Luggage check-in intrusion detection method and system, electronic equipment and storage medium

Abstract

The invention discloses a luggage check-in intrusion detection method and system, electronic equipment and a storage medium, and relates to the technical field of image processing. The method comprises the following steps: acquiring luggage state information and consignment video information of target luggage, wherein the luggage state information comprises luggage opening and closing data, pressure distribution data and temperature data; associating the consignment video information with the electronic identifier of the target luggage to obtain a space-time mapping relationship; performing anomaly detection on the luggage opening and closing data, the pressure distribution data and the temperature data to obtain an anomaly detection result, and determining a corresponding timestamp and a luggage position coordinate; extracting video data corresponding to the timestamps and the luggage position coordinates from the space-time mapping relation, and identifying suspicious persons in contact with the target luggage in the video data and operation behaviors of the suspicious persons; and when the operation behavior is determined as a non-preset authorization behavior, generating an intrusion event report. By implementing the technical scheme provided by the invention, the effect of improving the efficiency of luggage check-in intrusion detection is achieved.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and particularly to a method, system, electronic device and storage medium for detecting intrusion in checked luggage. Background Art

[0002] With the rapid development of the air transportation industry, the volume of checked luggage continues to grow, and the security issues during the checked luggage process have become increasingly prominent. The checked luggage process involves multiple scene conversions and personnel operations, and a perfect security management mechanism needs to be established to prevent security incidents such as illegal opening of luggage and theft of items during transportation.

[0003] Currently, the existing methods for detecting intrusion in checked luggage mainly adopt a combination of video surveillance and manual inspections. Staff judge whether there are abnormal situations by watching the surveillance footage and combining the inspection records. However, in actual applications, due to the large number of checked luggage, single abnormal situation detection by only video surveillance and manual inspections is prone to omissions, thus reducing the efficiency of detecting intrusion in checked luggage. Summary of the Invention

[0004] This application provides a method, system, electronic device and storage medium for detecting intrusion in checked luggage, which has the effect of improving the efficiency of detecting intrusion in checked luggage.

[0005] In a first aspect, this application provides a method for detecting intrusion in checked luggage, including: Obtaining the luggage status information and checked luggage video information of the target luggage, where the luggage status information includes luggage opening and closing data, pressure distribution data, and temperature data; Associating the checked luggage video information with the electronic identifier of the target luggage to obtain the spatio-temporal mapping relationship between the target luggage and the checked luggage video information; Performing anomaly detection on the luggage opening and closing data, pressure distribution data, and temperature data to obtain an anomaly detection result, and determining the time stamp and luggage position coordinates corresponding to the anomaly detection result; Extracting the video data corresponding to the time stamp and luggage position coordinates from the spatio-temporal mapping relationship, and identifying the suspicious persons who come into contact with the target luggage and the operation behaviors of the suspicious persons in the video data; When the operation behavior is confirmed as a non-preset authorized behavior, generating an intrusion event report for the suspicious person.

[0006] In a second aspect of this application, a system for detecting intrusion in checked luggage is provided. The system includes: An information acquisition module, configured to acquire the luggage status information and checked luggage video information of the target luggage, where the luggage status information includes luggage opening and closing data, pressure distribution data, and temperature data; A mapping relationship determination module, configured to associate the consignment video information with the electronic identifier of the target luggage to obtain a spatio-temporal mapping relationship between the target luggage and the consignment video information; An operation behavior recognition module, configured to perform anomaly detection on the luggage opening / closing data, pressure distribution data, and temperature data to obtain an anomaly detection result, and determine the timestamp and luggage position coordinates corresponding to the anomaly detection result; extract the video data corresponding to the timestamp and luggage position coordinates from the spatio-temporal mapping relationship, and identify the suspicious person who touches the target luggage and the operation behavior of the suspicious person in the video data; An intrusion detection module, configured to generate an intrusion event report of the suspicious person when the operation behavior is confirmed as a non-preset authorized behavior.

[0007] In a third aspect of the present application, an electronic device is provided, including a memory, a processor, and a program stored on the memory and executable on the processor. When the program is loaded and executed by the processor, it can implement a luggage consignment intrusion detection method.

[0008] In a fourth aspect of the present application, a computer-readable storage medium is provided. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements a luggage consignment intrusion detection method.

[0009] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: By adopting the above technical solution, a spatio-temporal mapping relationship of the target luggage is established based on the luggage status information, including luggage opening / closing data, pressure distribution data, and temperature data, and the consignment video information, realizing all-round monitoring of the luggage. Then, anomaly detection is performed on the luggage status information, and the timestamp and position coordinates of the anomaly detection result are determined. Combining the spatio-temporal mapping relationship, the suspicious person and their operation behavior in the corresponding video data are extracted, thereby realizing multi-dimensional luggage consignment intrusion detection. Finally, when it is found that the operation behavior does not belong to the preset authorized behavior, an intrusion event report of the suspicious person is automatically generated to timely discover and record the intrusion behavior. This solution constructs an automated and intelligent intrusion detection mechanism by combining the luggage status information with the consignment video information. Compared with the traditional method that relies solely on video monitoring and manual inspection, it can more comprehensively and timely discover anomalies during the luggage consignment process, significantly improving the efficiency of luggage consignment intrusion detection. Description of the Drawings

[0010] Figure 1 It is a flowchart of a luggage consignment intrusion detection method provided by an embodiment of the present application; Figure 2 It is a schematic structural diagram of a baggage consignment intrusion detection system provided by an embodiment of the present application; Figure 3 It is a schematic structural diagram of an electronic device provided by an embodiment of the present application.

[0011] Explanation of reference numerals: 300, electronic device; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory. Detailed implementation manners

[0012] In order to enable those skilled in the art to better understand the technical solutions in this specification, the following will clearly and completely describe the technical solutions in the embodiments of this specification with reference to the accompanying drawings in the embodiments of this specification. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments.

[0013] In the description of the embodiments of the present application, words such as "for example" or "for illustration" are used to represent examples, illustrations or explanations. Any embodiment or design solution described as "for example" or "for illustration" in the embodiments of the present application should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Exactly speaking, the use of words such as "for example" or "for illustration" is intended to present relevant concepts in a specific manner.

[0014] In the description of the embodiments of the present application, the meaning of the term "plurality" refers to two or more. For example, a plurality of systems refers to two or more systems, and a plurality of screen terminals refers to two or more screen terminals. In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the technical features indicated. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. The terms "include", "comprise", "have" and their variants all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0015] The embodiments of the present application provide a baggage consignment intrusion detection method. In one embodiment, please refer to Figure 1 , Figure 1 It is a flowchart of the baggage consignment intrusion detection method provided by the embodiments of the present application. This method can be implemented depending on a computer program, which can be integrated in an application or run as an independent tool-like application. This method can also be implemented depending on a single-chip microcomputer and can also run on a baggage consignment intrusion detection system based on the von Neumann architecture. Specifically, this method may include the following steps: Step 101: Obtain the luggage status information and consignment video information of the target luggage. The luggage status information includes luggage opening / closing data, pressure distribution data, and temperature data.

[0016] Among them, the luggage status information refers to a set of real-time data reflecting the physical state and external environment changes of the luggage during consignment, including luggage opening / closing data, pressure distribution data, and temperature data. In the embodiments of the present application, it can be understood that: the luggage opening / closing data is the information on the change of the box opening / closing state collected by a Hall sensor, used to monitor whether the luggage has been illegally opened; the pressure distribution data is the force condition on the surface of the box collected by an array of pressure sensors, used to detect whether the luggage has suffered abnormal external forces; the temperature data is the information on the temperature change around the box collected by a temperature sensor network, used to determine whether there is an abnormality in the environment where the luggage is located.

[0017] The consignment video information refers to the visual data stream collected by high-definition video monitoring devices deployed in the consignment scenario, recording visual information such as the movement trajectory of the luggage, the changes in the surrounding environment, and the operations of personnel during consignment. In the embodiments of the present application, it can be understood that: the consignment video information includes real-time video images of the luggage in different scenarios such as the consignment conveyor belt, sorting area, and temporary storage area. The video images include data such as the luggage position coordinates, time series information, and personnel activities, used for visual-level monitoring of the luggage consignment process.

[0018] Specifically, an opening / closing sensor, an array of pressure sensors, and a temperature sensor network are set on the luggage box to collect the luggage status information. Among them, the opening / closing sensor uses a Hall sensor and is installed at the opening / closing structure of the luggage box. By detecting the change in the magnetic field, it records the change in the opening / closing state of the luggage in real time, and converts the collected opening / closing signal into a digital quantity to generate luggage opening / closing data; the array of pressure sensors is composed of multiple pressure sensing units, which are distributed in a matrix on the surface of the luggage box to collect the pressure distribution on the surface of the luggage in real time, and convert the analog signal into a digital signal through a signal conditioning circuit to form pressure distribution data; the temperature sensor network includes multiple temperature sensing probes, which are arranged at different positions on the luggage box to continuously monitor the temperature change in the surrounding environment of the luggage, and generate temperature data after analog-to-digital conversion. At the same time, high-definition video monitoring devices are deployed in the consignment scenario to provide full coverage of the consignment area without dead angles and collect real-time video information during consignment. The luggage status information collected by various sensors and the consignment video information recorded by the video monitoring devices are uploaded to the monitoring system in real time through a wireless communication module to form a complete data collection system. This multi-dimensional data collection method can comprehensively reflect the physical state changes and surrounding environment conditions of the luggage during consignment, providing rich data support for subsequent abnormal behavior detection and effectively improving the accuracy and reliability of intrusion detection.

[0019] Step 102: Associate the consignment video information with the electronic identification of the target luggage to obtain the spatio-temporal mapping relationship between the target luggage and the consignment video information.

[0020] The electronic identification refers to the identity identification information stored in the electronic tag installed on the luggage body, which is used to achieve the unique identification and trajectory tracking of the luggage.

[0021] The spatio-temporal mapping relationship refers to the corresponding relationship established between the physical position, time information of the target luggage during the consignment process and the corresponding video surveillance data. In the embodiments of the present application, it can be understood that: the spatio-temporal mapping relationship is a mapping table of the luggage trajectory coordinate data and the time series data established based on the luggage ID. This mapping table records the spatial position information of the luggage at different time points and its corresponding video segments, which is used to establish the association between the physical trajectory of the luggage and the video surveillance data.

[0022] Specifically, first extract the luggage trajectory coordinate data and time series data from the consignment video information. The luggage trajectory coordinate data records the spatial position changes of the luggage during the consignment process, and the time series data records the corresponding time information. At the same time, read the electronic identification information stored in the electronic tag installed on the target luggage. This electronic identification information includes a unique luggage ID and timestamp information. Establish a mapping table of the luggage trajectory coordinate data and the time series data based on the luggage ID. This mapping table records the corresponding relationship between the luggage ID and its corresponding spatial position and time information. Then, retrieve the corresponding consignment video segment in the mapping table according to the timestamp in the electronic identification information, and use the retrieved video segment as the spatio-temporal mapping relationship between the target luggage and the consignment video information. The establishment of this mapping relationship realizes the precise correspondence between the physical space position of the luggage and the video surveillance data, enabling the system to quickly locate the corresponding video image according to the abnormal time point in the luggage status information, improving the efficiency and accuracy of abnormal behavior traceability. When detecting an abnormal luggage status, the corresponding video data at the corresponding time point can be immediately found through the spatio-temporal mapping relationship, and suspicious personnel and abnormal operation behaviors can be detected in a timely manner.

[0023] Based on the above embodiments, as an alternative embodiment, in step 102: Associating the consignment video information with the electronic identification of the target luggage to obtain the spatio-temporal mapping relationship between the target luggage and the consignment video information, this step may further include the following steps: Step 201: Obtain the luggage trajectory coordinate data and time series data in the consignment video information; read the electronic identification information of the target luggage, and the electronic identification information includes the luggage ID and timestamp.

[0024] Specifically, it is first necessary to obtain the luggage trajectory coordinate data and time series data in the consignment video information. Specifically, through image processing and object detection of the video stream collected by the high-definition video monitoring equipment in the consignment scenario, the position coordinates of the luggage in the video frame are extracted, and the luggage trajectory coordinate data is generated in combination with the video timestamp; at the same time, the video acquisition time corresponding to each coordinate point is recorded to generate the time series data. In addition, the system reads the electronic identification information stored in the electronic tag installed on the target luggage through the RFID reader to obtain the luggage ID and timestamp information. The acquisition of these data provides basic data support for establishing the spatio-temporal mapping relationship in the follow-up.

[0025] Step 202: Based on the luggage ID, establish a mapping table between the luggage trajectory coordinate data and the time series data.

[0026] Specifically, the system establishes a mapping table between the luggage trajectory coordinate data and the time series data based on the luggage ID. Specifically, the luggage ID is used as the primary key to establish a data table structure including trajectory coordinates, time information, and video segment indexes. The system records the obtained luggage trajectory coordinate data into the mapping table in chronological order, and each record includes the coordinate position (x, y, z), the corresponding time point, and the video segment index corresponding to this time point. The establishment of this mapping table realizes the association between the physical trajectory of the luggage and the time dimension, providing a data index for subsequent video positioning.

[0027] Step 203: Determine the corresponding consignment video segment in the mapping table according to the timestamp, and use it as the spatio-temporal mapping relationship between the target luggage and the consignment video information.

[0028] Specifically, the system determines the corresponding consignment video segment in the mapping table according to the timestamp in the electronic identification information. Specifically, the system first reads the timestamp information in the electronic identification, and then retrieves the record corresponding to this timestamp in the established mapping table. By comparing the timestamp with the time series data in the mapping table, the system can locate the spatial position of the luggage at a specific time point, and obtain the corresponding video monitoring image according to this position information and the video segment index. The corresponding relationship between these video segments and the luggage trajectory constitutes the spatio-temporal mapping relationship between the target luggage and the consignment video information. By establishing this mapping relationship, the system can quickly locate the video image at the moment when the luggage status is abnormal when detecting the abnormal status of the luggage, realizing the precise association between the abnormal status of the luggage and the video evidence, and improving the efficiency of tracing abnormal behaviors.

[0029] Step 103: Perform anomaly detection on the luggage opening and closing data, pressure distribution data, and temperature data to obtain the anomaly detection result, and determine the timestamp and luggage position coordinates corresponding to the anomaly detection result.

[0030] Among them, the abnormal detection result refers to the luggage physical state abnormal judgment data obtained by the system after analyzing the luggage status information, which reflects whether there are abnormal conditions such as abnormal opening and closing, abnormal pressure or abnormal temperature during the luggage consignment process.

[0031] The timestamp and luggage position coordinates refer to the data combination that records the time information and spatial position information when the abnormal event occurs.

[0032] Specifically, first, based on the luggage opening and closing data collected by the Hall sensor, abnormal judgment is performed by setting the opening and closing state threshold. When it is detected that the opening and closing angle of the box body exceeds the preset threshold, the system determines that the opening and closing is abnormal; at the same time, using the pressure distribution data collected by the pressure sensor array, the pressure distribution characteristics are analyzed through a deep learning model. When it is detected that the pressure distribution pattern deviates from the normal range or an abnormal pressure peak appears, the system determines that the pressure is abnormal; in addition, based on the temperature data collected by the temperature sensor network, by establishing a temperature change model, when it is detected that the temperature suddenly changes or significantly deviates from the expected change trend, the system determines that the temperature is abnormal. When any of the above abnormalities is detected, the system will record the specific timestamp when the abnormality occurs, and combine with the luggage real-time position tracking module to obtain the spatial coordinate position of the luggage when the abnormality occurs. This multi-dimensional abnormal detection method can not only comprehensively monitor the physical state changes of the luggage, but also provide accurate time and position positioning for subsequent video tracing by recording the spatio-temporal information of the abnormal event, effectively improving the accuracy and efficiency of tracing abnormal behaviors. For example, when the system detects that the luggage box is illegally opened, it can immediately locate the exact time and location of the abnormality, quickly retrieve the corresponding video evidence in combination with the spatio-temporal mapping relationship, and timely discover and stop illegal behaviors.

[0033] Based on the above embodiments, as an optional embodiment, in step 103: performing abnormal detection on the luggage opening and closing data, pressure distribution data, and temperature data to obtain the abnormal detection result. This step may further include the following steps: Step 301: Determine the opening and closing abnormal level according to the opening and closing frequency and duration in the luggage opening and closing data.

[0034] Specifically, the system determines the level of abnormal opening and closing based on the frequency and duration of opening and closing in the luggage opening and closing data. Specifically, the Hall sensor installed on the luggage body is used to monitor the change of the opening and closing state of the body in real time. For example, the opening and closing state data is collected every 100 ms. The system counts the number of times the luggage body opens and closes within a time window (such as 60 seconds) as the opening and closing frequency, and at the same time records the duration of each opening and closing state. Based on historical data analysis, the system sets a normal opening and closing frequency threshold (such as 3 times per minute) and a normal opening and closing duration threshold (such as 15 seconds). When it is detected that the opening and closing frequency exceeds the threshold or the duration of a single opening and closing is abnormal, the system calculates the level of abnormal opening and closing according to the degree of exceeding the threshold: for example, when the opening and closing frequency is between 3 - 5 times per minute or the duration of a single opening and closing is between 15 - 30 seconds, it is determined as a mild abnormality, and the abnormal level is 1; when the opening and closing frequency is between 5 - 8 times per minute or the duration of a single opening and closing is between 30 - 60 seconds, it is determined as a moderate abnormality, and the abnormal level is 2; when the opening and closing frequency exceeds 8 times per minute or the duration of a single opening and closing exceeds 60 seconds, it is determined as a severe abnormality, and the abnormal level is 3. This precise grading method can effectively identify possible illegal opening behaviors.

[0035] Step 302: Determine the level of pressure abnormality according to the displacement of the pressure center and the pressure peak value in the pressure distribution data.

[0036] Specifically, the system determines the level of pressure abnormality according to the displacement of the pressure center and the pressure peak value in the pressure distribution data. Specifically, a pressure sensor array (such as a 16×16 sensor matrix) is arranged on the surface of the luggage, and the sampling frequency is, for example, 200 Hz, to collect the pressure distribution data on the surface of the luggage body in real time. The system obtains the pressure center coordinates by calculating the centroid of the pressure distribution, and calculates the displacement distance through the pressure center coordinates at consecutive time points. At the same time, the system monitors the maximum pressure value in the pressure distribution as the pressure peak value. Based on the pressure characteristics of normal consignment operations, the system sets a pressure center displacement threshold (such as 50 mm) and a pressure peak value threshold (such as 500 N). When the pressure characteristics are abnormal, the system calculates the level of pressure abnormality according to the following criteria: for example, when the pressure center displacement is between 50 - 100 mm or the pressure peak value is between 500 - 800 N, it is determined as a mild abnormality, and the abnormal level is 1; when the pressure center displacement is between 100 - 200 mm or the pressure peak value is between 800 - 1200 N, it is determined as a moderate abnormality, and the abnormal level is 2; when the pressure center displacement exceeds 200 mm or the pressure peak value exceeds 1200 N, it is determined as a severe abnormality, and the abnormal level is 3. This method of abnormal grading based on quantitative analysis can accurately identify abnormal operations such as extrusion and impact.

[0037] Step 303: Obtain the level of temperature abnormality based on the mutation point and temperature gradient of the temperature data.

[0038] Specifically, the system obtains the temperature anomaly level based on the mutation points and temperature gradients of the temperature data. Specifically, temperature sensors are arranged at key positions of the luggage (such as the four corner points and the central position), and the sampling interval is, for example, 1 second, forming a temperature sensor network. The system uses the sliding window method to detect temperature mutation points. The window size is, for example, 30 seconds, and the mutation situation is identified by calculating the temperature change rate within the window. At the same time, the spatial temperature gradient is calculated using the temperature difference between adjacent sensors. Based on the temperature characteristics of the normal checked environment, the system sets a temperature mutation threshold (such as 5 °C / minute) and a temperature gradient threshold (such as 2 °C / cm). When the temperature characteristics are abnormal, the system calculates the temperature anomaly level according to the following criteria: for example, when the temperature mutation rate is between 5 - 8 °C / minute or the temperature gradient is between 2 - 4 °C / cm, it is determined as a mild anomaly, and the anomaly level is 1; when the temperature mutation rate is between 8 - 12 °C / minute or the temperature gradient is between 4 - 6 °C / cm, it is determined as a moderate anomaly, and the anomaly level is 2; when the temperature mutation rate exceeds 12 °C / minute or the temperature gradient exceeds 6 °C / cm, it is determined as a severe anomaly, and the anomaly level is 3. This fine-grained temperature anomaly classification helps to detect abnormal changes in the environment where the luggage is located in a timely manner.

[0039] Step 304: When any one of the opening / closing anomaly level, pressure anomaly level, and temperature anomaly level exceeds the anomaly level threshold, record the corresponding data time point and anomaly level as the anomaly detection result.

[0040] Specifically, the system sets an anomaly level threshold (for example, 2), that is, an alarm is triggered when the anomaly level of any monitoring dimension reaches or exceeds the moderate anomaly. When an anomaly is detected, the system will record the precise time point (accurate to the millisecond level) when the anomaly occurs, the anomaly type, and the anomaly level value, and call the luggage real-time position tracking module to obtain the three-dimensional space coordinates at that time. These pieces of information together constitute the anomaly detection result and are stored in the anomaly event database of the system. For example, when the system detects that the pressure anomaly level of the luggage at a certain moment is 2, it will record the time stamp at that moment, the type identifier of "pressure anomaly", the anomaly level value 2, and the precise position coordinates of the luggage at that time. This multi-dimensional anomaly detection and detailed information recording provide a complete event basis for subsequent video traceability and anomaly handling, significantly improving the accuracy and traceability of checked luggage safety monitoring.

[0041] Based on the above embodiments, as an alternative embodiment, in step 103: determining the time stamp and luggage position coordinates corresponding to the anomaly detection result, this step may further include the following steps: Step 305: Obtain the anomaly event trigger time point in the anomaly detection result; mark the data within a preset time period before and after the anomaly event trigger time point as the attention time window.

[0042] Specifically, by obtaining the trigger time point of the abnormal event in the anomaly detection result and performing marker analysis on the data before and after this time point. First, read the time point when the abnormal event was initially triggered as recorded in the anomaly detection result. Then, centered on this time point, expand a preset time length (for example, 30 seconds before and after) to mark all sensor data within this time range as the time window of interest. This data marking method based on the time window can completely capture the development process of the abnormal event, including the omen features before the anomaly occurs and the evolution features after the anomaly occurs, providing sufficient data support for accurately positioning the critical moment of the abnormal event.

[0043] Step 306: Extract the moment with the highest degree of abnormality from the time window of interest and use it as the time stamp corresponding to the anomaly detection result.

[0044] Specifically, the system deeply analyzes the data within the time window of interest to extract the moment with the highest degree of abnormality. First, uniformly quantify the opening / closing abnormality level, pressure abnormality level, and temperature abnormality level within the time window. For example, standardize each abnormality level to the range of 0 - 1. Then, the system uses a sliding time window (for example, the window size is 5 seconds) to calculate the comprehensive degree of abnormality point by point within the time window of interest. The calculation of the comprehensive degree of abnormality takes into account the weighted combination of various abnormality levels. By comparing the comprehensive degrees of abnormality at different moments, the system finds the moment point with the highest degree of abnormality and uses this moment as the time stamp corresponding to the anomaly detection result. This method for determining the time stamp based on the comprehensive degree of abnormality can accurately locate the most severe moment of the abnormal event, providing an accurate time positioning for subsequent video tracing.

[0045] Based on the above embodiments, as an alternative embodiment, in step 306: The step of extracting the moment with the highest degree of abnormality from the time window of interest may further include the following steps: Step 316: Calculate the weighted average of the opening / closing abnormality level, pressure abnormality level, and temperature abnormality level within the time window of interest and generate a comprehensive abnormality level curve corresponding to each weighted average.

[0046] Specifically, the system comprehensively processes the multi-dimensional anomaly level data within the attention time window to generate a time series curve reflecting the overall anomaly status. First, weight coefficients are assigned to the opening / closing anomaly level, pressure anomaly level, and temperature anomaly level respectively (for example, the weight of opening / closing anomaly is 0.4, the weight of pressure anomaly is 0.4, and the weight of temperature anomaly is 0.2). The setting of these weight coefficients is based on the impact degree of various anomalies on luggage safety. Then, the system uses the moving average method (for example, the window size is 3 seconds) within the attention time window to calculate the weighted average of each anomaly level. By performing weighted summation on the anomaly level values at each time point, a unified comprehensive anomaly level curve is generated. This data fusion method based on weighted average can smooth the influence of instantaneous fluctuations while retaining the main features of anomaly events, providing a reliable data basis for subsequent peak analysis.

[0047] Step 326: Perform peak detection on the comprehensive anomaly level curve to obtain a set of anomaly peak points.

[0048] Specifically, the system conducts peak detection and analysis on the generated comprehensive anomaly level curve. Specifically, a local maximum detection algorithm is adopted, and a detection window (for example, 5 seconds) and a peak threshold (for example, anomaly level 1.5) are set to search for time points on the comprehensive anomaly level curve that meet the peak conditions. When the anomaly level value at a certain moment is greater than the values at its previous and subsequent time points and exceeds the set peak threshold, that moment is marked as a peak point. The system stores all the detected peak points in the set of anomaly peak points in chronological order, and each peak point record contains a timestamp and the corresponding anomaly level value. This method based on peak detection can effectively identify the critical time points in anomaly events, providing candidate moments for determining the final anomaly timestamp.

[0049] Step 336: Calculate the peak amplitude and duration of each peak point in the set of anomaly peak points; take the moment corresponding to the peak point with the largest product of the peak amplitude and duration as the moment with the highest anomaly degree.

[0050] Specifically, the system determines the moment with the highest degree of abnormality by analyzing the characteristics of abnormal peak points. First, the characteristic parameters of each peak point are calculated: the peak amplitude is the difference between the abnormal level value of this point and the baseline value (such as abnormal level 1.0), and the duration is the length of the time period during which the abnormal level value remains above 80% of the peak (for example, in seconds). Then, the system calculates the product of the peak amplitude and the duration of each peak point, and this product reflects the combined effect of the severity of the abnormal event and the persistence of the impact. For example, if the peak amplitude of a certain peak point is 2.5 and the duration is 8 seconds, then the product is 20; for another peak point, the peak amplitude is 2.0 and the duration is 12 seconds, then the product is 24. The system selects the moment corresponding to the peak point with the largest product value as the moment with the highest degree of abnormality. This method of determining the moment based on comprehensive multi-feature evaluation not only considers the severity of the abnormality but also the continuous impact of the abnormality, and can more accurately locate the critical moment that poses the greatest threat to luggage safety, providing an accurate time reference point for subsequent video tracing and processing.

[0051] Step 307: Obtain the position tracking data corresponding to the time stamp, and extract the three-dimensional coordinates in the position tracking data as the luggage position coordinates corresponding to the abnormal detection result.

[0052] Specifically, the system obtains the corresponding luggage position information according to the determined time stamp. First, it accesses the database of the luggage real-time position tracking module and retrieves the position tracking data corresponding to the time stamp. The position tracking data contains the complete movement trajectory of the luggage during consignment and records the spatial position information of the luggage at different moments. The system extracts the three-dimensional coordinates (x, y, z) corresponding to this time stamp from the position tracking data, where the x and y coordinates represent the position of the luggage in the consignment scene plane, and the z coordinate represents the height information of the luggage. By using these spatial coordinate information as the luggage position coordinates corresponding to the abnormal detection result, the system realizes the accurate spatial positioning of the abnormal event. This spatio-temporal combined positioning method can not only determine the specific location where the abnormal event occurs but also provide an accurate spatial retrieval basis for subsequent video surveillance tracing, significantly improving the efficiency and accuracy of abnormal event investigation. For example, when the system detects a severe pressure abnormality in a suitcase, it can quickly locate the specific location where the abnormality occurs (such as a conveyor belt transfer point in the sorting area) and combine the video surveillance data at this location to promptly discover and handle the abnormal situation.

[0053] Step 104: Extract the video data corresponding to the time stamp and the luggage position coordinates from the spatio-temporal mapping relationship, and identify the suspicious persons who come into contact with the target luggage and the operating behaviors of the suspicious persons.

[0054] Among them, video data refers to the continuous image sequence information collected by surveillance cameras in the consignment scenario. The video data is used to record the real-time status of the luggage and its surrounding environment during the consignment process. Through the spatio-temporal mapping with the abnormal detection results, it realizes the visual traceability and restoration of abnormal events, providing intuitive visual evidence for confirming the identity of suspicious persons and judging the nature of operation behaviors.

[0055] Suspicious persons and their operation behaviors refer to the persons who have close contact with the target luggage where an abnormality occurs during the consignment process and the specific actions they perform on the luggage. The identification results of suspicious persons and their operation behaviors are used to determine the responsible entity and the nature of the behavior of the abnormal event, helping security management personnel quickly locate and handle the abnormal event, and at the same time providing a basis for subsequent event accountability and safety management optimization.

[0056] Specifically, based on the timestamps and luggage position coordinates in the abnormal detection results, the system retrieves the corresponding video data from the spatio-temporal mapping relationship database and performs intelligent analysis on the video data to identify suspicious persons and their behaviors. First, the monitoring area is determined according to the luggage position coordinates (for example, a circular area with a radius of 5 meters centered on the luggage coordinates), and all surveillance cameras covering this area are screened out. Then, the system extracts the video data of these cameras within a preset time period before and after the timestamp (for example, 2 minutes before and after). For the obtained video data, the system uses a deep learning object detection algorithm for analysis and processing: first, the luggage detection model is used to locate the position and contour of the target luggage, and at the same time, the person detection model is used to identify the person targets in the video; then, the target tracking algorithm is used to record the movement trajectories of the person targets, calculate the spatial distance and contact duration between the person targets and the target luggage; when it is detected that the distance between the person target and the target luggage is less than the threshold (for example, 1 meter) and the contact duration exceeds the set value (for example, 3 seconds), the system marks this person as a suspicious person. For the identified suspicious persons, the system further uses a behavior recognition algorithm to analyze their operation behaviors and classifies the behaviors into normal operations (such as routine inspections, conveyor belt transfers, etc.) and abnormal operations (such as unauthorized opening of boxes, violent handling, etc.). This traceability method based on video intelligent analysis can quickly locate the relevant persons and specific operation processes of the abnormal event, provide intuitive evidence for the restoration of the abnormal event for security management personnel, and significantly improve the processing efficiency and accuracy of the abnormal event. For example, when the system detects abnormal opening and closing of luggage in a certain sorting area, through video analysis, it can quickly confirm whether there is unauthorized box-opening inspection behavior and record the identity characteristics and specific operation processes of the relevant persons, providing strong support for subsequent security management.

[0057] Based on the above embodiments, as an alternative embodiment, in step 104: Extracting the video data corresponding to the timestamps and luggage position coordinates from the spatio-temporal mapping relationship, this step may further include the following steps: Step 401: Determine the corresponding video time point in the time-space mapping relationship according to the timestamp.

[0058] Specifically, the system searches for the corresponding video time point in the time-space mapping relationship database based on the timestamp in the anomaly detection result. First, the timestamp of the anomaly detection result is converted to a standard time format, and then time alignment is performed through the time mapping function pre-established in the time-space mapping relationship. Since the time base of the surveillance video may have a slight deviation from the sensor system, the system will calibrate the timestamp (for example, compensate based on the system clock synchronization error) to ensure that the video frame at the time when the abnormal event occurred is accurately located. This precise time alignment method can effectively avoid video traceability deviations caused by time errors and provide an accurate time starting point for subsequent video analysis.

[0059] Step 402: Taking the video time point as the center, capture a video segment of a preset length; and determine a target detection area in the video segment based on the luggage location coordinates.

[0060] Specifically, the system captures video clips based on a certain video time point and locates the target detection area in the video clip. Specifically, with the video time point as the center, video data of a preset length (for example, 90 seconds before and after) is captured to form a complete video clip. Then, the system converts the luggage position coordinates from the actual space coordinate system to the video image coordinate system, and calculates the projection position of the target luggage in the video screen taking into account the camera's installation position, shooting angle, and field of view. Based on the projection position, the system defines the target detection area (for example, a square area with a side length of 40% of the height of the video screen and a projection position as the center), which contains complete visual information of the target luggage and its surrounding environment. This area positioning method based on spatial mapping can accurately capture the video content of areas related to abnormal events while reducing interference from irrelevant areas.

[0061] Step 403: Perform image enhancement processing on the target detection area to obtain video data.

[0062] Specifically, the system performs image enhancement processing on the determined target detection area to improve the quality of video data. First, image preprocessing is performed on the target detection area, including operations such as denoising (e.g., using a Gaussian filter to reduce image noise), contrast enhancement (e.g., enhancing image details through histogram equalization), and sharpening (e.g., using the Laplacian operator to enhance edge features). Then, the system performs adaptive brightness adjustment according to the scene lighting conditions, performs light compensation on the underlit areas, and performs dynamic range compression on the overexposed areas. For motion-blurred images, the system uses a motion compensation algorithm to restore sharpness. Through these image enhancement processes, the system outputs high-quality video data, improving the accuracy of subsequent target detection and behavior recognition. For example, in a dimly lit scene, image enhancement can clearly display the detailed features on the surface of the suitcase and the operation actions of the personnel, providing a more reliable visual basis for the judgment of abnormal behaviors.

[0063] Based on the above embodiments, as an alternative embodiment, in step 104: identifying the suspicious personnel who come into contact with the target luggage and the operation behaviors of the suspicious personnel, this step may further include the following steps: Step 404: Perform personnel trajectory tracking on the video data to obtain personnel trajectory data.

[0064] Specifically, the system performs personnel trajectory tracking and analysis on the enhanced video data. First, a deep learning object detection algorithm (e.g., YOLOv5) is used to detect personnel targets in each frame of the video data, and the position, size, and pose features of the personnel are extracted. Then, the system uses a multi-object tracking algorithm (e.g., DeepSORT) to associate the same personnel targets in consecutive frames and assigns a unique ID identifier to each detected personnel. The system predicts and updates the motion state of the personnel targets through the Kalman filter algorithm, maintaining a stable tracking effect even when the personnel are temporarily occluded. Finally, the system generates personnel trajectory data including timestamps, position coordinates, and motion states, providing basic data support for subsequent contact analysis. This deep learning-based personnel tracking method can accurately capture the motion trajectories of all personnel in the scene, providing a reliable data basis for identifying suspicious personnel.

[0065] Step 405: Calculate the spatial distance between the personnel trajectory data and the position of the target luggage; when the spatial distance is less than the preset contact distance, mark the corresponding personnel as suspicious personnel.

[0066] Specifically, the system analyzes the contact situation between the personnel and the target luggage based on the personnel trajectory data. First, the personnel trajectory data and the target luggage position are converted into a unified coordinate system, and then the shortest spatial distance between the personnel and the luggage is calculated for each time point. The system sets a preset contact distance threshold (e.g., 1 meter). When the spatial distance between a certain personnel and the luggage continuously remains less than this threshold for more than a preset duration (e.g., 3 seconds), the system marks this personnel as a suspicious person. To improve the accuracy of judgment, the system also considers the orientation and posture information of the personnel. For example, it determines whether the personnel is facing the luggage through bone key point detection, and analyzes whether the personnel has an action of reaching out to touch the luggage through posture estimation. This multi-dimensional contact judgment method can accurately identify the suspicious personnel related to abnormal events, and effectively filter out the irrelevant personnel who are just passing by.

[0067] Step 406: Extract the frame image sequence of the suspicious person during the contact process, and match the frame image sequence based on a preset behavior recognition template to obtain the operation behavior of the suspicious person.

[0068] Specifically, the system conducts an analysis of the operation behavior of the identified suspicious person. First, a continuous frame image sequence of the suspicious person during the contact with the luggage is extracted from the video data, and spatio-temporal attention processing is performed on these images to highlight the key features of the personnel's operation behavior. Then, the system uses a pre-trained behavior recognition model to match the extracted frame image sequence with the preset behavior recognition templates. These behavior recognition templates contain various typical operation behavior features (such as the unpacking action template, the carrying action template, the inspection action template, etc.), and each template defines the spatio-temporal feature sequence of the corresponding behavior. The system identifies the specific operation behavior implemented by the suspicious person by calculating the similarity between the frame image sequence and each template. At the same time, the system also verifies the rationality of the recognition result in combination with the scene context information (such as the functional area where the personnel is located, the current operation process, etc.). This behavior recognition method based on template matching can accurately analyze the nature of the operation behavior of the suspicious person, providing a basis at the behavior level for the judgment of abnormal events. For example, the system can distinguish between normal security inspection unpacking operations and suspicious unauthorized unpacking behaviors, thus helping security management personnel quickly make correct disposal decisions.

[0069] Step 105: When the operation behavior is confirmed as a non-preset authorized behavior, generate an intrusion event report for the suspicious person.

[0070] Among them, the preset authorized actions refer to the set of standardized operation actions that personnel in different positions are allowed to perform in the consignment scenario. In the embodiments of the present application, it can be understood that: the preset authorized actions are a multi-dimensional behavior definition system including the operation subject, operation scenario, operation type, and operation specifications. Among them, the operation subject refers to the staff with specific identity permissions (such as security inspection personnel, transportation personnel, maintenance personnel, etc.), the operation scenario refers to specific functional areas (such as security inspection area, sorting area, transfer area, etc.), the operation type refers to the specified behavior actions (such as unpacking inspection, loading and unloading, maintenance, etc.), and the operation specifications refer to the specific requirements to be followed when performing each operation (such as the unpacking inspection time not exceeding 5 minutes, keeping horizontal and stable during handling, etc.). The preset authorized actions are used to establish the behavior criterion standards during the consignment process. By comparing with the actually detected operation behaviors, the compliance of the operation behaviors is judged, so as to identify and prevent the illegal operations of unauthorized personnel and ensure the safety of the consignment process.

[0071] The intrusion event report refers to the standardized event record document automatically generated by the system for the unauthorized operation behaviors detected during the consignment process.

[0072] Specifically, the system judges whether an intrusion event report needs to be generated by comparing the identified operation behaviors with the preset authorized action list. First, it calls the preset authorized action database, which stores the authorized operation types of personnel in different positions in different functional areas (for example, security inspection personnel can perform unpacking inspections in the security inspection area, and transportation personnel can perform standardized handling in the conveyor belt area). The system matches and analyzes the operation behaviors of the suspicious personnel with the authorized behaviors in the corresponding scenarios. When it is found that the operation behaviors are not within the authorized scope or the operation methods do not meet the specification requirements, the system automatically generates an intrusion event report. This report contains detailed event information: the time and location of the abnormality (based on the timestamp and the luggage position coordinates), the description of the characteristics of the suspicious personnel (such as physical features, dressing characteristics), the specific process of the unauthorized operation (based on the behavior recognition results), the associated abnormal detection data (such as sensor data of abnormal opening and closing, abnormal pressure, etc.), and the key video screenshots of the event site. At the same time, the system will classify the risk level of the report according to the severity of the event (for example, based on the danger level and duration of the operation behavior), and push alarm information to relevant management personnel according to the preset disposal process. This intelligent intrusion event report generation mechanism can timely discover and record the unauthorized operation behaviors during the consignment process, provide complete event information and disposal basis for security management personnel, and effectively improve the timeliness and accuracy of consignment security management. For example, when the system finds that a non-security inspection personnel performs unpacking operations on luggage in a non-security inspection area, it can immediately generate a detailed intrusion event report to help security personnel quickly intervene and dispose of it to prevent safety accidents from occurring.

[0073] Refer to Figure 2, which is a baggage check-in intrusion detection system provided by an embodiment of the present application. The system includes: an information acquisition module, a mapping relationship determination module, an operation behavior recognition module, and an intrusion detection module, where: The information acquisition module is used to acquire the baggage status information and checked baggage video information of the target baggage. The baggage status information includes baggage opening and closing data, pressure distribution data, and temperature data; The mapping relationship determination module is used to associate the checked baggage video information with the electronic identification of the target baggage to obtain the spatio-temporal mapping relationship between the target baggage and the checked baggage video information; The operation behavior recognition module is used to perform anomaly detection on the baggage opening and closing data, pressure distribution data, and temperature data to obtain an anomaly detection result, and determine the timestamp and baggage position coordinates corresponding to the anomaly detection result; extract the video data corresponding to the timestamp and baggage position coordinates from the spatio-temporal mapping relationship, and identify the suspicious personnel who come into contact with the target baggage and the operation behavior of the suspicious personnel; The intrusion detection module is used to generate an intrusion event report of the suspicious personnel when the operation behavior is confirmed as a non-preset authorized behavior.

[0074] Based on the above embodiment, the mapping relationship determination module is further used to acquire the baggage trajectory coordinate data and time series data in the checked baggage video information; read the electronic identification information of the target baggage, and the electronic identification information includes the baggage ID and timestamp; establish a mapping table between the baggage trajectory coordinate data and the time series data based on the baggage ID; determine the corresponding checked baggage video segment in the mapping table according to the timestamp, and use it as the spatio-temporal mapping relationship between the target baggage and the checked baggage video information.

[0075] Based on the above embodiment, the operation behavior recognition module is further used to determine the opening and closing anomaly level according to the opening and closing frequency and duration in the baggage opening and closing data; determine the pressure anomaly level according to the pressure center displacement and pressure peak value in the pressure distribution data; obtain the temperature anomaly level according to the mutation point and temperature gradient of the temperature data; when any one of the opening and closing anomaly level, pressure anomaly level, and temperature anomaly level exceeds the anomaly level threshold, record the corresponding data time point and anomaly level as the anomaly detection result.

[0076] Based on the above embodiment, the operation behavior recognition module is further used to obtain the anomaly event trigger time point in the anomaly detection result; mark the data within a preset time period before and after the anomaly event trigger time point as the attention time window; extract the moment with the highest anomaly degree from the attention time window and use it as the timestamp corresponding to the anomaly detection result; obtain the position tracking data corresponding to the timestamp, and extract the three-dimensional coordinates in the position tracking data as the baggage position coordinates corresponding to the anomaly detection result.

[0077] On the basis of the above embodiments, the operation behavior recognition module is also used to calculate the weighted average of the opening and closing abnormality level, the pressure abnormality level and the temperature abnormality level within the focus time window, and generate a comprehensive abnormality level curve corresponding to each weighted average value; perform peak detection on the comprehensive abnormality level curve to obtain a set of abnormal peak points; calculate the peak amplitude and duration of each peak point in the set of abnormal peak points; and take the moment corresponding to the peak point with the largest product of the peak amplitude and the duration as the moment with the highest degree of abnormality.

[0078] On the basis of the above-mentioned embodiment, the operation behavior recognition module is further used to determine the corresponding video time point in the time-space mapping relationship according to the timestamp; capture a video segment of a preset length with the video time point as the center; determine the target detection area in the video segment based on the luggage location coordinates; and perform image enhancement processing on the target detection area to obtain video data.

[0079] On the basis of the above embodiment, the operation behavior recognition module is also used to track the trajectory of the person in the video data to obtain the trajectory data of the person; calculate the spatial distance between the trajectory data of the person and the target luggage position; when the spatial distance is less than the preset contact distance, mark the corresponding person as a suspicious person; extract the frame image sequence of the suspicious person during the contact process, and match the frame image sequence based on the preset behavior recognition template to obtain the operation behavior of the suspicious person.

[0080] It should be noted that: when the device provided in the above embodiment realizes its function, only the division of the above functional modules is used as an example. In actual application, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0081] The present application also discloses an electronic device. Figure 3 , Figure 3 The electronic device 300 may include: at least one processor 301 , at least one network interface 304 , a user interface 303 , a memory 305 , and at least one communication bus 302 .

[0082] The communication bus 302 is used to realize the connection and communication between these components.

[0083] The user interface 303 may include a display interface and a camera interface. Optionally, the user interface 303 may also include a standard wired interface and a wireless interface.

[0084] Among them, the network interface 304 may optionally include a standard wired interface, a wireless interface (such as a WI-FI interface).

[0085] Among them, the processor 301 may include one or more processing cores. The processor 301 connects various parts within the entire server through various interfaces and lines. By running or executing instructions, programs, code sets, or instruction sets stored in the memory 305, and by calling the data stored in the memory 305, it executes various functions of the server and processes data. Optionally, the processor 301 may be implemented in at least one hardware form of digital signal processing (DSP), field-programmable gate array (FPGA), or programmable logic array (PLA). The processor 301 may integrate a combination of one or several of a central processing unit (CPU), a graphics processing unit (GPU), and a modem. Among them, the CPU mainly processes the operating system, user interface graphics, and application programs, etc.; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communication. It can be understood that the above-mentioned modem may not be integrated into the processor 301 and may be implemented separately by a single chip.

[0086] Among them, the memory 305 may include random access memory (RAM), and may also include read-only memory. Optionally, the memory 305 includes a non-transitory computer-readable storage medium. The memory 305 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 305 may include a program storage area and a data storage area. Among them, the program storage area may store instructions for implementing the operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store the data involved in the above-mentioned various method embodiments. The memory 305 is optionally also at least one storage device located far from the aforementioned processor 301. Refer to Figure 3 , as a computer storage medium, the memory 305 may include an operating system, a network communication module, a user interface module, and an application program for a method of detecting intrusion in checked luggage.

[0087] InFigure 3 In the electronic device 300 shown, the user interface 303 is mainly used to provide an interface for the user to input and obtain the data input by the user; and the processor 301 can be used to call the application program stored in the memory 305 for a method of detecting intrusion in checked luggage. When executed by one or more processors 301, the electronic device 300 is caused to execute the method as described in one or more of the above embodiments. It should be noted that for the foregoing method embodiments, for simplicity of description, they are all expressed as a series of combinations of actions. However, those skilled in the art should know that this application is not limited by the described order of actions, because according to this application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0088] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0089] In several implementation manners provided by this application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be through some service interfaces. The indirect coupling or communication connection of the device or unit can be in an electrical or other form.

[0090] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place, or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0091] In addition, in each embodiment of this application, the functional units can be integrated in one processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units.

[0092] When the integrated unit 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 memory. Based on this understanding, the technical solution of the present application, in essence, 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. This computer software product is stored in a memory and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned memory includes various media that can store program codes, such as USB flash drives, mobile hard disks, magnetic disks, or optical discs.

[0093] The above are only exemplary embodiments of the present disclosure, and the scope of the present disclosure cannot be limited thereby. That is, any equivalent changes and modifications made in accordance with the teachings of the present disclosure still fall within the scope covered by the present disclosure. Those skilled in the art will readily think of other implementation manners of the present disclosure after considering the specification and the practice of the present disclosure.

[0094] The present application aims to cover any variations, uses, or adaptive changes of the present disclosure. These variations, uses, or adaptive changes follow the general principles of the present disclosure and include common general knowledge or conventional technical means in the technical field not recorded in the present disclosure. The specification and the embodiments are only regarded as exemplary.

Claims

1. A baggage check-in intrusion detection method, characterized in that: include: Acquire baggage status information and checked video information of the target baggage, wherein the baggage status information includes baggage opening and closing data, pressure distribution data, and temperature data; Associating the checked video information with the electronic identification of the target luggage to obtain a spatiotemporal mapping relationship between the target luggage and the checked video information; Performing anomaly detection on the luggage opening and closing data, pressure distribution data, and temperature data to obtain an anomaly detection result, and determining a timestamp and luggage position coordinates corresponding to the anomaly detection result; Extracting the video data corresponding to the timestamp and the luggage location coordinates from the time-space mapping relationship, and identifying the suspicious person who has contacted the target luggage and the operation behavior of the suspicious person in the video data; When the operation behavior is confirmed to be a non-preset authorized behavior, an intrusion event report of the suspicious person is generated.

2. The baggage check-in intrusion detection method according to claim 1, characterized in that: The step of associating the checked video information with the electronic identification of the target luggage to obtain a spatiotemporal mapping relationship between the target luggage and the checked video information includes: Obtaining luggage trajectory coordinate data and time series data in the checked video information; Reading the electronic identification information of the target luggage, wherein the electronic identification information includes a luggage ID and a timestamp; Based on the baggage ID, a mapping table between the baggage trajectory coordinate data and the time series data is established; The corresponding checked-in video segment is determined in the mapping table according to the timestamp, and serves as the spatiotemporal mapping relationship between the target luggage and the checked-in video information.

3. The baggage check-in intrusion detection method according to claim 1, characterized in that: The abnormality detection is performed on the luggage opening and closing data, the pressure distribution data and the temperature data to obtain the abnormality detection result, including: Determining an opening and closing abnormality level according to the opening and closing frequency and duration in the luggage opening and closing data; Determining the pressure anomaly level according to the pressure center displacement and the pressure peak value in the pressure distribution data; Obtaining a temperature anomaly grade according to a mutation point and a temperature gradient of the temperature data; When any abnormal level among the opening and closing abnormal level, the pressure abnormal level and the temperature abnormal level exceeds the abnormal level threshold, the corresponding data time point and abnormal level are recorded as the abnormal detection result.

4. The baggage check-in intrusion detection method according to claim 1, characterized in that: The determining the timestamp and luggage location coordinates corresponding to the abnormal detection result includes: Obtaining the abnormal event triggering time point in the abnormality detection result; Marking the data in a preset time period before and after the abnormal event triggering time point as a focus time window; Extracting the moment with the highest degree of abnormality from the concerned time window and using it as the timestamp corresponding to the abnormality detection result; The position tracking data corresponding to the timestamp is acquired, and the three-dimensional coordinates in the position tracking data are extracted as the luggage position coordinates corresponding to the abnormality detection result.

5. The baggage check-in intrusion detection method according to claim 4, characterized in that: The step of extracting the moment with the highest abnormality from the concerned time window comprises: Calculate the weighted average of the opening and closing abnormality level, the pressure abnormality level and the temperature abnormality level within the concerned time window, and generate a comprehensive abnormality level curve corresponding to each of the weighted average values; Performing peak detection on the comprehensive abnormality level curve to obtain an abnormal peak point set; Calculate the peak amplitude and duration of each peak point in the abnormal peak point set; The time corresponding to the peak point where the product of the peak amplitude and the duration is the largest is taken as the time at which the abnormality is the highest.

6. The baggage check-in intrusion detection method according to claim 1, characterized in that: The extracting the video data corresponding to the timestamp and the luggage location coordinates from the time-space mapping relationship includes: Determine a corresponding video time point in the time-space mapping relationship according to the timestamp; Taking the video time point as the center, intercepting a video segment of a preset length; determining a target detection area in the video clip based on the luggage location coordinates; Perform image enhancement processing on the target detection area to obtain video data.

7. The baggage check-in intrusion detection method according to claim 1, characterized in that: The identifying the suspicious person who has contacted the target luggage and the operation behavior of the suspicious person in the video data includes: Tracking the trajectory of a person on the video data to obtain trajectory data of the person; Calculating the spatial distance between the personnel trajectory data and the target luggage position; When the spatial distance is less than the preset contact distance, the corresponding person is marked as a suspicious person; A frame image sequence of the suspicious person during the contact process is extracted, and the frame image sequence is matched based on a preset behavior recognition template to obtain the operation behavior of the suspicious person.

8. A baggage check-in intrusion detection system, characterized in that: The system comprises: An information acquisition module, used to acquire luggage status information and checked video information of the target luggage, wherein the luggage status information includes luggage opening and closing data, pressure distribution data, and temperature data; A mapping relationship determination module, used to associate the checked video information with the electronic identification of the target luggage to obtain a spatiotemporal mapping relationship between the target luggage and the checked video information; an operation behavior recognition module, configured to perform anomaly detection on the luggage opening and closing data, pressure distribution data, and temperature data, obtain an anomaly detection result, and determine a timestamp and luggage location coordinates corresponding to the anomaly detection result; extract video data corresponding to the timestamp and luggage location coordinates from the time-space mapping relationship, and recognize a suspicious person who has contacted the target luggage and the operation behavior of the suspicious person in the video data; The intrusion detection module is used to generate an intrusion event report of the suspicious person when the operation behavior is confirmed to be a non-preset authorized behavior.

9. An electronic device, characterized in that: It includes a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes the baggage check-in intrusion detection method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed, the baggage check-in intrusion detection method according to any one of claims 1 to 7 is executed.

Images