A foothold analysis method, device and medium

Abstract

The present application discloses a landing point analysis method, device, and medium, wherein the method includes: after obtaining each monitoring data and the time span of each monitoring data, calling at least two analysis methods for different time spans to analyze each monitoring data, and incorporating each target landing point obtained by each analysis method into a target landing point set. Since different analysis methods are used for different time spans, the situation where the target landing point in the monitoring data with a small time span is ignored is avoided, thereby improving the accuracy and reliability of the target landing point analysis. In addition, the landing point analysis device and medium provided by the present application correspond to the above-mentioned landing point analysis method and have the same effect as above.

Description

Technical Field

[0001] The present application relates to the field of safety prevention technology, and in particular to a foothold analysis method, device and medium. Background Art

[0002] With the rapid development of technology in the security industry, the intelligent security monitoring systems in cities are becoming increasingly complete and sound. Therefore, the location information and time information displayed by the photos taken by the intelligent security monitoring system can be used to form the target's activity trajectory, so as to facilitate the analysis of the target's whereabouts.

[0003] Currently, target location analysis is performed by calculating the target's dwell time at each point in the monitoring data and identifying points with dwell times greater than a preset value as the target's destination. Because multiple monitoring data sets are collected, each with a different time span, determining the target's destination based solely on whether the target's dwell time exceeds the preset value can result in overlooking some target destinations in monitoring data with a shorter time span, as their dwell time does not exceed the preset value. This reduces the accuracy and reliability of target location analysis.

[0004] Therefore, the present application provides a landing point analysis method, device and medium, the purpose of which is to improve the accuracy and reliability of the landing point of the analysis target. Summary of the Invention

[0005] The purpose of this application is to provide a foothold analysis method, device and medium.

[0006] To solve the above technical problems, this application provides a foothold analysis method, including:

[0007] Obtaining each monitoring data and the time span of each monitoring data;

[0008] In the case where the time spans are inconsistent, calling at least two analysis methods to analyze the monitoring data;

[0009] Each target landing point obtained by the analysis method is included in the target landing point set.

[0010] Preferably, when the time spans are inconsistent, the step of calling at least two analysis methods to analyze the monitoring data includes:

[0011] In the case where there are both time spans not greater than the first threshold and time spans greater than the first threshold in the time spans, respectively calling the first analysis method and the second analysis method to analyze the monitoring data;

[0012] Preferably, the step of incorporating each target landing point obtained by the analysis method into a target landing point set comprises:

[0013] Using the first analysis method, the first monitoring data in each of the time spans that is not greater than the first threshold is analyzed to obtain a first target landing point set;

[0014] Using the second analysis method, the second monitoring data greater than the first threshold in each of the time spans is analyzed to obtain a second target landing point set;

[0015] The first target landing point set and the second target landing point set are included in the target landing point set.

[0016] Preferably, the step of analyzing the first monitoring data not greater than the first threshold in each of the time spans by the first analysis method to obtain the first target landing point set includes:

[0017] Obtain the target's identity information;

[0018] determining a set of suspicious addresses based on the identity information;

[0019] determining a first movement trajectory of the target according to the first monitoring data;

[0020] Calculating the distance difference between each suspicious address in the suspicious address set and each first point corresponding to the first activity trajectory;

[0021] If the distance difference does not exceed a second threshold, the corresponding suspicious address is included in the first target destination set;

[0022] When all the distance differences exceed the second threshold, the probability of each of the first points is calculated, and the first point corresponding to the maximum value of the probabilities is included in the first target landing point set.

[0023] Preferably, the step of analyzing the second monitoring data greater than the first threshold in each of the time spans by the second analysis method to obtain the second target landing point set includes:

[0024] dividing the second monitoring data into preset intervals;

[0025] Determining a second movement trajectory of the target corresponding to each of the preset intervals and a time set corresponding to the second movement trajectory;

[0026] Calculate an active foothold set and an inactive foothold set based on the second activity trajectory and the time collection;

[0027] The active landing point set and the inactive landing point set are included in the second target landing point set.

[0028] Preferably, calculating the active landing point set includes:

[0029] Selecting a second target movement trajectory corresponding to a preset target interval, and calculating a state probability vector corresponding to each second point in the second target movement trajectory;

[0030] Determining an average dwell time of each of the second points according to a target time set corresponding to the second target activity trajectory;

[0031] Calculating the state probability corresponding to each second point position according to each of the average residence times and each of the state probability vectors;

[0032] Determine the second point corresponding to the maximum value of the state probability as the target active landing point;

[0033] The target active landing point corresponding to the value with the highest occurrence frequency among all the target active landing points is included in the active landing point set.

[0034] Preferably, after selecting the second target movement trajectory corresponding to the preset target interval, the method further includes:

[0035] Calculating an average of the number of occurrences of all second points corresponding to the second activity trajectory of the target;

[0036] The second point position that is higher than the average value is selected as the point position participating in the calculation of the active landing point set.

[0037] Preferably, calculating the inactive landing point set includes:

[0038] Selecting a target second movement track corresponding to the target preset interval and a target time set corresponding to the target second movement track;

[0039] Calculating a third point position that appears repeatedly according to the second moving trajectory of the target;

[0040] Calculate the longest time interval between occurrences of each of the third points according to the target time set;

[0041] Determine the third point corresponding to the maximum value in the longest time interval as the target inactive landing point;

[0042] The target inactive landing point corresponding to the value with the highest occurrence frequency among all the target inactive landing points is included in the inactive landing point set.

[0043] Preferably, the monitoring data includes: human body image monitoring data, electronic fence monitoring data and vehicle checkpoint monitoring data.

[0044] In order to solve the above technical problems, the present application also provides a foothold analysis device, comprising:

[0045] A first acquisition module is used to acquire each monitoring data and the time span of each monitoring data;

[0046] A first calling module is configured to call at least two analysis methods to analyze the monitoring data when the time spans are inconsistent;

[0047] The first inclusion module is used to include each target landing point obtained by the analysis method into the target landing point set.

[0048] In order to solve the above technical problems, the present application also provides a foothold analysis device, comprising:

[0049] Memory for storing computer programs;

[0050] A processor is used to implement the steps of the above-mentioned foothold analysis method when executing the computer program.

[0051] In order to solve the above technical problems, the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the landing point analysis method as described above are implemented.

[0052] The landing point analysis method provided in this application, after obtaining each piece of monitoring data and its time span, analyzes each piece of monitoring data using at least two analysis methods for different time spans, and incorporates the target landing points obtained by each analysis method into a target landing point set. By using different analysis methods for different time spans, it avoids overlooking target landing points in monitoring data with a short time span, thereby improving the accuracy and reliability of target landing point analysis.

[0053] In addition, the present application provides a landing point analysis device and medium, which correspond to the above-mentioned landing point analysis method and have the same effect as above. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0055] Figure 1 A flowchart of a foothold analysis method provided in an embodiment of the present application;

[0056] Figure 2 A flowchart of another landing point analysis method provided in an embodiment of the present application;

[0057] Figure 3 A flowchart of a first analysis method provided in an embodiment of the present application;

[0058] Figure 4 A flowchart of a second analysis method provided in an embodiment of the present application;

[0059] Figure 5 An overall flow chart of a foothold analysis method provided in an embodiment of the present application;

[0060] Figure 6 A flowchart of calculating an active landing point set provided in an embodiment of the present application;

[0061] Figure 7 A statistical graph of the frequency of transitions between the second points in the target second activity trajectory on the first day provided in an embodiment of the present application;

[0062] Figure 8 Another flow chart for calculating the active landing point set provided by this application;

[0063] Figure 9 A flowchart of calculating an inactive landing point set provided in an embodiment of the present application;

[0064] Figure 10 A schematic structural diagram of a foothold analysis device provided in an embodiment of the present application;

[0065] Figure 11 A schematic structural diagram of another landing point analysis device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0066] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0067] The core of this application is to provide a foothold analysis method, device and medium.

[0068] In order to enable those skilled in the art to better understand the present application, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0069] Figure 1 This is a flowchart of a landing point analysis method provided in an embodiment of the present application. Figure 1 As shown, the method includes:

[0070] S10: Acquire each monitoring data and the time span of each monitoring data.

[0071] In the embodiments of the present application, the monitoring data is photos or videos taken by each monitoring device and including location data and time data. The time span of the monitoring data is the time range of the data stored in each monitoring device. For example, if an elevator monitoring device stores a maximum of seven days of monitoring data, the time span of the monitoring data is seven days. It is understood that the time span of different monitoring data may be different or the same.

[0072] S11: When the time spans are inconsistent, at least two analysis methods are called to analyze the monitoring data.

[0073] Specifically, when the time spans of each monitoring data item are divided into two time span categories, two analysis methods are invoked: the analysis method corresponding to the first time span category is invoked to analyze the monitoring data corresponding to the first time span category, and the analysis method corresponding to the second time span category is invoked to analyze the monitoring data corresponding to the second time span category. It is understood that, when the time spans of each monitoring data item are divided into three time span categories, three analysis methods are invoked to analyze the monitoring data.

[0074] It is understandable that the more detailed the categories in which the time span of each monitoring data is divided by time, the more analysis methods are called, and the more comprehensive and detailed the analysis results of the monitoring data are. However, the more complex the analysis process of the monitoring data is, the longer it takes. Therefore, in actual implementation, the number of categories in which the time span of each monitoring data is divided by time needs to match actual needs. If it is necessary to obtain the analysis results of the monitoring data quickly, the time span of each monitoring data can be divided into two categories. If it is necessary to obtain comprehensive analysis results of the monitoring data, the time span of each monitoring data can be divided into three or more categories.

[0075] S12: Adding each target landing point obtained by the analysis method into the target landing point set.

[0076] It should be noted that the corresponding target destinations obtained by different analysis methods must be included in the target destination set, that is, when the time span of each monitoring data is divided into two types of time spans according to time, the target destination points obtained by the analysis method corresponding to the first type of time span and the target destination points obtained by the analysis method corresponding to the second type of time span are both included in the target destination set.

[0077] It can be understood that when the time span of each monitoring data is divided into two types of time spans according to time, at least one target destination can be obtained by calling at least two analysis methods, that is, the analysis method corresponding to the first type of time span can obtain one target destination or multiple target destinations, and the analysis method corresponding to the second type of time span can obtain one target destination or multiple target destinations. When the target destination obtained by the analysis method corresponding to the first type of time span is repeated with the target destination obtained by the analysis method corresponding to the second type of time span, the corresponding target destination will be included in the target destination set after deduplication. Therefore, the target destination set contains only one target destination or multiple target destinations.

[0078] The landing point analysis method provided in the embodiments of the present application, after obtaining each piece of monitoring data and its time span, analyzes each piece of monitoring data using at least two analysis methods for different time spans, and incorporates each target landing point obtained by each analysis method into a target landing point set. By using different analysis methods for different time spans, the target landing points in monitoring data with a short time span are avoided from being overlooked, thereby improving the accuracy and reliability of the target landing point analysis.

[0079] On the basis of the above embodiment, when the time spans are inconsistent, the step of calling at least two analysis methods to analyze the monitoring data includes: when there are both time spans that are not greater than the first threshold and time spans that are greater than the first threshold in each time span, calling the first analysis method and the second analysis method respectively to analyze the monitoring data.

[0080] It should be noted that there is no restriction on the value of the first threshold. In order to quickly and accurately obtain the analysis results of the monitoring data, in the specific implementation, the first threshold can be set to 1 day, that is, each monitoring data is divided into the first category of monitoring data with a time span of less than 1 day and the second category of monitoring data with a time span of more than 1 day.

[0081] Figure 2 This is a flow chart of another landing point analysis method provided in an embodiment of the present application. Figure 2 As shown, based on the above embodiment, S12 specifically includes:

[0082] S20: Using a first analysis method, analyze the first monitoring data in each time span that is not greater than a first threshold to obtain a first target landing point set.

[0083] Among them, the first target landing point set can be an empty set or not. If it is not an empty set, there can be one first target landing point or multiple first target landing points. If the first target landing point set is an empty set, it means that there is no target landing point in the first monitoring data.

[0084] S21: Using a second analysis method, analyze the second monitoring data greater than the first threshold in each time span to obtain a second target landing point set.

[0085] Among them, the second target landing point set can be an empty set or not. If it is not an empty set, there can be one second target landing point or multiple second target landing points. If the second target landing point set is an empty set, it means that there is no target landing point in the second monitoring data.

[0086] S22: Add the first target landing point set and the second target landing point set into the target landing point set.

[0087] The landing point analysis method provided in the embodiment of the present application is a method in which each monitoring data is divided into first monitoring data and second monitoring data according to a first threshold value, and two analysis methods are used to analyze the monitoring data respectively. Therefore, compared with using at least three analysis methods, using two analysis methods reduces the workload of analyzing the monitoring data and improves the speed of analyzing the monitoring data.

[0088] Figure 3 This is a flow chart of a first analysis method provided in an embodiment of the present application. Figure 3 As shown, based on the above embodiment, S20 specifically includes:

[0089] S30: Obtain the target's identity information.

[0090] S31: Determine a suspicious address set based on the identity information.

[0091] It should be noted that the suspicious address set may include the target's permanent addresses such as the target's registered residence address, house purchase address, rental address, company address, etc.

[0092] S32: Determine a first movement trajectory of the target according to the first monitoring data.

[0093] In the embodiment of the present application, the first activity track is composed of the address data in the first monitoring data arranged in time sequence, and the first activity track is recorded as P n .

[0094] S33: Calculate the distance difference between each suspicious address in the suspicious address set and each first point corresponding to the first activity trajectory, and record the distance difference as K.

[0095] S34: Determine whether all distance differences exceed the second threshold. If yes, proceed to S35; if no, proceed to S36.

[0096] There is no restriction on the value of the second threshold. In a specific implementation, the second threshold can be set to 3 kilometers.

[0097] S35: Calculate the probability of each first point, and include the first point corresponding to the maximum value in the probability into the first target landing point set. n .

[0098] It should also be noted that the calculation method of the probability of each first point is as follows: each first point is divided into types, different weights are assigned to different types, and the probability of each first point is calculated based on the product of the probability of occurrence of each first point and the weight of the type corresponding to each first point, and one or more first points corresponding to the maximum value of the probability are included in the first target landing point set.

[0099] S36: Add the suspicious addresses corresponding to the distance differences that do not exceed the second threshold into the first target destination set.

[0100] It is understandable that if there are multiple suspicious addresses corresponding to distance differences that do not exceed the second threshold, all corresponding suspicious addresses are included in the first target destination set.

[0101] In order to make those skilled in the art more clear about the flowchart of the first analysis method provided in the embodiment of the present application, this embodiment will be specifically described by example:

[0102] After obtaining the target's identity information, determine the target's settlement address, house purchase address, and rental address based on the identity information. The settlement address is recorded as T1, the house purchase address is recorded as T2, and the rental address is recorded as T3. With n=3, the first activity trajectory P n Take P1, P2, and P3 as examples. Calculate the distance difference K between T1 and P1 respectively. 11 , the distance difference K between T1 and P2 12 , the distance difference K between T1 and P3 13 , the distance difference K between T2 and P1 21 , the distance difference K between T2 and P2 22 , the distance difference K between T2 and P3 23 , the distance difference K between T3 and P1 31 , the distance difference K between T3 and P2 32 , and the distance difference K between T3 and P3 33 . Determine whether all the above distance differences exceed the second threshold. If the distance difference K between T1 and P1 is 11If the second threshold is not exceeded, the settlement address T1 corresponding to K11 is included in the first target settlement point set. If all the above distance differences exceed the second threshold, all the first points are divided into types, and different types are assigned different weights. The probability X of each first point is calculated according to the probability formula. n , the probability X n The first point corresponding to the maximum value in is included in the first target landing point set, where the probability formula is:

[0103] X n =C n *α n

[0104] Among them, X n is the probability of the first point n, C n is the number of times the first point n appears, α n is the weight of the type corresponding to the first point n.

[0105] Table 1 shows the three types of residential, commercial, and industrial areas and their weights. Taking Table 1 as an example, when the first activity trajectory is P1, P2, and P3, where P1 is a residential area, P2 is a commercial area, and P3 is an industrial area, and P1, P2, and P3 appear only once, then the probability of P1 is X P1 is 0.5, the probability of P2 is X P2 is 0.3, the probability of P3 is X P3 is 0.2. Among these three probabilities, X P1 Maximum, so X P1 The corresponding P1 is included in the first target landing point set.

[0106] Table 1

[0107] type residential area commercial district Industrial Zone Weight 0.5 0.3 0.2

[0108] The landing point analysis method provided in the embodiment of the present application calculates the distance difference between each suspicious address and each first point. If the distance difference is not greater than the second threshold, the corresponding point is included in the target landing point set. If the distance difference is less than the second threshold, the probability of each first point is weighted according to the point type, and the point corresponding to the maximum probability is included in the target landing point set. Therefore, compared with the existing technical methods, the workload of calculating the distance difference and / or calculating the weighted probability is smaller and the accuracy is higher, thereby further improving the speed of analyzing monitoring data and the accuracy of the analysis results.

[0109] In the above, the first analysis method is described in detail, and the second analysis method mentioned in this application will be described in detail below.

[0110] Figure 4This is a flow chart of a second analysis method provided in an embodiment of the present application. Figure 4 As shown, based on the above embodiment, S21 specifically includes:

[0111] S40: Divide the second monitoring data into preset intervals.

[0112] In the embodiment of the present application, the preset interval is not specifically limited. In order to quickly and comprehensively obtain the analysis results of the second monitoring data, in a specific implementation, the preset interval can be set to 1 day.

[0113] S41: Determine a second movement trajectory of the target corresponding to each preset interval and a time set corresponding to the second movement trajectory.

[0114] The second activity trajectory is a set of location data in the second monitoring data within each preset interval, and the time set is a set of time data corresponding to the location data in the second monitoring data within each preset interval.

[0115] S42: Calculate an active landing point set and an inactive landing point set according to the second activity trajectory and the time collection corresponding to the second activity trajectory.

[0116] It should be noted that active locations are the locations where the target appears most frequently in each preset interval and has a high probability of reappearing. The active location set is the collection of one or more active locations in each preset interval. Inactive locations are locations that appear less frequently in each preset interval but conform to social activity patterns. For example, an office worker may only appear in their residential complex twice a day, but this location could be a target's location. The inactive location set is the collection of one or more inactive locations in each preset interval.

[0117] S43: Add the active destination set and the inactive destination set into the second target destination set.

[0118] It is understandable that the second target destination point set may include only one active destination point and / or one inactive destination point, or may include multiple active destination points and / or multiple inactive destination points.

[0119] In order to make those skilled in the art more clear about the process of the landing point analysis method provided in this application, Figure 5As shown, an embodiment of the present application also provides an overall flow chart of a landing point analysis method, wherein S100 corresponds to the step of respectively calling the first analysis method and the second analysis method to analyze the monitoring data when there are both time spans not greater than the first threshold and time spans greater than the first threshold in each time span, and S101 and S102 correspond to S42, so the specific implementation methods of S100, S101 and S102 are not described here in detail.

[0120] In order to avoid errors in the monitoring data and further improve the accuracy and reliability of the landing point of the analysis target, as a preferred embodiment, before S40, the following step may be further included: filtering interference data.

[0121] The interference data includes data with facial recognition below the third threshold and error data. It should be noted that error data refers to points that need to be reached at a speed beyond the target capability range, such as three adjacent points A, B, and C in the monitoring data, where the distance between A and B is N. AB , the time difference is T AB , the distance between B and C is N BC , the time difference is T BC , the distance between A and C is N AC , the time difference is T AC .if:

[0122]

[0123] Point B is considered erroneous data. It should be noted that while the examples provided in this application describe the maximum speed as the maximum speed on a highway, this does not necessarily mean that this is the only maximum speed. In specific implementations, the maximum speed can be set to match the requirements. It is understood that there is no restriction on the value of the third threshold; in specific implementations, the third threshold can be set to 80%.

[0124] The landing point analysis method provided in the embodiment of the present application analyzes the target's landing point from two aspects: active landing points and inactive landing points for monitoring data with a large time span. Therefore, compared with the existing method, this embodiment not only considers active landing points with a high probability of occurrence, but also considers inactive landing points with a low probability of occurrence that conforms to the laws of social activities, thereby improving the comprehensiveness of analyzing monitoring data with a large time span.

[0125] Figure 6 This is a flow chart of calculating an active landing point set provided in an embodiment of the present application. Figure 6 As shown, based on the above embodiment, calculating the active landing point set specifically includes:

[0126] S50: Selecting a second target moving track corresponding to the preset target interval.

[0127] S51: Calculating a state probability vector corresponding to each second point in the second target movement trajectory according to the target second movement trajectory.

[0128] It should be noted that the state probability vector is calculated as follows: the transition probability matrix is ​​calculated by the ratio of the transfer frequency between each second point in the target's second activity trajectory to the number of times each second point appears, and then the state probability vector of each second point is obtained according to the following formula.

[0129] According to the sum of all state probabilities being 1, we can get:

[0130]

[0131] According to the post-ineffectiveness of the Markov process and the Bayesian conditional probability formula, we can get:

[0132]

[0133] By recursion, we can get:

[0134] π(k)=π(k-1)P=π(0)P k

[0135] Among them, π j (k) is the state probability vector of the target at point j after the kth transfer in the second activity trajectory, P ij is the transition probability matrix from point i to point j; P is the transition probability matrix corresponding to the second activity trajectory of the target; π(0) is the initial state probability vector; N is the number of second points in the second activity trajectory of the target; n is the nth point in the second activity trajectory of the target.

[0136] Take the second target activity trajectory as D, E, F as an example, where D to E is the first transfer and E to F is the second transfer, then π E (1) is the probability that the target is at point E after the first transfer in the second activity trajectory, π(0) = (1, 0, 0).

[0137] Let the state probability vector π=[π1,π2,…,π n ], then:

[0138] π i =log k→∞ π i (k) (i=1,2…n)

[0139] log k→∞ π i (k) = log k→∞ π i (k+1)=π

[0140] Among them, π i is the state probability vector of the i-th point in the second activity trajectory of the target; i (k) is the state probability vector of the target at point i after the k-th transfer in the second activity trajectory.

[0141] By reorganizing and calculating the above formula, we can finally obtain π = π P. Based on π = π P and the transition probability matrix corresponding to the second target activity trajectory, the state probability vector corresponding to each second point in the second target activity trajectory can be calculated.

[0142] S52: Determine the average stay time of each second point according to the target time set corresponding to the second target activity trajectory.

[0143] It is understandable that the average dwell time is the average of each dwell time at the second point. Taking point A in the second activity trajectory of the target as an example, the average dwell time is calculated as follows:

[0144]

[0145] Where, ΔT A is the average residence time of point A in the second activity trajectory of the target; T1 is the residence time between the first appearance of point A and the second appearance of point A; T2 is the residence time between the second appearance of point A and the third appearance of point A; T m It is the dwell time between the moment when point A appears for the mth time and the moment when point A appears for the m+1th time.

[0146] S53: Calculate the state probability corresponding to each second point according to each average residence time and each state probability vector.

[0147] S54: Determine the second point corresponding to the maximum value in the state probability as the target active landing point.

[0148] S55: Add the target active landing point corresponding to the value with the highest occurrence frequency among all target active landing points into the active landing point set.

[0149] It can be understood that there can be one target active landing point or multiple target active landing points in the active landing point set. If the frequency of occurrence of all target active landing points is the same, the active landing point set is an empty set. For example, the second monitoring data can only be divided into monitoring data of the first preset interval and monitoring data of the second preset interval. The target active landing point corresponding to the monitoring data of the first preset interval is E, and the target active landing point corresponding to the monitoring data of the second preset interval is F. The number of occurrences of point E and point F are both 1. Then the target active landing points corresponding to the monitoring data of the first preset interval and the monitoring data of the second preset interval are not included in the active landing point set. At this time, the active landing point set is an empty set.

[0150] In order to make those skilled in the art more clear about the flowchart of calculating the active foothold set provided in the embodiment of the present application, this embodiment will be specifically described by example, in which the preset interval is set to 1 day:

[0151] According to the second monitoring data, a second activity trajectory and a set of time points corresponding to the second activity trajectory are obtained:

[0152] P n ={[(A, B, C,...), (A, F, E,...),..., (B, G, D,...)]}

[0153] T n ={[(T A , T B , T C ,...),(T A , T F , T E ,...),…,(T B , T G , T D , ...)]}

[0154] The time series corresponding to the target's second activity trajectory on the first day was selected, and the average dwell time at each second point was calculated. The number of occurrences of each second point in the target's second activity trajectory on the first day was obtained, as shown in Table 2.

[0155] Table 2

[0156] Point Occurrences A 10 B 10 C 10

[0157] At the same time, the transfer frequency between the second points is obtained according to the target second activity trajectory on the first day, such as Figure 7As shown, it was transferred from point A to point A 5 times, point A to point B 3 times, point A to point C 2 times, point B to point A 3 times, point B to point B 4 times, point B to point C 3 times, point C to point A 3 times, point C to point B 3 times, and point C to point C 4 times.

[0158] The first day's transition probability matrix is ​​obtained by the ratio of the transfer frequency between each point in the target second activity trajectory on the first day to the number of occurrences of each point. For example, if there are 5 transfers from point A to point A and point A appears 10 times, then the ratio of 5 to 10 is the data of the first row and the first case in the first day's transition probability matrix. Similarly, the first day's transition probability matrix can be obtained as follows:

[0159]

[0160] Let the state probability vector π=[π A ,π B ,π C ], and substituting π=πP, we get:

[0161]

[0162] The above formula is used to calculate π A =0.375,π B =0.333,π C =0.292.

[0163] By P A =π A *ΔT A , P B =π B *ΔT B , P C =π C *ΔT C Calculate the state probabilities of each second point corresponding to the target's second activity trajectory on the first day, and select the maximum probability as the target active landing point corresponding to the target's second activity trajectory on the first day. Finally, add the target active landing point with the highest occurrence frequency among all target active landing points to the active landing point set.

[0164] Figure 8 Another flow chart for calculating the active landing point set provided by this application. Figure 8 As shown, after S50, it also includes:

[0165] S60: Calculate the average number of occurrences of all second points corresponding to the second target movement trajectory.

[0166] S61: Select a second point that is higher than the average value, so as to be used as a point for participating in the calculation of the active landing point set.

[0167] The landing point analysis method provided in the embodiment of the present application selects points that are higher than the average value as points involved in calculating active landing points, thereby avoiding the problem of too many points participating in the calculation of active landing points resulting in a large computational workload, thereby improving the speed of analyzing the target landing points.

[0168] Figure 9 This is a flow chart of calculating an inactive landing point set provided by an embodiment of the present application. Figure 9 As shown, the calculation method of the inactive landing point set includes:

[0169] S70: Selecting a target second movement track corresponding to the target preset interval and a target time set corresponding to the target second movement track.

[0170] S71: Calculate the third point position that appears repeatedly according to the second moving trajectory of the target.

[0171] It should be noted that the third point position is a set of second point positions that appear repeatedly in the second moving track of the target.

[0172] S72: Calculate the longest time interval between the appearances of the third points according to the target time set.

[0173] It should also be noted that the longest time interval is the interval between the first and last appearance times of each third point.

[0174] S73: Determine the third point corresponding to the maximum value in the longest time interval as the target inactive landing point.

[0175] S74: Add the target inactive landing point corresponding to the value with the highest occurrence frequency among all target inactive landing points into the inactive landing point set.

[0176] It can be understood that there can be one target inactive landing point or multiple target inactive landing points in the inactive landing point set. If the frequency of appearance of all target inactive landing points is the same, the inactive landing point set is an empty set. For example, the second monitoring data can only be divided into monitoring data of the first preset interval and monitoring data of the second preset interval. The target inactive landing point corresponding to the monitoring data of the first preset interval is G, and the target inactive landing point corresponding to the monitoring data of the second preset interval is H. The number of occurrences of point G and point H are both 1. Then the target inactive landing points corresponding to the monitoring data of the first preset interval and the monitoring data of the second preset interval are not included in the inactive landing point set. At this time, the inactive landing point set is an empty set.

[0177] In order to make those skilled in the art more clear about the flowchart of calculating the inactive foothold set provided in the embodiment of the present application, this embodiment will be specifically described by example, in which the preset interval is set to 1 day:

[0178] According to the second monitoring data, a second activity trajectory and a set of time points corresponding to the second activity trajectory are obtained:

[0179] P n ={[(A, B, C,...), (A, F, E,...),..., (B, G, D,...)]}

[0180] T n ={[(T A , T B , T C ,...),(T A , T F , T E ,...),…,(T B , T G , T D , ...)]}

[0181] The target second activity trajectory on the first day is P1 = [A, B, C, D, A, B, A], and the time set corresponding to the target second activity trajectory on the first day is T1 = [T A1 ,T B1 ,T C ,T D ,T A2 ,T B2 ,T A3 ].

[0182] From the second activity trajectory of the target on the first day, we know that the third point that appears repeatedly is point A and point B. Calculate the longest time interval ΔT of point A respectively. A The longest time interval ΔT between point B and point B B , where ΔT A and ΔT B is calculated as follows:

[0183] ΔT A =T A3 -T A1

[0184] ΔT B =T B2 -T B1

[0185] If ΔT A >ΔT B, then the target inactive landing point corresponding to the second target activity trajectory on the first day is point A. Finally, the target inactive landing point corresponding to the highest frequency value among all target inactive landing points is included in the inactive landing point set.

[0186] The landing point analysis method provided in the embodiment of the present application determines the target inactive landing point by calculating the longest time interval between repeatedly appearing points. The calculation method of the longest time interval is simple and the workload is small. Therefore, the analysis results of the inactive landing points in the second monitoring data can be quickly obtained, thereby improving the speed of analyzing the target landing point.

[0187] Based on the above embodiment, the monitoring data may include: human image monitoring data, electronic fence monitoring data, and vehicle checkpoint monitoring data. The target's movement trajectory can be obtained from the human image monitoring data, the target's mobile phone movement trajectory can be obtained from the electronic fence monitoring data, and the target's vehicle movement trajectory can be obtained from the vehicle checkpoint monitoring data.

[0188] In the embodiment of the present application, the three types of data, namely, the target's movement trajectory, the target's mobile phone movement trajectory, and the target's vehicle movement trajectory, are merged in time order, and after deduplication of data of the same time (within one minute), the same location, and different types, a first activity trajectory or a second activity trajectory, and a time set corresponding to the first activity trajectory or the second activity trajectory are obtained:

[0189] P n ={[(A, B, C,...), (A, F, E,...),..., (B, G, D,...)]}

[0190] T n ={[(T A , T B , T C ,...),(T A , T F , T E ,...),…,(T B , T G , T D , ...)]}

[0191] The method for analyzing the location of a target provided by an embodiment of the present application uses monitoring data including human image monitoring data, electronic fence monitoring data, and vehicle checkpoint monitoring data. Based on the monitoring data, three types of data are generated: the target's movement trajectory, the target's mobile phone movement trajectory, and the target's vehicle movement trajectory. After merging and deduplicating these three types of data, a first activity trajectory or a second activity trajectory is obtained. Because the first activity trajectory or the second activity trajectory includes all three types of monitoring data, the scope of the monitoring data is expanded, the coverage of points in the first activity trajectory or the second activity trajectory is increased, and the accuracy and reliability of analyzing the target's location of a target is further improved.

[0192] In the above embodiments, the foothold analysis method is described in detail. This application also provides corresponding embodiments of the foothold analysis device. It should be noted that this application describes the embodiments of the device from two perspectives: one is based on the functional module perspective, and the other is based on the hardware perspective.

[0193] Figure 10 This is a schematic diagram of the structure of a foothold analysis device provided in an embodiment of the present application. Figure 10 As shown, based on the perspective of functional modules, the device includes:

[0194] The first acquisition module 10 is used to acquire each monitoring data and the time span of each monitoring data.

[0195] The first calling module 11 is configured to call at least two analysis methods to analyze the monitoring data when the time spans are inconsistent.

[0196] The first incorporation module 12 is used to incorporate each target landing point obtained by the analysis method into the target landing point set.

[0197] As a preferred embodiment, the first inclusion module 12 includes:

[0198] The first analysis module is used to analyze the first monitoring data that is not greater than a first threshold in each time span using a first analysis method to obtain a first target landing point set.

[0199] The second analysis module is used to analyze the second monitoring data greater than the first threshold in each time span using a second analysis method to obtain a second target landing point set.

[0200] The second inclusion module is used to include the first target destination set and the second target destination set into the target destination set.

[0201] The first analysis module includes:

[0202] The second acquisition module is used to obtain the identity information of the target.

[0203] The first determining module is configured to determine a suspicious address set according to identity information.

[0204] The second determining module is configured to determine a first activity trajectory of the target according to the first monitoring data.

[0205] The first calculation module is configured to calculate the distance difference between each suspicious address in the suspicious address set and each first point corresponding to the first activity trajectory.

[0206] The third inclusion module is configured to include the corresponding suspicious address into the first target destination set when the distance difference does not exceed the second threshold.

[0207] The fourth inclusion module is used to calculate the probability of each first point when all distance differences exceed the second threshold, and include the first point corresponding to the maximum value among the probabilities into the first target landing point set.

[0208] The second analysis module includes:

[0209] The first dividing module is used to divide the second monitoring data into preset intervals.

[0210] The third determining module is configured to determine a second movement trajectory of the target corresponding to each preset interval and a time set corresponding to the second movement trajectory.

[0211] The second calculation module is used to calculate the active foothold set and the inactive foothold set according to the second activity trajectory and the time collection.

[0212] The fifth inclusion module is used to include the active destination set and the inactive destination set into the second target destination set.

[0213] The second calculation module includes:

[0214] The third calculation module is used to select the second target movement trajectory corresponding to the preset target interval, and calculate the state probability vector corresponding to each second point in the second target movement trajectory.

[0215] The fourth determining module is configured to determine an average dwell time of each second point according to a target time set corresponding to the target second activity trajectory.

[0216] The fourth calculation module is used to calculate the state probability corresponding to each second point according to each average residence time and each state probability vector.

[0217] The fifth determination module is used to determine that the second point corresponding to the maximum value in the state probability is the target active landing point.

[0218] The sixth inclusion module is used to include the target active landing point corresponding to the value with the highest occurrence frequency among all target active landing points into the active landing point set.

[0219] The second calculation module also includes:

[0220] The fifth calculation module is used to calculate the average number of occurrences of all second points corresponding to the second activity trajectory of the target.

[0221] The first selection module is used to select a second point position that is higher than the average value so as to be used as a point position for participating in the calculation of the active landing point set.

[0222] The second calculation module also includes:

[0223] The second selection module is configured to select a target second movement track corresponding to a target preset interval and a target time set corresponding to the target second movement track.

[0224] The sixth calculation module is used to calculate the third point position that appears repeatedly according to the second moving trajectory of the target.

[0225] The seventh calculation module is used to calculate the longest time interval between the appearances of each third point position according to the target time set.

[0226] The sixth determining module is configured to determine a third point corresponding to the maximum value in the longest time interval as a target inactive landing point.

[0227] The seventh inclusion module is used to include the target inactive landing point corresponding to the value with the highest occurrence frequency among all target inactive landing points into the inactive landing point set.

[0228] Since the embodiments of the apparatus part correspond to the embodiments of the method part, please refer to the description of the embodiments of the method part for the embodiments of the apparatus part, and they will not be repeated here.

[0229] The landing point analysis device provided in the embodiments of the present application, after acquiring each piece of monitoring data and its time span, analyzes each piece of monitoring data using at least two analysis methods for different time spans, and incorporates each target landing point obtained by each analysis method into a target landing point set. By using different analysis methods for different time spans, the target landing points in monitoring data with a short time span are avoided from being overlooked, thereby improving the accuracy and reliability of target landing point analysis.

[0230] Figure 11 This is a schematic diagram of the structure of another landing point analysis device provided in an embodiment of the present application. Figure 11 As shown, based on the hardware structure, the device includes:

[0231] Memory 20, for storing computer programs;

[0232] The processor 21 is configured to implement the steps of the foothold analysis method in the above embodiment when executing a computer program.

[0233] Among them, the processor 21 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 21 can be implemented in at least one hardware form of digital signal processing (DSP), field-programmable gate array (FPGA), and programmable logic array (PLA). The processor 21 may also include a main processor and a coprocessor. The main processor is a processor for processing data in the awake state, also known as a central processing unit (CPU); the coprocessor is a low-power processor for processing data in the standby state. In some embodiments, the processor 21 may be integrated with a graphics processing unit (GPU), which is responsible for rendering and drawing the content to be displayed on the display screen.

[0234] The memory 20 may include one or more computer-readable storage media, which may be non-transitory. The memory 20 may also include a high-speed random access memory, and a non-volatile memory, such as one or more disk storage devices, flash memory storage devices. In this embodiment, the memory 20 is at least used to store the following computer program 201, wherein, after the computer program is loaded and executed by the processor 21, it can implement the relevant steps of the landing point analysis method disclosed in any of the aforementioned embodiments. In addition, the resources stored in the memory 20 may also include an operating system 202 and data 203, etc., and the storage method may be temporary storage or permanent storage. Among them, the operating system 202 may include Windows, Unix, Linux, etc. The data 203 may include but is not limited to the data involved in the landing point analysis method, etc.

[0235] In some embodiments, the foothold analysis device may further include a display screen 22 , an input / output interface 23 , a communication interface 24 , a power supply 25 , and a communication bus 26 .

[0236] Those skilled in the art will understand that Figure 11 The structure shown in the figure does not constitute a limitation of the landing point analysis device, and may include more or fewer components than shown in the figure.

[0237] The landing point analysis device provided in an embodiment of the present application includes a memory and a processor. When executing a program stored in the memory, the processor can implement the following method: after obtaining each piece of monitoring data and the time span of each piece of monitoring data, the processor uses at least two analysis methods for different time spans to analyze each piece of monitoring data, and incorporates each target landing point obtained by each analysis method into a target landing point set. Because different analysis methods are used for different time spans, the target landing points in monitoring data with a short time span are avoided from being overlooked, thereby improving the accuracy and reliability of the target landing point analysis.

[0238] Finally, the present application also provides an embodiment corresponding to a computer-readable storage medium. The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps described in the above method embodiment.

[0239] It is understandable that if the method in the above embodiment is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium and executes all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program code, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0240] The computer-readable storage medium provided in an embodiment of the present application stores a computer program. When executed by a processor, the computer program can implement the following method: after obtaining each piece of monitoring data and the time span of each piece of monitoring data, at least two analysis methods are invoked for different time spans to analyze each piece of monitoring data, and each target landing point obtained by each analysis method is incorporated into a target landing point set. Because different analysis methods are used for different time spans, the situation in which target landing points in monitoring data with a short time span are overlooked is avoided, thereby improving the accuracy and reliability of the target landing point analysis.

[0241] The above is a detailed introduction to a foothold analysis method, device and medium provided by the present application. The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same and similar parts between the various embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of this application, several improvements and modifications can be made to the present application, and these improvements and modifications also fall within the scope of protection of the claims of this application.

[0242] It should also be noted that, in this specification, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus comprising the element.

Claims

1. A foothold analysis method, characterized in that: include: Obtaining each monitoring data and the time span of each monitoring data; In the case where the time spans are inconsistent, calling at least two analysis methods to analyze the monitoring data; Adding each target landing point obtained by the analysis method into the target landing point set; Wherein, when the time spans are inconsistent, the step of calling at least two analysis methods to analyze the monitoring data includes: In the case where there are both time spans not greater than the first threshold and time spans greater than the first threshold in the time spans, respectively calling the first analysis method and the second analysis method to analyze the monitoring data; The step of incorporating the target landing points obtained by the analysis method into the target landing point set includes: Using the first analysis method, the first monitoring data in each of the time spans that is not greater than the first threshold is analyzed to obtain a first target landing point set; Using the second analysis method, the second monitoring data greater than the first threshold in each of the time spans is analyzed to obtain a second target landing point set; The first target landing point set and the second target landing point set are included in the target landing point set.

2. The foothold analysis method according to claim 1, characterized in that: The step of analyzing the first monitoring data in each time span that is not greater than the first threshold by using the first analysis method to obtain a first target landing point set includes: Obtain the target's identity information; determining a set of suspicious addresses based on the identity information; determining a first movement trajectory of the target according to the first monitoring data; Calculating the distance difference between each suspicious address in the suspicious address set and each first point corresponding to the first activity trajectory; If the distance difference does not exceed a second threshold, the corresponding suspicious address is included in the first target destination set; When all the distance differences exceed the second threshold, the probability of each of the first points is calculated, and the first point corresponding to the maximum value of the probabilities is included in the first target landing point set.

3. The foothold analysis method according to claim 1, characterized in that: The step of analyzing the second monitoring data greater than the first threshold in each of the time spans by the second analysis method to obtain a second target landing point set includes: dividing the second monitoring data into preset intervals; Determining a second movement trajectory of the target corresponding to each of the preset intervals and a time set corresponding to the second movement trajectory; Calculate an active foothold set and an inactive foothold set based on the second activity trajectory and the time collection; The active landing point set and the inactive landing point set are included in the second target landing point set.

4. The foothold analysis method according to claim 3, characterized in that: Calculating the active landing point set includes: Selecting a second target movement trajectory corresponding to a preset target interval, and calculating a state probability vector corresponding to each second point in the second target movement trajectory; Determining an average dwell time of each of the second points according to a target time set corresponding to the second target activity trajectory; Calculating the state probability corresponding to each second point position according to each of the average residence times and each of the state probability vectors; Determine the second point corresponding to the maximum value of the state probability as the target active landing point; The target active landing point corresponding to the value with the highest occurrence frequency among all the target active landing points is included in the active landing point set.

5. The foothold analysis method according to claim 4, characterized in that: After selecting the second target movement track corresponding to the preset target interval, the method further includes: Calculating an average of the number of occurrences of all second points corresponding to the second activity trajectory of the target; The second point position that is higher than the average value is selected as the point position participating in the calculation of the active landing point set.

6. The foothold analysis method according to claim 3, characterized in that: Calculating the inactive landing point set includes: Selecting a target second movement track corresponding to the target preset interval and a target time set corresponding to the target second movement track; Calculate a third point position that appears repeatedly according to the second moving trajectory of the target; Calculate the longest time interval between occurrences of each of the third points according to the target time set; Determine the third point corresponding to the maximum value in the longest time interval as the target inactive landing point; The target inactive landing point corresponding to the value with the highest occurrence frequency among all the target inactive landing points is included in the inactive landing point set.

7. The foothold analysis method according to any one of claims 1 to 6, characterized in that: The monitoring data includes: human body image monitoring data, electronic fence monitoring data and vehicle checkpoint monitoring data.

8. A foothold analysis device, characterized in that: include: A first acquisition module is used to acquire each monitoring data and the time span of each monitoring data; A first calling module is configured to call at least two analysis methods to analyze the monitoring data when the time spans are inconsistent; A first inclusion module is used to include each target landing point obtained by the analysis method into a target landing point set; The step of analyzing the monitoring data by the first calling module includes: when there are both time spans not greater than a first threshold and time spans greater than the first threshold in each of the time spans, respectively calling the first analysis method and the second analysis method to analyze the monitoring data; The first inclusion module includes: a first analysis module, configured to analyze the first monitoring data not greater than the first threshold in each of the time spans using the first analysis method to obtain a first target landing point set; a second analysis module, configured to analyze the second monitoring data greater than the first threshold in each of the time spans using the second analysis method to obtain a second target landing point set; The second inclusion module is used to include the first target destination set and the second target destination set into the target destination set.

9. A foothold analysis device, characterized in that: include: memory for storing computer programs; A processor, configured to implement the steps of the foothold analysis method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the foothold analysis method according to any one of claims 1 to 7 are implemented.

Images