taking the value obtained by dividing the number of the acquisition points with the speed less than the set speed threshold value by the number of the acquisition points corresponding to the quasi-staying points as the staying rate corresponding to the quasi-staying points. The speed threshold value can be set according to actual requirements, for example, the speed threshold value can be set to be 3km/h or 4 km/h. The stay rate threshold can be set according to actual requirements, for example, it can be set to 0.7 or 0.8. When the stopping rate corresponding to the quasi stopping point is larger than the set stopping rate threshold value, the speed of the target object moving between the acquisition points corresponding to the quasi stopping point is slow, and the target object may stop moving frequently, stay or wander in an area or place of interest, and therefore the quasi stopping point is used as the stopping point. It should be noted that, the purification parameters corresponding to the quasi-staying point are determined according to the track characteristic parameters of the collection point corresponding to the quasi-staying point, and the quasi-staying point is purified according to the relationship between the purification parameters and the setting conditions to obtain the staying point, or: determining the moving distance corresponding to the quasi-staying point and the speed corresponding to each acquisition point in the acquisition points corresponding to the quasi-staying point according to the time and the geographic position of the acquisition points corresponding to the quasi-staying point; dividing the number of the acquisition points with the speed smaller than a set speed threshold value in the acquisition points corresponding to the quasi-staying point by the moving distance corresponding to the quasi-staying point to obtain a value as the staying rate corresponding to the quasi-staying point; and when the stopping rate corresponding to the quasi stopping point is determined to be larger than a set stopping rate threshold value, taking the quasi stopping point as a stopping point. Therefore, the quasi-staying point with the staying rate larger than the set staying rate threshold value is used as the staying point, the accuracy of staying point acquisition is improved, and the accuracy of the subsequently acquired interest area is further improved.

In an alternative embodiment, the purification parameter comprises a rate of change of speed; the method comprises the following steps of determining purification parameters corresponding to the quasi-staying points according to track characteristic parameters of the collection points corresponding to the quasi-staying points, purifying the quasi-staying points according to the relation between the purification parameters and set conditions, and acquiring the staying points, and comprises the following steps:

acquiring the speed change rate corresponding to each acquisition point in the acquisition points corresponding to the quasi-staying point according to the time and the geographic position of the acquisition points corresponding to the quasi-staying point;

determining the speed change rate corresponding to the quasi-staying point according to the number of the acquisition points corresponding to the quasi-staying point and the number of the acquisition points of which the speed change rate in the acquisition points corresponding to the quasi-staying point is greater than a set first speed change rate threshold value;

and when the speed change rate corresponding to the quasi-staying point is determined to be greater than a set second speed change rate threshold value, taking the quasi-staying point as a staying point.

Specifically, according to the time and the geographic position of the acquisition points corresponding to the quasi-staying point, the speed change rate corresponding to each acquisition point in the acquisition points corresponding to the quasi-staying point is acquired; determining the number of the acquisition points of which the speed change rate is greater than a set first speed change rate threshold value in the acquisition points corresponding to the quasi-staying point according to the speed change rate corresponding to each acquisition point; determining the speed change rate corresponding to the quasi-staying point according to the number of the acquisition points of which the speed change rate is greater than a set first speed change rate threshold value and the number of the acquisition points corresponding to the quasi-staying point; and when the speed change rate corresponding to the quasi-staying point is determined to be greater than a set second speed change rate threshold value, taking the quasi-staying point as a staying point.

Here, the obtaining, according to the time and the geographic position of the acquisition point corresponding to the quasi-staying point, a speed change rate corresponding to each acquisition point in the acquisition points corresponding to the quasi-staying point includes: determining the speed corresponding to each acquisition point according to the time and the geographic position of two adjacent acquisition points in the acquisition points corresponding to the quasi-staying point; and determining the speed change rate corresponding to each acquisition point according to the speed corresponding to the two adjacent acquisition points. It is understood that the velocity corresponding to any one acquisition point is the distance between the acquisition point and the next adjacent acquisition point divided by the time interval between the acquisition point and the next adjacent acquisition point, said next adjacent acquisition point being the next adjacent acquisition point that is subsequent in time to the acquisition point. Taking the speed corresponding to the first acquisition point corresponding to the quasi-staying point as an example, the speed corresponding to the first acquisition point is the distance between the second acquisition point and the first acquisition point divided by the time interval between the second acquisition point and the first acquisition point. The speed change rate corresponding to any one acquisition point is the absolute value of the difference between the speed corresponding to the acquisition point and the speed corresponding to the next acquisition point adjacent to the acquisition point, and the speed corresponding to the acquisition point is divided by the speed corresponding to the acquisition point. Taking the speed change rate corresponding to the first acquisition point corresponding to the quasi-staying point as an example, the speed change rate corresponding to the first acquisition point is the absolute value of the difference between the speed corresponding to the second acquisition point and the speed corresponding to the first acquisition point divided by the speed corresponding to the first acquisition point. The determining the speed change rate corresponding to the quasi-staying point according to the number of the collecting points of which the speed change rate is greater than a set first speed change rate threshold value and the number of the collecting points corresponding to the quasi-staying point comprises the following steps: and taking the value obtained by dividing the number of the acquisition points with the speed change rate larger than the set first speed change rate threshold value by the number of the acquisition points corresponding to the quasi-staying points as the speed change rate corresponding to the quasi-staying points. The first speed change rate threshold may be set according to actual requirements, for example, may be set to 0.4 or 0.5. The second speed change rate threshold may be set according to actual requirements, for example, may be set to 0.5 or 0.6. When the speed change rate corresponding to the quasi-staying point is larger than a set second speed change rate threshold value, it is indicated that the speed change of the target object moving between the acquisition points corresponding to the quasi-staying point is fast, and the target object may move around an interested area or a place, so that the quasi-staying point is used as a staying point. It should be noted that, the purification parameters corresponding to the quasi-staying point are determined according to the track characteristic parameters of the collection point corresponding to the quasi-staying point, and the quasi-staying point is purified according to the relationship between the purification parameters and the setting conditions to obtain the staying point, or: determining the moving distance corresponding to the quasi-staying point and the speed change rate corresponding to each acquisition point in the acquisition points corresponding to the quasi-staying point according to the time and the geographic position of the acquisition points corresponding to the quasi-staying point; dividing the number of the acquisition points with the speed change rate larger than a set first speed change rate threshold value in the acquisition points corresponding to the quasi-staying point by the moving distance corresponding to the quasi-staying point to obtain a value as the speed change rate corresponding to the quasi-staying point; and when the speed change rate corresponding to the quasi-staying point is determined to be greater than a set second speed change rate threshold value, taking the quasi-staying point as a staying point. Therefore, the quasi-stay point with the speed change rate larger than the set second speed change rate threshold value is used as the stay point, the accuracy of stay point acquisition is improved, and the accuracy of the subsequently acquired interest area is further improved.

In an alternative embodiment, the purification parameters include an angle conversion rate; the method comprises the following steps of determining purification parameters corresponding to the quasi-staying points according to track characteristic parameters of the collection points corresponding to the quasi-staying points, purifying the quasi-staying points according to the relation between the purification parameters and set conditions, and acquiring the staying points, and comprises the following steps:

acquiring the angle corresponding to each acquisition point except the first and last acquisition points in the acquisition points corresponding to the quasi-staying point according to the geographic position of the acquisition point corresponding to the quasi-staying point;

acquiring an angle conversion rate corresponding to the quasi-staying point according to the number of the acquisition points except the first and last acquisition points in the acquisition points corresponding to the quasi-staying point and the number of the acquisition points of which the angles are larger than a set angle threshold value in the acquisition points corresponding to the quasi-staying point;

and when the angle conversion rate corresponding to the quasi-staying point is determined to be larger than a set angle conversion rate threshold value, taking the quasi-staying point as a staying point.

Specifically, according to the geographic position of the acquisition point corresponding to the quasi-staying point, the angle corresponding to each acquisition point except the first and last acquisition points in the acquisition points corresponding to the quasi-staying point is acquired; determining the number of the acquisition points with the angles larger than a set angle threshold in the acquisition points corresponding to the quasi-staying points according to the angle corresponding to each acquisition point; acquiring an angle conversion rate corresponding to the quasi-staying point according to the number of the acquisition points of which the angle is greater than a set angle threshold value and the number of the acquisition points except the first and last acquisition points in the acquisition points corresponding to the quasi-staying point; and when the angle conversion rate corresponding to the quasi-staying point is determined to be larger than a set angle conversion rate threshold value, taking the quasi-staying point as a staying point.

Here, the obtaining, according to the geographic position of the acquisition point corresponding to the quasi-staying point, an angle corresponding to each acquisition point except the first and last acquisition points in the acquisition points corresponding to the quasi-staying point includes: and determining the angle corresponding to each acquisition point according to the geographic position of the adjacent acquisition point in the acquisition points corresponding to the quasi-staying point. It can be understood that, in the acquisition points corresponding to the quasi-staying point, except for the first and last two acquisition points, an angle corresponding to any one acquisition point is an included angle formed among the acquisition point, the last acquisition point adjacent to the acquisition point and the next acquisition point adjacent to the acquisition point, that is, an included angle formed between a vector from the acquisition point to the last acquisition point adjacent to the acquisition point and a vector from the acquisition point to the next acquisition point adjacent to the acquisition point by taking the starting point as the acquisition point. The last acquisition point adjacent to the acquisition point is an adjacent acquisition point which is prior in time to the acquisition point, and the next acquisition point adjacent to the acquisition point is an adjacent acquisition point which is subsequent in time to the acquisition point. Taking the angle corresponding to the second collecting point corresponding to the quasi-staying point as an example, the angle corresponding to the second collecting point is an included angle formed among the first collecting point, the second collecting point and the third collecting point. The angle threshold can be set according to actual requirements, for example, the angle threshold can be set to 110 degrees or 120 degrees. The acquiring the angle conversion rate corresponding to the quasi-staying point according to the number of the acquisition points of which the angle is greater than the set angle threshold and the number of the acquisition points except the first and last acquisition points in the acquisition points corresponding to the quasi-staying point comprises: and taking the value obtained by dividing the number of the acquisition points of which the angle is greater than the set angle threshold value by the number of the acquisition points except the first and last acquisition points in the acquisition points corresponding to the quasi-staying point as the angle conversion rate corresponding to the quasi-staying point. The angle conversion rate threshold can be set according to actual requirements, for example, it can be set to 0.6 or 0.8. When the angle conversion rate corresponding to the quasi-staying point is greater than a set angle conversion rate threshold value, it indicates that the target object may move between the acquisition points corresponding to the quasi-staying point in a manner of slow speed and changeable moving direction, such as a walking manner, and the target object may hover in an area or place of interest, so that the quasi-staying point is taken as the staying point. Therefore, the quasi-staying point with the angle conversion rate larger than the set angle conversion rate threshold value is used as the staying point, the accuracy of staying point acquisition is improved, and the accuracy of the subsequently acquired interest region is further improved.

It should be noted that, when the purification parameters include an average speed, a residence rate, a speed change rate, and an angle change rate, the obtained residence point needs to satisfy the following conditions: the average speed corresponding to the stay point is smaller than a set average speed threshold, the stay rate corresponding to the stay point is larger than a set stay rate threshold, the speed change rate corresponding to the stay point is larger than a set second speed change rate threshold, and the angle transition rate corresponding to the stay point is larger than a set angle transition rate threshold.

Step S103: and clustering the stop points, and obtaining the interest area of the target object according to the clustering result.

Here, the clustering of the staying points may be performed by performing clustering operation on the staying points by using a density clustering algorithm, so as to obtain a distribution of the staying points, and obtain an interest region of the target object according to the distribution of the staying points. In this embodiment, density clustering is performed on the stopover points as an example, assuming that the maximum radius of a neighborhood is Eps, and the minimum number of stopover points in an Eps-neighborhood is MinPts, if the number of stopover points included in the Eps-neighborhood of a stopover point is equal to or greater than MinPts, the stopover point is a core stopover point, and a cluster is formed by the stopover point and all stopover points whose stopover point densities can reach, which indicates that the geographic positions of the stopover points are close to each other, even the geographic positions are the same. When the cluster is identified by a geographic region, it is an area of interest of the target object.

Here, the clustering of the stopover points may also be based on classification attributes of the stopover points, for example, the stopover points may be classified according to geographic attributes, or the stopover points may be classified according to consumption attributes, and the like, and by clustering different classification attributes of the stopover points to obtain corresponding clustering results, the distribution of the stopover points based on different classification attributes may be obtained, so as to obtain the interest area of the target object according to the distribution of the stopover points, and information and the like related to the interest area of the target object may be recommended to the target object.

In summary, in the method for extracting an interest region provided in the above embodiment, the quasi-staying points determined based on the acquisition points representing the movement track information of the target object are purified, so that the interfered staying point data can be reduced, the abnormal staying points affecting the accuracy of the interest region are deleted, the staying points obtained after purification are clustered, and the accuracy of the obtained interest region of the target object is improved.

In order to implement the foregoing method, an interest region extracting apparatus is further provided corresponding to an embodiment of the present invention, as shown in fig. 2, the apparatus includes:

the extraction module 10 is configured to determine a quasi-stop point according to an acquisition point of movement trajectory information for representing a target object;

the purification module 20 is configured to determine a purification parameter corresponding to the quasi-staying point according to a track characteristic parameter of the collection point corresponding to the quasi-staying point, and purify the quasi-staying point according to a relationship between the purification parameter and a setting condition to obtain a staying point;

and the clustering module 30 is configured to cluster the stop points and obtain the interest region of the target object according to a clustering result.

In summary, the interesting region extracting apparatus provided in the above embodiment determines the quasi-staying point through the acquisition point for representing the movement track information of the target object, refines the quasi-staying point according to the relationship between the refining parameter corresponding to the quasi-staying point and the setting condition to obtain the staying point, and clusters the staying point to obtain the interesting region of the target object. Therefore, the quasi-staying points determined by the acquisition points based on the information representing the moving track of the target object are purified, so that the interfering staying point data can be reduced, the abnormal staying points influencing the accuracy of the interest area are deleted, the staying points obtained after purification are clustered, and the accuracy of the obtained interest area of the target object is improved.

In an alternative embodiment, the purification parameter comprises an average velocity; the purification module 20 is specifically configured to:

obtaining the average speed corresponding to the quasi-staying point according to the staying time and the moving distance; and when the average speed corresponding to the quasi-staying point is determined to be smaller than a set average speed threshold value, taking the quasi-staying point as a staying point.

Therefore, the quasi-stop point with the average speed smaller than the set average speed threshold value is used as the stop point, the accuracy of stop point acquisition is improved, and the accuracy of the subsequently acquired interest area is further improved.

In an alternative embodiment, the purification parameters include residence rate; the purification module 20 is specifically configured to:

when the stopping rate corresponding to the quasi stopping point is determined to be larger than a set stopping rate threshold value, taking the quasi stopping point as a stopping point; alternatively, the first and second electrodes may be,

acquiring the speed corresponding to each acquisition point in the acquisition points corresponding to the quasi-staying point and the moving distance corresponding to the quasi-staying point according to the time and the geographic position of the acquisition points corresponding to the quasi-staying point;

acquiring a stopping rate corresponding to the quasi stopping point according to the moving distance corresponding to the quasi stopping point and the number of the acquisition points of which the speed is less than a set speed threshold in the acquisition points corresponding to the quasi stopping point;

Therefore, the quasi-staying point with the staying rate larger than the set staying rate threshold value is used as the staying point, the accuracy of staying point acquisition is improved, and the accuracy of the subsequently acquired interest area is further improved.

In an alternative embodiment, the purification parameter comprises a rate of change of speed; the purification module 20 is specifically configured to:

when the speed change rate corresponding to the quasi-staying point is determined to be larger than a set second speed change rate threshold value, taking the quasi-staying point as a staying point; alternatively, the first and second electrodes may be,

acquiring the speed change rate corresponding to each acquisition point in the acquisition points corresponding to the quasi-staying point and the moving distance corresponding to the quasi-staying point according to the time and the geographic position of the acquisition points corresponding to the quasi-staying point;

acquiring the speed change rate corresponding to the quasi-staying point according to the moving distance corresponding to the quasi-staying point and the number of the acquisition points of which the speed change rate in the acquisition point corresponding to the quasi-staying point is greater than a set first speed change rate threshold value;

Therefore, the quasi-stay point with the speed change rate larger than the set second speed change rate threshold value is used as the stay point, the accuracy of stay point acquisition is improved, and the accuracy of the subsequently acquired interest area is further improved.

In an alternative embodiment, the purification parameters include an angle conversion rate; the purification module 20 is specifically configured to:

Therefore, the quasi-staying point with the angle conversion rate larger than the set angle conversion rate threshold value is used as the staying point, the accuracy of staying point acquisition is improved, and the accuracy of the subsequently acquired interest region is further improved.

In an optional embodiment, the extraction module 10 is specifically configured to:

determining an initial acquisition point;

and determining corresponding stopping points according to the initial acquisition points and the acquisition points between the initial acquisition points and the target acquisition points.

taking the target acquisition point as an updated initial acquisition point;

Therefore, the speed of acquiring the quasi-staying point is improved by updating the initial acquisition point in time.

An embodiment of the present invention provides an interest region extraction apparatus, as shown in fig. 3, the apparatus includes: a processor 310 and a memory 311 for storing computer programs capable of running on the processor 310; the processor 310 illustrated in fig. 3 is not used to refer to the number of the processors 310 as one, but is only used to refer to the position relationship of the processor 310 relative to other devices, and in practical applications, the number of the processors 310 may be one or more; similarly, the memory 311 shown in fig. 3 is also used in the same sense, i.e. it is only used to refer to the position relationship of the memory 311 with respect to other devices, and in practical applications, the number of the memory 311 may be one or more.

The processor 310 is configured to execute the following steps when executing the computer program:

In an alternative embodiment, the purification parameter comprises an average velocity; the processor 310 is further configured to execute the following steps when the computer program is executed:

In an alternative embodiment, the purification parameters include residence rate; the processor 310 is further configured to execute the following steps when the computer program is executed:

In an alternative embodiment, the purification parameter comprises a rate of change of speed; the processor 310 is further configured to execute the following steps when the computer program is executed:

when the speed change rate corresponding to the quasi-staying point is determined to be larger than a set second speed change rate threshold value, taking the quasi-staying point as a staying point;

or acquiring the speed change rate corresponding to each acquisition point in the acquisition points corresponding to the quasi-staying point and the moving distance corresponding to the quasi-staying point according to the time and the geographic position of the acquisition point corresponding to the quasi-staying point;

In an alternative embodiment, the purification parameters include an angle conversion rate; the processor 310 is further configured to execute the following steps when the computer program is executed:

In an alternative embodiment, the processor 310 is further configured to execute the following steps when the computer program is executed:

determining an initial acquisition point;

taking the target acquisition point as an updated initial acquisition point;

The device also includes: at least one network interface 312. The various components of the device are coupled together by a bus system 313. It will be appreciated that the bus system 313 is used to enable communications among the components connected. The bus system 313 includes a power bus, a control bus, and a status signal bus in addition to the data bus. For clarity of illustration, however, the various buses are labeled as bus system 313 in FIG. 3.

The memory 311 may be a volatile memory or a nonvolatile memory, or may include both volatile and nonvolatile memories. Among them, the nonvolatile Memory may be a Read Only Memory (ROM), a Programmable Read Only Memory (PROM), an Erasable Programmable Read-Only Memory (EPROM), an Electrically Erasable Programmable Read-Only Memory (EEPROM), a magnetic random access Memory (FRAM), a Flash Memory (Flash Memory), a magnetic surface Memory, an optical disk, or a Compact Disc Read-Only Memory (CD-ROM); the magnetic surface storage may be disk storage or tape storage. Volatile memory can be Random Access Memory (RAM), which acts as external cache memory. By way of illustration and not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Synchronous Static Random Access Memory (SSRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), Enhanced Synchronous Dynamic Random Access Memory (Enhanced DRAM), Synchronous Dynamic Random Access Memory (SLDRAM), Direct Memory (DRmb Access), and Random Access Memory (DRAM). The memory 311 described in connection with the embodiments of the invention is intended to comprise, without being limited to, these and any other suitable types of memory.

The memory 311 in the embodiment of the present invention is used to store various types of data to support the operation of the apparatus. Examples of such data include: any computer program for operating on the device, such as operating systems and application programs; contact data; telephone book data; a message; a picture; video, etc. The operating system includes various system programs, such as a framework layer, a core library layer, a driver layer, and the like, and is used for implementing various basic services and processing hardware-based tasks. The application programs may include various application programs such as a Media Player (Media Player), a Browser (Browser), etc. for implementing various application services. Here, the program that implements the method of the embodiment of the present invention may be included in an application program.

The present embodiment also provides a computer storage medium, in which a computer program is stored, where the computer storage medium may be a Memory such as a magnetic random access Memory (FRAM), a Read Only Memory (ROM), a Programmable Read Only Memory (PROM), an Erasable Programmable Read-Only Memory (EPROM), an Electrically Erasable Programmable Read-Only Memory (EEPROM), a Flash Memory (Flash Memory), a magnetic surface Memory, an optical Disc, or a Compact Disc Read-Only Memory (CD-ROM); or may be a variety of devices including one or any combination of the above memories, such as a mobile phone, computer, tablet device, personal digital assistant, etc.

A computer storage medium having a computer program stored therein, the computer program, when executed by a processor, performing the steps of:

In an alternative embodiment, the refining parameter comprises an average speed, and the computer program, when executed by the processor, further performs the steps of:

In an alternative embodiment, the purification parameters include a retention rate, and the computer program, when executed by the processor, further performs the steps of:

acquiring the staying rate corresponding to the quasi staying point according to the number of the acquisition points corresponding to the quasi staying point and the number of the acquisition points of which the speed is less than a set speed threshold in the acquisition points corresponding to the quasi staying point;

In an alternative embodiment, the refining parameter comprises a rate of change of speed, and the computer program, when executed by the processor, further performs the steps of:

In an alternative embodiment, the purification parameters include an angle conversion, and the computer program, when executed by the processor, further performs the steps of:

acquiring an angle conversion rate corresponding to the quasi-staying point according to the number of the acquisition points corresponding to the quasi-staying point and the number of the acquisition points of which the angles are larger than a set angle threshold value in the acquisition points corresponding to the quasi-staying point;

In an alternative embodiment, the computer program, when executed by the processor, further performs the steps of:

determining an initial acquisition point;

taking the target acquisition point as an updated initial acquisition point;

The following describes an embodiment of the present invention in further detail by using specific examples, and in this embodiment, a target object is taken as an example. The space-time data of a large number of moving objects, namely acquisition points, can be obtained through related network equipment, but one single acquisition point can only explain the space-time information of the moving objects, and a dwell point data model with semantic information needs to be established when a user wants to obtain the space-time position with the semantic information. The construction of the data model of the stop point of the moving object mainly comprises the following three parts:

first, stop point data model definition

Acquisition Point (Point): the spatiotemporal information of the mobile object obtained through the signaling data receiving apparatus includes information of Longitude (Longitude), Latitude (Latitude), altitude, and time, as shown in fig. 4(a), where P is_iI is more than or equal to 1 and less than or equal to n, and the 1 st acquisition point and the nth acquisition point are respectively marked as P₁＝{lat₁,long₁,T₁}、P_n＝{lat_n,long_n,T_n}。

Movement Log (Log): the sequence of acquisition points in the movement log can be expressed by a matrix, as shown in fig. 4 (a).

Movement Trajectory (Trajectory): in three dimensions, will be differentThe collection of collection points formed by connecting collection points according to chronological order is shown in fig. 4(b), and in this example, the collection of all collection points is denoted as P ═ P₁,P₂,...,P_n}。

The stop point data model represents a geographical area and has special semantic information, such as a place where a user works or lives, a restaurant frequently going to eat, or a scenic spot during travel, etc., which can be represented by a stop point, as shown in fig. 5, and the original collection point sequence { P in fig. 4 can be represented by a stop point₁,P₂,...,P_nConverts to an abstract sequence of stop points S₁,S₂,...,S_m}. Wherein each stop point S_iAll consist of longitude, latitude, start time, end time and duration span, i is more than or equal to 1 and less than or equal to m. The starting time is the time of the first acquisition point corresponding to the staying point, the ending time is the time of the last acquisition point corresponding to the staying point, and the duration span is the time obtained by subtracting the starting time from the ending time.

Therefore, it is necessary to extract the stopover point data model, both from the quality of the extracted information and from storage and computational overhead considerations. The feature definition of the stop point data model is based on various practical situations, and considers a plurality of factors such as the stop time in the same stop area, the distance between the collection points, the average speed in the stop area, the stop rate in the unit distance, the speed change rate, the angle conversion rate and the like, and after obtaining the stop point set, density reachable clustering is carried out to obtain the user interest area.

Second, initial extraction of stop points

The extraction of the stop points is divided into two cases: one situation is stay, for example, when a person enters a building, the user's location information stays under a certain base station until it appears outdoors again, at which point the stay time exceeds a certain time span; another situation is loitering, for example when people travel outdoors to a certain landscape or are attracted to the surrounding environment, where the dwell time exceeds a certain threshold while the distance between successive collection points is less than a certain distance threshold.

In this embodiment, the time distance and the spatial distance are used as initial extraction conditions, and the initial extracted stopover point is used as a quasi-stopover point. The quasi-stop point represents that the user stays in a certain area for more than a time threshold T_rAnd the distance between any two acquisition points is less than a threshold value D_r。Dist(p_i,p_j) Representing two acquisition points p_iAnd p_jDistance between, Int (p)_i,p_j)＝|p_i.t_i-p_j.t_jI denotes two acquisition points p_iAnd p_jThe time interval in between. The quasi-stop point can be considered as a virtual stop point position extracted by a set of consecutive acquisition points. Conform to

So that Dist (p)_m,p_i)≤D_r、Dist(p_m,p_n+1)＞D_rAnd Int (p)_m,p_n)≥T_rThus, quasi-dwell point s ═ x, y, t_a,t_l) Is composed of

Wherein, in formula (1-1) and formula (1-2), x and y respectively represent each acquisition point P in the acquisition point set P_iAverage coordinates in the x-axis and y-axis; s.t_a＝p_m.t_mIndicating the arrival time of the quasi-stop point, i.e. the start time, s.t_l＝p_n.t_nIndicating the departure time, i.e. the end time, of the quasi-stop point. Compared with the original acquisition point, the quasi-stop point contains richer semantic information, can better represent the potential information contained in the original data set, and can reduce the operation scale to improve the calculation efficiency for the subsequent calculation. As shown in FIG. 6, Stay Point 1 and Stay Point 2 respectively represent two quasi-stop points, wherein Stay Point 1 tableShowing a user at a certain acquisition Point p₃Has exceeded a time threshold T_rHowever, most quasi-lingering points will be similar to Stay Point 2, indicating that the user wanders in a region consisting of a sequence of acquisition points, e.g., P ═ P, for a period of time₅,p₆,p₇,p₈> (ii). The above limiting conditions are to perform the initialized extraction on the massive data, and can efficiently select the quasi-stop point, and then further purify the quasi-stop point to obtain the stop point.

The generated quasi-stop points are further analyzed, and the practical situation of the user is considered, if the destination of the user is clear, for example, the user goes home directly from the working place, the home point with semantic information is extracted as the quasi-stop point, and other acquisition points do not have any practical semantic information. In addition, if the destination of the user is not clear, the mobile data of the user is processed by the method for extracting the quasi-stop point, so that a plurality of quasi-stop points can be generated, for example, the user self-service travels in a certain city, the random of the access place is high, the interest point is uncertain, and the generation of the quasi-stop points is also more. The semantic information contained in the stopover points generated by the later is richer, and more potential information of people can be mined.

Method for extracting interest area

The areas capable of generating the stopping points are obtained through observation and analysis of actual conditions, are often places which attract people to stop and have rich semantic information and are places which can be reached by people in a walking mode, and the stopping point set is subjected to clustering analysis to obtain an interest area set of each user.

First, the dwell point needs to be aligned for refinement to obtain the dwell point. The average velocity is used as the purification condition for the above-mentioned quasi-stagnation point. When people travel outdoors to attract a certain landscape or ambient environment, the places really attracting people often need people to reach by walking, so the average speed of continuous acquisition points in the extracted quasi-stop point is within the walking speed range of the user, the average speed of the continuous acquisition points is not more than the walking maximum speed threshold of the user, and if the walking maximum speed threshold is more than the walking maximum speed threshold, the user is explained to transfer vehicles, and the area represented by the quasi-stop point is not interesting to the user. For example, as shown in fig. 7, Stay Point 1, Stay Point 2, and Stay Point 3 respectively represent three quasi-stop points, where a user waits for a bus when reaching a bus stop P9, and the user leaves the place with the bus after the bus arrives, but the quasi-stop points are extracted by using the above method for extracting stop points, Stay Point 3 will be extracted as a quasi-stop Point, but Stay Point 3 does not contain any semantic information of the user, and "noise Point" of the quasi-stop Point set, which is Stay Point 3, will be eliminated, and the key Point of purification by using the average speed is to calculate an average speed corresponding to the quasi-stop points according to the continuous collection points in the quasi-stop points, and determine whether the quasi-stop points have actual semantic information according to the average speed.

The residence rate, rate of change of velocity and angle change rate will be used as the above-mentioned quasi-residence point for further purification. As people need to change vehicles frequently to reach a destination, a variety of travel modes may be involved over a range of distances. The speeds of different travel modes are easily affected by traffic environment and weather, and if the user's travel mode is determined to be walking only by the average speed, an erroneous determination may occur. For example, on a crowded road, it is intuitive to feel that driving is slower than the walking speed of a person. Therefore, only considering the average velocity as the purification condition of the stop point, a meaningless stop point will occur. Therefore, in this example, the residence rate, the rate of change in velocity and the angle change rate are defined as the conditions for further purification at the residence point.

Retention Rate (SR, Stop Rate): indicating that the speed value of the acquisition point within a unit distance is below a set speed threshold V_sThe number of acquisition points size. Using a speed threshold V_sThe collection points corresponding to each quasi-staying point can be extracted, so that a group of collection point sets, which are marked as P, can be obtained_s＝{p_i|p_i∈P,p_i.V＜V_sEach acquisition point in the set of acquisition points has a velocity lower than that of the acquisition pointV_s. According to the formula SR ═ P_sAnd | the Distance obtains the SR corresponding to each quasi-staying point, wherein the Distance represents the moving Distance corresponding to the quasi-staying point. As shown in fig. 8, fig. 8(a), fig. 8(b), and fig. 8(c) show the stay rates of Driving (Driving), (Bus), and Walking (Walking), respectively, and the speed (Velocity) is less than the speed threshold V from Driving, Bus, to Walking within the same distance_sThe number of the collection points is increased in turn, and the walking retention rate is obviously higher than the non-walking retention rate SR (Walk) > SR (non-Walk).

When different travel modes are adopted, the speed change trend of the bicycle is obviously changed. In the same distance range, people adopt different travel modes, and the stay times can be greatly changed. In essence, driving stops are due to waiting for traffic lights at the intersection, while passengers getting on and off the bus will need to stop when the bus arrives. Meanwhile, the number of stops on foot is significantly more frequent than other travel patterns, possibly due to waiting for a bus, chatting with a pedestrian, being attracted by the surrounding environment, and the like.

Speed Change Rate (SCR): the rate of change of the speed corresponding to each quasi-staying point can be obtained by the formulas (1-3) and (1-4). Setting a certain speed threshold V_rCounting that the speed change rate in the acquisition point corresponding to each quasi-staying point is higher than V_rThe number of the collection points, wherein the speed change rate P corresponding to the collection points can be calculated according to the formula (1-3)_iVRate, the rate of change of speed SCR corresponding to the quasi-stay point can be calculated according to the formula (1-4):

P_i.VRate＝|V_i+1-V_i|/V_i(1-3)

SCR＝|P_v|/Distance (1-4)

wherein, P_v＝{p_i|p_i∈P,p_i.VRate＞V_r}，|P_vI represents that the speed change rate is higher than V_rCollection point set P of_vIn (c) p_iThe Distance represents the moving Distance corresponding to the quasi-staying point. For a dwell point, it is understood that the SCR corresponding to the dwell point should be atThe unit distance is above a set rate of change of speed threshold.

Angle Change Rate (ACR, Angle Change Rate): setting an angle threshold A according to the angle between each acquisition point in the track and two adjacent acquisition points connected with the acquisition point_rCounting that the corresponding angle in the acquisition point corresponding to each quasi-staying point is higher than A_rThe number of the collection points can calculate the angle conversion rate ACR corresponding to each stopping point through a formula (1-5):

ACR＝|A_v|/(Pionts-2) (1-5)

wherein A is_v＝{A_i|a_i∈A,a_i.ARate＞A_r}，|A_vI represents an angle higher than A_rCollection Point set A of_vIn (a)_iAnd Pionts represents the number of acquisition points corresponding to the quasi-staying point. As shown in FIG. 9, acquisition Point P2 corresponds to an angle with acquisition Point P2 and with adjacent acquisition points P1 and P3, and acquisition point P3 corresponds to an angle with acquisition point P3 and with adjacent acquisition points P2 and P4. For the dwell point, it can be understood that the ACR corresponding to the dwell point should be above the set angle transition rate threshold.

Residence point p ∈ neighborhood: the density of the dwell points p refers to a region with p as a center and epsilon as a radius, and can be represented by N _ epsilon (p), wherein N _ epsilon (p) is { q epsilon D | dist (p, q) ≦ epsilon }, and D represents a set of dwell points. Meanwhile, the density of the stop points in the area with the E as the radius near the stop point p can also be represented by the number of the stop points contained in N _ E (p). The general requirement for the user's interest area is that for any point p in the interest area, the number of points in N _ e (p) is greater than the set minimum number value (MinPts).

Density value of unit area of stop point: from the set of stop points obtained from the above constraint, if a stop point p is reachable from point q with respect to the direct density of e, MinPts, the following condition needs to be satisfied:

p∈N_∈(q)；

|N_∈(p)|≥MinPts。

the density can reach: if there is a set of stop points D ═ p₁₁,p₁,......,p_nIn which p is₁＝q、p_nP, and satisfies p_i+1Is from p_iWith respect to ∈, MinPts direct density reachable, we call the dwell point p density reachable from the dwell point q.

The user interest area: assuming that D is a set of stop points, if there is a subset C of D that is a cluster for e, MinPts, the following condition is satisfied:

1)

q, if p belongs to C and q is directly reachable from p with respect to the density belonging to C and MinPts, then q belongs to C;

2)

q ∈ C q is reachable from p with respect to ∈, MinPts density.

Based on the previous dwell point data model definition, the overall dwell point data model extraction algorithm is briefly introduced below and is divided into two parts: the first part is that the mass acquisition point data is initialized into a quasi-stop point set, namely a quasi-stop point extraction algorithm; and the second part carries out purification on the set of stop points, namely a stop point purification algorithm. And finally, clustering based on the density of the stop points to obtain the interest area of the user.

The quasi-stagnation point extraction algorithm is used for initially extracting historical position data of each user, and mainly adopts two basic modes from the life habit extraction and track continuity consideration of the users: stay or wander in a certain area. When the staying time of a user in a certain area exceeds a certain threshold value, the time distance and the space distance are provided as characteristic conditions of a staying point data model, so that the high efficiency of calculation is ensured, and the extracted quasi-staying point has experimental analysis capability.

1) Principle of quasi-stagnation point extraction algorithm

Inputting: a collection point set P of a user, a distance threshold distThreh and a time threshold timeThreh;

and (3) outputting: quasi-stagnation point set CSP, CS ═ CSP };

initializing to enable i to be 0, pointNum to be | P |, and sumDistance to be 0; v/| P | represents the number of acquisition points contained in the acquisition point set P;

2) principle of stop point purification algorithm

Inputting: a quasi-stop point set CS of a user, an average speed threshold speedThreh, a speed threshold Vs, a stop rate threshold SRThreh, a speed change rate threshold SCRTHhreh, an angle threshold Ar and an angle change rate threshold ACRTThreh;

and (3) outputting: a set of stop points SP, S ═ SP };

initializing to make i equal to 0, cspNum equal to | P |, sumDistance equal to 0, VchangeCount equal to 0, ACCount equal to 0, and stopCount equal to 0; the// | P | represents the number of acquisition points contained in the CSP;

3) region of interest clustering algorithm principle

Inputting: a set of stop points S, epsilon and MinPts;

and (3) outputting: the user's interest area set G ═ { C1, C2, …, Cm };

get Si from S; i/0 < i < k, Si is a subset of S, which is the set of objects and their positions at time ti;

in the method for extracting the interest region provided by the embodiment, the source position data is preprocessed according to the time threshold and the distance threshold to obtain the quasi-stop point, so that the processing speed of the basic position data and the accuracy of extracting the quasi-stop point are improved; then, the elimination of the noise points in the alignment stop point set is achieved by taking the speed change rate, the stay rate and the angle conversion rate as the purification conditions of the quasi stop points. The key point of purification is that the retention rate, the speed change rate and the angle change rate corresponding to the quasi-retention point are calculated according to the obtained collection point set in the quasi-retention points; finally, performing density clustering on the historical stop point set of the user according to the obtained stop point set, and obtaining the user interest area by using the density value of the unit area of the stop point and the limit of reachable density; the interest region extraction method provided by the embodiment solves the problem that the accuracy of storing and extracting the interest region of the explosively-increased mobile position data is not high. Particularly, the extraction coverage and accuracy of the interest region are improved: according to the method for purifying the stop points based on the time threshold, the distance threshold, the speed threshold, the stay rate, the speed change rate and the angle conversion rate in the embodiment, the position information on a position data set and a boundary which are distributed randomly can be processed; by using the density clustering processing method for the stop points in the embodiment, the position data sets of all the user stop points are analyzed, so that the coverage comprehensiveness and accuracy of user position information are ensured, and the accuracy of extracting the user interest areas is improved; secondly, the processing efficiency is improved, and the storage resources are reduced: based on the dwell point data modeling method in the embodiment, the time complexity of the algorithm is linear time O (n), and compared with the time complexity of the prior art, the time complexity is O (n)²) The processing speed is obviously improved. The method for extracting the interest area provided by the embodiment can effectively reduce the data volume of the key position of the user, thereby improving the storage period and effectively solving the problem of explosively increased data storage of the mobile position.

The above description is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, and improvement made within the spirit and scope of the present invention are included in the protection scope of the present invention.

Claims

1. A method for extracting a region of interest, the method comprising:

2. The method of claim 1, wherein the purification parameters include an average velocity; the method comprises the following steps of determining purification parameters corresponding to the quasi-staying points according to track characteristic parameters of the collection points corresponding to the quasi-staying points, purifying the quasi-staying points according to the relation between the purification parameters and set conditions, and acquiring the staying points, and comprises the following steps:

3. The method of claim 1, wherein the purification parameters include a residence rate; the method comprises the following steps of determining purification parameters corresponding to the quasi-staying points according to track characteristic parameters of the collection points corresponding to the quasi-staying points, purifying the quasi-staying points according to the relation between the purification parameters and set conditions, and acquiring the staying points, and comprises the following steps:

4. The method of claim 1, wherein the purification parameter comprises a rate of change of velocity; the method comprises the following steps of determining purification parameters corresponding to the quasi-staying points according to track characteristic parameters of the collection points corresponding to the quasi-staying points, purifying the quasi-staying points according to the relation between the purification parameters and set conditions, and acquiring the staying points, and comprises the following steps:

5. The method of claim 1, wherein the purification parameters include an angle transition rate; the method comprises the following steps of determining purification parameters corresponding to the quasi-staying points according to track characteristic parameters of the collection points corresponding to the quasi-staying points, purifying the quasi-staying points according to the relation between the purification parameters and set conditions, and acquiring the staying points, and comprises the following steps:

6. The method according to claim 1, wherein the determining the quasi-stop point according to the acquisition point of the movement track information for characterizing the target object comprises:

determining an initial acquisition point;

7. The method of claim 6, further comprising, after determining corresponding quasi-dwell points from the initial acquisition points and the acquisition points between the initial acquisition points and the target acquisition points:

taking the target acquisition point as an updated initial acquisition point;

8. An apparatus for extracting a region of interest, comprising:

9. An apparatus for extracting a region of interest, comprising: a processor and a memory for storing a computer program capable of running on the processor,

wherein the processor is configured to implement the region of interest extraction method according to any one of claims 1 to 7 when the computer program is executed.

10. A computer storage medium, in which a computer program is stored, which, when executed by a processor, implements the region of interest extraction method according to any one of claims 1 to 7.