A short-term rental house tailgating warning system and method based on big data

Abstract

The present invention discloses a short-term rental house tailgating warning system and method based on big data, and relates to the field of intelligent security technology. The present invention realizes monitoring of personnel flow in each unit time period by analyzing the update change of facial feature information collected by a monitoring and detection device in each unit time period, and evaluates the risk coefficient of tenants of target short-term rental houses being tailed in each unit time period; based on the risk coefficient, the minimum danger duration referred to in the anti-tailgating warning analysis in each unit time period is dynamically adjusted. At the same time, the present invention refers to the movement trajectory that exists regularly in a facial feature information database, that is, it is very likely to be a permanent resident when performing the anti-tailgating warning analysis, and is committed to providing a method for analyzing and processing reference warning data that changes with the information of surrounding personnel for tenants of short-term rental houses whose rental environments are usually more complicated, so as to improve the rental safety of tenants of target short-term rental houses.

Description

Technical Field

[0001] The present invention relates to the field of intelligent security technology, and in particular to a short-term rental house tailgating warning system and method based on big data. Background Art

[0002] Short-term rental, also known as mobile, daily rental, self-service, daily rental, is an emerging industry. Short-term rental is a form of business in which the owner rents out the vacant part of his residence to others for a short period of time. In the traditional rental market, the tenant's rental period generally ranges from half a year to several years. Short-term rental shortens this period and provides rental services from one day to several months. It integrates the functions of residence, hotel, and club, and has two major functions: self-use and rental.

[0003] Nowadays, there are more and more anti-tailgating devices on the market, and most of them are anti-tailgating systems installed in the gate area, which use the restricted movement characteristics of the gate to perform anti-tailgating detection and early warning. This anti-tailgating method cannot perform anti-tailgating detection in areas other than devices such as gates, especially when the tailgating person tracks the user's home when the user returns home.

[0004] The above-mentioned anti-tailgating method cannot perform anti-tailgating detection in areas other than equipment such as gates. For tenants of short-term rentals, the environment of mobile personnel renting short-term rentals is relatively complex, which is prone to safety hazards caused by poor personnel management. Summary of the invention

[0005] The purpose of the present invention is to provide a short-term rental house tailgating warning system and method based on big data to solve the problems raised in the prior art.

[0006] To achieve the above purpose, the present invention provides the following technical solution: a short-term rental house tailgating warning method based on big data, the method comprising:

[0007] Step S1: according to the installation position of the monitoring detection device at the door of the target short-term rental house and the maximum detection range, a monitoring area is set at the door of the target short-term rental house, a monitoring video stream of the monitoring area is obtained in real time, and facial feature information of the persons appearing in the monitoring video stream is recognized;

[0008] Step S2: At each unit period, the facial feature information collected in the monitoring area during each unit period of the day is collected and stored, and a facial feature information database corresponding to each unit period of the monitoring area is established;

[0009] Step S3: By analyzing the information update changes of the facial feature information database constructed in each unit time period, the risk coefficient of the tenant of the target short-term rental house being followed in each unit time period is evaluated and calculated;

[0010] Step S4: In all historical facial feature information databases of each unit time period, facial feature information with cumulative repetition times greater than the threshold value is set as the tag information corresponding to each unit time period; in all surveillance video streams acquired to obtain each tag information, the range area where the person corresponding to each tag information first appears and last appears in the display screen of the monitoring detection device is circled;

[0011] Step S5: identifying and extracting the characteristic moving range of each mark information according to the distribution of the range area circled by the person corresponding to each mark information;

[0012] Step S6: receiving the estimated time of return sent in advance by the tenant of the target short-term rental house, and providing the tenant with an anti-tailgating warning reminder based on the risk coefficient of the unit time period in which the estimated time of return falls.

[0013] Preferably, step S3 includes:

[0014] Step S3-1: extracting historical facial feature information bases constructed in chronological order for each unit time period in a day; collecting the total number K of facial feature information contained in each historical facial feature information base, and the time interval between each two adjacent facial feature information being collected and stored in each historical facial feature information base, and obtaining the average time interval T presented in each historical facial feature information base;

[0015] Among them, if the monitoring detection device collects more facial feature information from the monitoring area in a unit time period, and if the monitoring detection device collects and stores facial feature information more frequently, that is, the average time length T is smaller, it means that the monitoring area has a large flow of people in the unit time period and is relatively safer;

[0016] Step S3-2: Calculate the information update index β=K×(1 / T) for each historical facial feature information base in each unit time period in turn; in all historical facial feature information bases in each unit time period, the total number N of facial feature information with a cumulative number of repeated occurrences greater than the number threshold;

[0017] If the monitoring detection device collects more facial feature information with a repetition frequency greater than the threshold value in the monitoring area in a unit time period, it means that the flow of people in the monitoring area in the unit time period is more stable and the relative security is higher;

[0018] Step S3-3: Evaluate and calculate the risk coefficient of the tenant of the target short-term rental house being followed in each unit time period 1 / α=[(β 1 +β 2 +...+β d ) / d]×N; where β 1 , β 2 , ..., βd Respectively represent the information update index corresponding to the 1st, 2nd, ..., dth historical face feature information database in each unit time period;

[0019] Among them, the smaller the risk coefficient 1 / α is, the higher the security of the monitoring area in the unit time period, and the lower the risk of the tenants of the target short-term rental house being followed in each unit time period; the larger the risk coefficient 1 / α is, the lower the security of the monitoring area in the unit time period, and the greater the risk of the tenants of the target short-term rental house being followed in each unit time period.

[0020] Preferably, step S4 comprises:

[0021] Step S4-1: extracting edge lines of the display screen of the monitoring detection device, constructing a plane graphic of the display screen of the monitoring detection device; calling the monitoring video stream presented by the monitoring detection device when acquiring each mark information;

[0022] Step S4-2: Mark the position point where the person corresponding to each mark information first appears in the monitoring video stream in the plane graph as a first-category position point, and mark the position point where the person corresponding to each mark information last appears in the monitoring video stream in the plane graph as a second-category position point;

[0023] Step S4-3: constructing a minimum circle A that can cover all first-category position points and a minimum circle B that can cover all second-category position points for each piece of marking information.

[0024] Preferably, step S5 comprises:

[0025] Step S5-1: for each piece of marking information, respectively, identify the overlap area P(A) between the minimum circle A and the plane figure of the monitoring and detection device display screen, and the overlap area P(B) between the minimum circle B and the plane figure of the monitoring and detection device display screen, wherein the area of ​​the overlap area P(A) is less than or equal to the area of ​​the minimum circle A, and the area of ​​the overlap area P(B) is less than or equal to the area of ​​the minimum circle B;

[0026] Step S5-2: Extract the contour graphic of the overlapping area P(A), and translate the contour graphic in the plane graphic of the display screen of the monitoring and detection device along the direction of the overlapping area P(B) until the intersection area with the overlapping area P(B) is greater than the area threshold, and mark the relevant areas along the path of the contour graphic in the plane graphic of the display screen of the monitoring and detection device, and use the area with the characteristic mark in the plane graphic as the characteristic movement range of each mark information.

[0027] Preferably, step S6 comprises:

[0028] Step S6-1: If the estimated time of returning home is within a certain unit time period, extract the characteristic movement range corresponding to each mark information of the certain unit time period, and mark the characteristic movement range corresponding to each mark information as a safe area; extract the risk coefficient of the tenant of the target short-term rental house being followed in a certain unit time period; set the minimum danger duration T according to the risk coefficient; wherein, the higher the risk coefficient corresponding to a certain unit time period, the smaller the corresponding minimum danger duration T;

[0029] Step S6-2: When the monitoring detection device identifies facial feature information that does not belong to the marking information of a certain unit time period, and monitors that the person corresponding to the facial feature information is continuously in an area not marked as a safe area within the minimum danger duration T, the sound and light warning device is triggered.

[0030] In order to better implement the above method, a short-term rental house tailgating warning system is also proposed, which includes an information recognition and extraction module, an information management module, a risk coefficient calculation and evaluation module, an area delineation management module, a feature movement range extraction module, and an anti-tailgating warning prompt management module;

[0031] The information recognition and extraction module is used to set a monitoring area at the door of the target short-term rental house according to the installation position of the monitoring detection device at the door of the target short-term rental house and the maximum detection range, obtain the monitoring video stream of the monitoring area in real time, and recognize the facial feature information of the people appearing in the monitoring video stream;

[0032] The information management module is used to collect and store the facial feature information collected in each unit period of the monitoring area during the day at each unit period, and establish a facial feature information database corresponding to each unit period in the monitoring area;

[0033] The risk coefficient calculation and evaluation module is used to evaluate and calculate the risk coefficient of the tenant of the target short-term rental house being followed in each unit time period by analyzing the information update changes of the facial feature information database constructed in each unit time period;

[0034] The area delineation management module is used to set the facial feature information with a cumulative number of repeated appearances greater than the number threshold in all historical facial feature information databases of each unit time period as the marking information corresponding to each unit time period; in all the monitoring video streams collected to obtain each marking information, the range area where the person corresponding to each marking information first appears and last appears in the display screen of the monitoring detection device is delineated;

[0035] The feature movement range extraction module is used to identify and extract the feature movement range of each mark information according to the distribution of the range area circled by the person corresponding to each mark information;

[0036] The anti-tailgating warning management module is used to receive the estimated return time sent in advance by the tenant of the target short-term rental house, and provide the tenant with an anti-tailgating warning based on the risk coefficient of the unit time period in which the estimated return time falls.

[0037] Preferably, the risk coefficient calculation and evaluation module includes an information update change management unit and a risk coefficient calculation and evaluation unit;

[0038] An information update change management unit, used to analyze the information update change of the face feature information database constructed in each unit time period;

[0039] The risk coefficient calculation and evaluation unit is used to evaluate and calculate the risk coefficient of the tenant of the target short-term rental house being followed in each unit time period.

[0040] Preferably, the area delineation management module includes a marking information extraction unit and a range area delineation management unit;

[0041] A marking information extraction unit is used to set the facial feature information with a cumulative number of repeated appearances greater than a threshold value in all historical facial feature information databases of each unit time period as marking information corresponding to each unit time period;

[0042] The range area delineation management unit is used to delineate the range area where the person corresponding to each marking information first appears and last appears in the display screen of the monitoring detection device in all the monitoring video streams collected to obtain each marking information.

[0043] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention divides a day into several unit time periods, and observes the flow of personnel in each unit time period by analyzing the update changes of the facial feature information collected by the monitoring and detection device in each unit time period, thereby realizing the evaluation and calculation of the risk coefficient of the tenants of the target short-term rental house being followed in each unit time period; based on the risk coefficient, the minimum danger duration referred to in the anti-tailing warning analysis in each unit time period is dynamically adjusted, and at the same time, the present invention refers to the movement trajectory that exists regularly in the facial feature information database, that is, it is very likely to be the movement trajectory of permanent residents when performing the anti-tailing warning analysis, and is committed to providing a method for analyzing and processing reference warning data that changes with the information of the surrounding personnel for short-term rental tenants whose rental environment is usually more complex, thereby improving the rental safety of the tenants of the target short-term rental house. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a flow chart of a short-term rental house tailgating warning method based on big data of the present invention;

[0045] Figure 2 The present invention is a structural schematic diagram of a short-term rental house tailgating warning system based on big data. DETAILED DESCRIPTION

[0046] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without making any creative work shall fall within the scope of protection of the present invention.

[0047] Example: Figure 1-Figure 2 As shown, the present invention provides a short-term rental house tailgating warning method based on big data, the method comprising:

[0048] Step S1: according to the installation position of the monitoring detection device at the door of the target short-term rental house and the maximum detection range, a monitoring area is set at the door of the target short-term rental house, a monitoring video stream of the monitoring area is obtained in real time, and facial feature information of the persons appearing in the monitoring video stream is recognized;

[0049] Step S2: At each unit period, the facial feature information collected in the monitoring area during each unit period of the day is collected and stored, and a facial feature information database corresponding to each unit period of the monitoring area is established;

[0050] Step S3: By analyzing the information update changes of the facial feature information database constructed in each unit time period, the risk coefficient of the tenant of the target short-term rental house being followed in each unit time period is evaluated and calculated;

[0051] Wherein, step S3 comprises:

[0052] Step S3-1: extracting historical facial feature information bases constructed in chronological order for each unit time period in a day; collecting the total number K of facial feature information contained in each historical facial feature information base, and the time interval between each two adjacent facial feature information being collected and stored in each historical facial feature information base, and obtaining the average time interval T presented in each historical facial feature information base;

[0053] Step S3-2: Calculate the information update index β=K×(1 / T) for each historical facial feature information base in each unit time period in turn; in all historical facial feature information bases in each unit time period, the total number N of facial feature information with a cumulative number of repeated occurrences greater than the number threshold;

[0054] Step S3-3: Evaluate and calculate the risk coefficient of the tenant of the target short-term rental house being followed in each unit time period 1 / α=[(β 1 +β 2 +...+β d ) / d]×N; where β 1 , β 2 , ..., β dRespectively represent the information update index corresponding to the 1st, 2nd, ..., dth historical face feature information database in each unit time period;

[0055] Step S4: setting facial feature information whose cumulative number of repeated appearances is greater than a threshold value in all historical facial feature information databases of each unit time period as marking information corresponding to each unit time period;

[0056] For example, the monitoring detection device collects and generates 7 historical facial feature information databases for the unit time period from 7:00 am to 9:00 am. Among them, the facial feature information of the same person exists in 5 historical facial feature information databases at the same time, and the preset number threshold is 4. Therefore, the facial feature information of the person can be set as the label information of the unit time period corresponding to 7:00 am to 9:00 am;

[0057] In all the surveillance video streams collected to obtain each tag information, the range area where the person corresponding to each tag information first appears and last appears in the display screen of the surveillance detection device is circled;

[0058] Wherein, step S4 comprises:

[0059] Step S4-1: extracting edge lines of the display screen of the monitoring detection device, constructing a plane graphic of the display screen of the monitoring detection device; calling the monitoring video stream presented by the monitoring detection device when acquiring each mark information;

[0060] Step S4-2: Mark the position point where the person corresponding to each mark information first appears in the monitoring video stream in the plane graph as a first-category position point, and mark the position point where the person corresponding to each mark information last appears in the monitoring video stream in the plane graph as a second-category position point;

[0061] Step S4-3: constructing a minimum circle A that can cover all first-category position points, and a minimum circle B that can cover all second-category position points for each piece of marking information;

[0062] Step S5: identifying and extracting the characteristic moving range of each mark information according to the distribution of the range area circled by the person corresponding to each mark information;

[0063] Wherein, step S5 comprises:

[0064] Step S5-1: for each piece of marking information, respectively, identify the overlap area P(A) between the minimum circle A and the plane figure of the monitoring and detection device display screen, and the overlap area P(B) between the minimum circle B and the plane figure of the monitoring and detection device display screen, wherein the area of ​​the overlap area P(A) is less than or equal to the area of ​​the minimum circle A, and the area of ​​the overlap area P(B) is less than or equal to the area of ​​the minimum circle B;

[0065] Step S5-2: extract the contour graph of the overlapped area P(A), translate the contour graph in the plane graph of the display screen of the monitoring and detection device along the direction of the overlapped area P(B), and stop when the intersection area with the overlapped area P(B) is greater than the area threshold, and mark the relevant areas along the path of the contour graph in the plane graph of the display screen of the monitoring and detection device, and use the area with the characteristic mark in the plane graph as the characteristic movement range of each mark information;

[0066] Step S6: receiving the estimated time of return sent in advance by the tenant of the target short-term rental house, and providing the tenant with an anti-tailgating warning reminder based on the risk coefficient of the unit time period in which the estimated time of return falls;

[0067] Wherein, step S6 comprises:

[0068] Step S6-1: If the expected time of returning home is within a certain unit time period, extract the characteristic movement range corresponding to each mark information of the certain unit time period, and mark the characteristic movement range corresponding to each mark information as a safe area; extract the risk coefficient of the tenant of the target short-term rental house being followed in a certain unit time period; set the minimum danger duration T according to the risk coefficient;

[0069] Step S6-2: When the monitoring detection device identifies facial feature information that does not belong to the marking information of a certain unit time period, and monitors that the person corresponding to the facial feature information is continuously in an area not marked as a safe area within the minimum danger duration T, the sound and light warning device is triggered.

[0070] In order to better implement the above method, a short-term rental house tailgating warning system is also proposed, which includes an information recognition and extraction module, an information management module, a risk coefficient calculation and evaluation module, an area delineation management module, a feature movement range extraction module, and an anti-tailgating warning prompt management module;

[0071] The information recognition and extraction module is used to set a monitoring area at the door of the target short-term rental house according to the installation position of the monitoring detection device at the door of the target short-term rental house and the maximum detection range, obtain the monitoring video stream of the monitoring area in real time, and recognize the facial feature information of the people appearing in the monitoring video stream;

[0072] The information management module is used to collect and store the facial feature information collected in each unit period of the monitoring area during the day at each unit period, and establish a facial feature information database corresponding to each unit period in the monitoring area;

[0073] The risk coefficient calculation and evaluation module is used to evaluate and calculate the risk coefficient of the tenant of the target short-term rental house being followed in each unit time period by analyzing the information update changes of the facial feature information database constructed in each unit time period;

[0074] Among them, the risk coefficient calculation and evaluation module includes an information update change management unit and a risk coefficient calculation and evaluation unit;

[0075] An information update change management unit, used to analyze the information update change of the face feature information database constructed in each unit time period;

[0076] A risk coefficient calculation and evaluation unit is used to evaluate and calculate the risk coefficient of the tenant of the target short-term rental house being followed in each unit time period;

[0077] The area delineation management module is used to set the facial feature information with a cumulative number of repeated appearances greater than the number threshold in all historical facial feature information databases of each unit time period as the marking information corresponding to each unit time period; in all the monitoring video streams collected to obtain each marking information, the range area where the person corresponding to each marking information first appears and last appears in the display screen of the monitoring detection device is delineated;

[0078] Among them, the area delineation management module includes a marking information extraction unit and a range area delineation management unit;

[0079] A marking information extraction unit is used to set the facial feature information with a cumulative number of repeated appearances greater than a threshold value in all historical facial feature information databases of each unit time period as marking information corresponding to each unit time period;

[0080] The range area delineation management unit is used to delineate the range area where the person corresponding to each mark information first appears and last appears in the display screen of the monitoring detection device in all the monitoring video streams collected to obtain each mark information;

[0081] The feature movement range extraction module is used to identify and extract the feature movement range of each mark information according to the distribution of the range area circled by the person corresponding to each mark information;

[0082] The anti-tailgating warning management module is used to receive the estimated return time sent in advance by the tenant of the target short-term rental house, and provide the tenant with an anti-tailgating warning based on the risk coefficient of the unit time period in which the estimated return time falls.

[0083] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A short-term rental house tailgating warning method based on big data, characterized by: The method comprises: Step S1: according to the installation position of the monitoring detection device at the door of the target short-term rental house and the maximum detection range, a monitoring area is set at the door of the target short-term rental house, a monitoring video stream of the monitoring area is obtained in real time, and facial feature information of the persons appearing in the monitoring video stream is recognized; Step S2: At each unit period, the facial feature information collected in the monitoring area during each unit period of the day is collected and stored, and a facial feature information database corresponding to each unit period of the monitoring area is established; Step S3: By analyzing the information update changes of the facial feature information database constructed in each unit time period, the risk coefficient of the tenant of the target short-term rental house being followed in each unit time period is evaluated and calculated; Step S4: In all historical facial feature information databases of each unit time period, facial feature information with cumulative repetition times greater than the threshold value is set as the tag information corresponding to each unit time period; in all surveillance video streams acquired to obtain each tag information, the range area where the person corresponding to each tag information first appears and last appears in the display screen of the monitoring detection device is circled; Step S5: identifying and extracting the characteristic moving range of each mark information according to the distribution of the range area circled by the person corresponding to each mark information; Step S6: receiving the estimated time of return sent in advance by the tenant of the target short-term rental house, and providing the tenant with an anti-tailgating warning reminder based on the risk coefficient of the unit time period in which the estimated time of return falls; The step S4 comprises: Step S4-1: extracting edge lines of the display screen of the monitoring detection device, constructing a plane graphic of the display screen of the monitoring detection device; calling the monitoring video stream presented by the monitoring detection device when acquiring each mark information; Step S4-2: marking the position point where the person corresponding to each mark information first appears in the monitoring video stream in the plane figure as a first type of position point, and marking the position point where the person corresponding to each mark information last appears in the monitoring video stream in the plane figure as a second type of position point; Step S4-3: constructing a minimum circle A that can cover all first-category position points, and a minimum circle B that can cover all second-category position points for each piece of marking information; The step S5 comprises: Step S5-1: for each piece of marking information, respectively, identify the overlap area P(A) between the minimum circle A and the plane figure of the display screen of the monitoring and detection device, and the overlap area P(B) between the minimum circle B and the plane figure of the display screen of the monitoring and detection device, wherein the area of ​​the overlap area P(A) is less than or equal to the area of ​​the minimum circle A, and the area of ​​the overlap area P(B) is less than or equal to the area of ​​the minimum circle B; Step S5-2: extracting the contour graph of the overlapped area P(A), translating the contour graph in the plane graph of the display screen of the monitoring detection device along the direction of the overlapped area P(B), and stopping when the intersection area with the overlapped area P(B) is greater than the area threshold, marking the relevant areas along the path of the contour graph in the plane graph of the display screen of the monitoring detection device, and using the areas with the characteristic marks in the plane graph as the characteristic movement range of each mark information; The step S6 comprises: Step S6-1: If the expected time of returning home is within a certain unit time period, extract the characteristic movement range corresponding to each mark information of the certain unit time period, and mark the characteristic movement range corresponding to each mark information as a safe area; extract the risk coefficient of the tenant of the target short-term rental house being followed in the certain unit time period; set the minimum danger duration T according to the risk coefficient; Step S6-2: When the monitoring detection device identifies facial feature information that does not belong to the marking information of the unit time period, and monitors that the person corresponding to the facial feature information is continuously in the area not marked as a safe area within the minimum danger duration T, the sound and light warning device is triggered.

2. According to claim 1, a short-term rental house tailgating warning method based on big data is characterized by: The step S3 comprises: Step S3-1: extracting historical facial feature information bases constructed in chronological order for each unit time period in a day; collecting the total number K of facial feature information contained in each historical facial feature information base, and the time interval between each two adjacent facial feature information being collected and stored in each historical facial feature information base, and obtaining the average time interval T presented in each historical facial feature information base; Step S3-2: Calculate the information update index β=K×(1 / T) for each historical facial feature information base in each unit time period in turn; in all historical facial feature information bases in each unit time period, the total number N of facial feature information with a cumulative number of repeated occurrences greater than the number threshold; Step S3-3: Evaluate and calculate the risk coefficient of the tenant of the target short-term rental house being followed in each unit time period 1 / α=[(β 1 +β 2 +...+β d ) / d]×N; where β 1 , β 2 , ..., β d They respectively represent the information update indexes corresponding to the 1st, 2nd, ..., dth historical face feature information databases in the unit time periods.

3. A short-term rental house tailgating warning system, used to execute a short-term rental house tailgating warning method based on big data as described in any one of claims 1-2, characterized in that: The system includes an information identification and extraction module, an information management module, a risk coefficient calculation and evaluation module, an area delineation management module, a feature movement range extraction module, and an anti-tailgating warning prompt management module; The information recognition and extraction module is used to set a monitoring area at the door of the target short-term rental house according to the installation position of the monitoring detection device at the door of the target short-term rental house and the maximum detection range, obtain the monitoring video stream of the monitoring area in real time, and recognize the facial feature information of the person appearing in the monitoring video stream; The information management module is used to collect and store the facial feature information collected in each unit period of the monitoring area during the day at each unit period, and establish a facial feature information database corresponding to each unit period of the monitoring area; The risk coefficient calculation and evaluation module is used to evaluate and calculate the risk coefficient of the tenant of the target short-term rental house being followed in each unit time period by analyzing the information update changes of the facial feature information database constructed in each unit time period; The area delineation management module is used to set the facial feature information with a cumulative number of repeated appearances greater than a threshold in all historical facial feature information databases of each unit time period as the marking information corresponding to each unit time period; In all the surveillance video streams collected to obtain each tag information, the range area where the person corresponding to each tag information first appears and last appears in the display screen of the surveillance detection device is circled; The feature movement range extraction module is used to identify and extract the feature movement range of each mark information according to the distribution of the range area circled by the person corresponding to each mark information; The anti-tailgating warning prompt management module is used to receive the estimated time of return sent in advance by the tenant of the target short-term rental house, and provide the tenant with an anti-tailgating warning prompt based on the risk coefficient of the unit time period in which the estimated time of return falls.

4. A short-term rental house tailgating warning system according to claim 3, characterized in that: The risk coefficient calculation and evaluation module includes an information update change management unit and a risk coefficient calculation and evaluation unit; The information update change management unit is used to analyze the information update change of the face feature information database constructed in each unit time period; The risk coefficient calculation and evaluation unit is used to evaluate and calculate the risk coefficient of the tenant of the target short-term rental house being followed in each unit time period.

5. The short-term rental house tailgating warning system according to claim 3 is characterized by: The area delineation management module includes a marking information extraction unit and a range area delineation management unit; The marking information extraction unit is used to set the facial feature information with a cumulative number of repeated appearances greater than a threshold value in all historical facial feature information databases in each unit time period as the marking information corresponding to each unit time period; The range area delineation management unit is used to delineate the range area where the person corresponding to each mark information first appears and last appears in the display screen of the monitoring detection device in all the monitoring video streams collected to obtain each mark information.

Images