CN-122002221-A - Crime place identification method based on track data and base station positioning data

CN122002221ACN 122002221 ACN122002221 ACN 122002221ACN-122002221-A

Abstract

The invention discloses a crime place identification method based on track data and base station positioning data, and relates to the technical field of electronic data evidence obtaining and data mining. The method comprises the steps of S1, obtaining suspect track positioning data and common mobile phone base station positioning data, adding source confidence and application labels to the obtained data, S2, carrying out abnormal cleaning and data standardization on the obtained data to construct a standardized track data set, S3, defining the base stations as units and time periods as sets, calculating the activity values of the base stations in different sets, S4, calculating the common degree value of the base stations based on the coverage condition of the sets to obtain calculated activity-common degree values, and determining crime places. The invention meets the analysis requirements of the analyst on the suspected crime place. An analyst can more accurately analyze crime venues from suspect trajectory data and a large number of nonsensical base station locations.

Inventors

WANG JUNYE
ZENG CHAO
ZHANG LIXIANG
CHEN WEIQIANG
YAN YUBIN

Assignees

国投智能(南京)信息科技有限公司

Dates

Publication Date: 20260508
Application Date: 20251127

Claims (9)

1. A crime place identification method based on track data and base station positioning data is characterized by comprising the following steps: s1, acquiring suspect track positioning data and common mobile phone base station positioning data, and adding source confidence and use labels to the acquired data; s2, carrying out abnormal cleaning and data standardization on the acquired data to construct a standardized track data set; s3, defining the base stations as units and the time period as a set, and calculating the active values of each base station in different sets; And S4, calculating the common degree value of each base station based on the coverage condition of the set to obtain a calculation activity-common degree value, and determining the crime place.
2. The crime scene recognition method based on trajectory data and base station positioning data according to claim 1, wherein the abnormality washing comprises: Re-estimating coordinates of the trajectory positioning data of the suspected person according to the reachable speed and the acceleration threshold, and deleting if the situation that the positioning coordinates are obviously abnormal exists; and positioning data of the mobile phone base station of the common person, and removing repeated records and unreasonable connection time data.
3. The crime scene recognition method based on the trajectory data and the base station positioning data according to claim 2, wherein the data normalization includes: uniformly converting a coordinate system of the suspect track positioning data and the common mobile phone base station positioning data into a common geographic coordinate system; Projecting the coordinate track to a road network, and reducing the positioning error propagation of the base station sector; Aligning the asynchronous samples to a unified window; The time information is uniformly formatted as a time stamp in seconds.
4. The crime scene recognition method based on trajectory data and base station positioning data according to claim 3, wherein the calculation of the activity value comprises: Wherein, the Indicating the active value of the ith base station in the jth set, Indicating the number of times the ith base station appears in the jth set.
5. The crime scene recognition method based on the trajectory data and the base station positioning data according to claim 4, wherein the calculation of the ordinary degree value comprises: Wherein, the Represents the normal value of the i-th base station, N represents the total number of sets, Indicating the number of sets containing the i-th base station.
6. The crime scene recognition method based on trajectory data and base station positioning data according to claim 5, wherein the calculation of the activity-level value obtains the activity-level value of the base station by multiplying the activity value and the level value of each base station.
7. The method for identifying a crime scene based on trajectory data and base station positioning data according to claim 6, wherein said determining a crime scene comprises the steps of: For each base station, calculating the ratio of the activity-common degree value of the base station in the suspect track positioning data to the activity-common degree value of the common person mobile phone base station positioning data; according to the ratio, calculating the probability that each base station is the same as the suspicious person in positioning: Wherein, therein Indicating the probability that the ith base station is the same as the suspect location, A value representing the activity-common level of the ith base station in the suspect data, An activity-common level value indicating an i-th base station in the common person data; based on the source confidence level and 1 to 24 hours The dynamic distribution diagram sets a probability threshold, and when the probability that a certain base station is the same as the suspicious person in positioning exceeds the probability threshold, the area covered by the corresponding base station is determined as a possible crime place.
8. A computer device comprises a memory and a processor, wherein the memory stores a computer program, and the computer program is characterized in that the processor executes the computer program to realize the steps of the crime place identification method based on track data and base station positioning data according to any one of claims 1 to 7.
9. A computer-readable storage medium having stored thereon a computer program, wherein the computer program when executed by a processor implements the steps of the crime scene recognition method based on trajectory data and base station positioning data as set forth in any one of claims 1 to 7.

Description

Crime place identification method based on track data and base station positioning data Technical Field The invention relates to the technical field of electronic data evidence obtaining data mining, in particular to a crime place identification method based on track data and base station positioning data. Background In the crime investigation process, the crime location is critical for the investigation of cases. Conventional ways of identifying crime places often depend on means such as on-site investigation, witness and dialect, however, these methods have certain limitations. With the development of information technology, electronic data is increasingly widely applied to crime investigation. The trajectory positioning data of the suspects can provide the activity range and the position information about the suspects, but an effective method for accurately positioning the crime scene by fully utilizing the data is lacking at present. How to scientifically and efficiently analyze and compare the trajectory positioning data of the suspects with the positioning data of the massive base stations, so that the crime places are accurately determined, and the method becomes a problem to be solved in the field. The existing track data analysis technology generally determines whether a locating place is a crime place or not by analyzing locating data of a suspected person in a period of time, analyzing the locating place of the suspected person in the same locating place for a plurality of times, or analyzing the locating place by collision of locating data of a plurality of suspected persons. Such analytical methods suffer from the following disadvantages: First, in the daily activity track of a suspect, public areas such as a mall, a station, a hospital, etc., are inevitably involved, and these areas are places where the general population moves at high frequency. The conventional method only analyzes the trajectories of the suspects, and is easy to misjudge the public area as a crime place, because whether the suspects appear in the area is normal activity or crime-related behaviors cannot be distinguished. Secondly, the traditional method is to count the occurrence frequency of the suspects at each locating point, and take the high-frequency occurrence point as a suspected crime place. The method lacks of a scientific quantitative model support, the judgment result is greatly influenced by subjective experiences of investigation staff, and positioning points with similar frequencies but different properties are difficult to distinguish accurately. Furthermore, the spatio-temporal relevance of the positioning data is not considered. Criminals often have specific space-time characteristics, and the analysis of the suspicious positioning data by the traditional method is limited to a single time or space dimension, so that the inherent correlation of the positioning data of different time periods and different areas is not fully mined, and the crime places with concealment are difficult to identify. The traditional method only focuses on the positioning data of the individuals of the suspects, ignores the regional activity rule information contained in the positioning data of massive common people, and cannot exclude the normal activity region by comparison, so that the investigation range is narrowed. Disclosure of Invention This section is intended to summarize some aspects of embodiments of the application and to briefly introduce some preferred embodiments, which may be simplified or omitted in this section, as well as the description abstract and the title of the application, to avoid obscuring the objects of this section, description abstract and the title of the application, which is not intended to limit the scope of this application. The present invention has been made in view of the above-described problems occurring in the prior art. The technical scheme includes that S1, suspicious person track positioning data and common person mobile phone base station positioning data are obtained, source confidence and application labels are added to the obtained data, S2, abnormal cleaning and data standardization are conducted on the obtained data, a standardized track data set is built, S3, base stations are defined as units and time periods are defined as sets, active values of the base stations in different sets are calculated, S4, common degree values of the base stations are calculated based on coverage conditions of the sets, calculated active-common degree values are obtained, and crime places are determined. The method comprises the steps of carrying out anomaly cleaning on suspicious trajectory positioning data, re-estimating coordinates according to a reachable speed and an acceleration threshold, deleting the positioning coordinates if obvious anomalies exist, and removing repeated records and unreasonable connection time data of common mobile phone base station positioning data. The method comprises