A method for three-dimensional comparison and analysis of explosive information
By using a three-dimensional comparative analysis method, the problems of accuracy and traceability efficiency in explosives detection in airport security checks have been solved, realizing automated handling of dangerous goods and personnel tracking, and improving the safety and efficiency of security checks.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-06-09
- Publication Date
- 2026-03-27
AI Technical Summary
Current airport security checks do not pay enough attention to the surface material of items inside packages when detecting explosives, which reduces the accuracy of detection, increases the incidence of aircraft accidents, and the subjective judgment of staff may endanger life safety. In addition, the tracing process is time-consuming and wasteful of resources.
Using a three-dimensional comparison and analysis method for explosive information, images are acquired through package delivery detection. The initial similarity coefficient and apparent grayscale value of the object and the dangerous goods are analyzed to automatically determine whether the package is a dangerous goods, track the associated personnel, and automatically push it to the explosion-proof area.
It has improved the accuracy of airport explosives detection, reduced the incidence of aircraft accidents, lowered security inspection and maintenance costs, reduced the risk to the lives of staff, shortened the traceability time, and enhanced the value of dangerous package analysis.
Smart Images
Figure CN116682063B_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of explosive information comparison, in particular to an explosive information three-dimensional comparison analysis method. BACKGROUND
[0002] Airport security is a very important link, airport security is the first line of defense to protect passengers and crew safety, through the inspection of all passengers on commercial flights, to ensure that any dangerous goods will not pass through the security checkpoint, so as to prevent terrorist attacks and illegal acts, airport security can help prevent air accident, which may be caused by dangerous goods in the high altitude caused by fire or explosion and other safety problems, through airport security, can prevent dangerous goods from being taken on the plane, and regulate the requirements of stacking and carrying goods, reduce the risk of accidents, among the many dangerous goods, explosive goods have strong destructive power, if the detection is not accurate, it will affect the safety of the subsequent aircraft and airport, therefore, it is extremely necessary to detect explosives in the airport.
[0003] The existing airport security can meet the basic requirements in the detection of explosive goods, but there are still some defects, which are embodied in the following aspects: (1) the existing airport security pays little attention to the apparent material of the goods in the package during the detection of explosive goods, which reduces the accuracy of airport explosive detection to a certain extent, thereby increasing the incidence of aircraft accidents and air disasters, which is not conducive to the long-term sustainable development of the airport industry.
[0004] (2) When the existing airport security detects a package that may contain explosives, it is mostly handled by the staff wearing anti-explosive devices, which increases the maintenance cost of airport security and reduces the related income of the airport to a certain extent. On the other hand, due to the subjective judgment of the staff, there may be inappropriate handling of explosive goods, which poses a great threat to the safety of the staff.
[0005] (3) When the existing airport security traces the dangerous package, the staff mostly trace it through monitoring, which is relatively slow due to the subjectivity of the staff, thereby prolonging the tracing time of the dangerous package to a certain extent, wasting a lot of resources when checking the relevant personnel, reducing the value of dangerous package analysis, and thus affecting the reputation of the relevant airlines. SUMMARY
[0006] In order to overcome the shortcomings in the background art, the present application provides an explosive information three-dimensional comparison analysis method, which can effectively solve the problems in the above background art.
[0007] The object of the application can be realized by the following technical scheme: a kind of explosive information three-dimensional comparison analysis method, comprising: S1. conveying package detection: when package conveying is carried out in security inspection conveyor, each conveying package is detected, and then the image of each conveying package in each detection time period is obtained.
[0008] S2. conveying package analysis: according to the image of each conveying package in each detection time period, the initial similarity coefficient corresponding to each object and each dangerous article of each conveying package in each detection time period is analyzed, then each to-be-resolved package in each detection time period is screened, and the corresponding target object is obtained, so as to obtain each target dangerous article corresponding to each target object of each to-be-resolved package in each detection time period, and the similarity evaluation index of each target object of each to-be-resolved package in each detection time period and specified dangerous article is analyzed.
[0009] S3. conveying package danger judgment: whether each to-be-resolved package in each detection time period belongs to dangerous article is judged, and then each target to-be-handled dangerous package in each detection time period is obtained.
[0010] S4. dangerous article processing: the predicted arrival time point corresponding to each target to-be-handled dangerous package in each detection time period is analyzed, and each target to-be-handled dangerous package in each detection time period is pushed to the corresponding explosion-proof area at the predicted arrival time point.
[0011] S5. dangerous article associated personnel analysis: according to each target to-be-handled dangerous package in each detection time period, the face image and whole-process flow video of each associated personnel corresponding to each target to-be-handled dangerous package in each detection time period are analyzed.
[0012] S6. associated personnel related information processing, the face image and whole-process flow video of each associated personnel corresponding to each target to-be-handled dangerous package in each detection time period are displayed.
[0013] As a preferred scheme, the initial similarity coefficient corresponding to each object and each dangerous article of each conveying package in each detection time period is analyzed, and the specific method is as follows: according to the image of each conveying package in each detection time period, the contour of each object corresponding to each conveying package in each detection time period is obtained.
[0014] The image corresponding to each dangerous article is extracted from the cloud database, and the contour corresponding to each dangerous article is obtained.
[0015] The contour of each object corresponding to each conveying package in each detection time period is compared with the contour corresponding to each dangerous article, and then the overlapping volume of each object corresponding to each conveying package and each dangerous article in each detection time period is obtained wherein i represents the number of each detection period, i = 1, 2, …, n, m represents the number of each delivery package, m = 1, 2, …, l, p represents the number of each object, p = 1, 2, …, q, and k represents the number of each dangerous object, k = 1, 2, …, j.
[0016] The image corresponding to each dangerous object is extracted from the cloud database, and then the size information corresponding to each dangerous object is obtained, and the length, width, height, and volume of each dangerous object are marked as C' k , K' k , G' k , and V' k , respectively.
[0017] The initial similarity coefficients corresponding to each object and each dangerous object of each delivery package belonging to each detection period are analyzed. wherein is the size similarity coefficient corresponding to the pth object and the kth dangerous object of the mth delivery package belonging to the ith detection period, and γ1 and γ2 represent the preset proportion factors of the object and the dangerous object corresponding to the size similarity and the contour similarity, respectively.
[0018] As a preferred scheme, the specific analysis method of the size similarity coefficient corresponding to each object and each dangerous object of each delivery package belonging to each detection period is as follows: the size information of each object corresponding to each delivery package belonging to each detection period is obtained according to the image of each delivery package belonging to each detection period, wherein the size information includes length width height and volume
[0019] The size similarity coefficient corresponding to each object and each dangerous object of each delivery package belonging to each detection period is analyzed. wherein wherein λ1, λ2, λ3, and λ4 represent the weight influence factors corresponding to the preset length similarity, width similarity, height similarity, and volume similarity, respectively.
[0020] As a preferred scheme, the specific method of the similarity evaluation index of each target object and the specified dangerous object of each delivery package belonging to each detection period is as follows: the specified dangerous object corresponding to each target object of each delivery package belonging to each detection period is analyzed according to the initial similarity coefficient corresponding to each object and each dangerous object of each delivery package belonging to each detection period.
[0021] Extracting the apparent image corresponding to each dangerous article from the cloud database, and randomly selecting each gray value therefrom, and then constructing the apparent reference gray value range corresponding to each dangerous article according to the same, so as to obtain the apparent reference gray value range of the specified dangerous article corresponding to each target object of each to-be-analyzed parcel belonging to each detection time period.
[0022] Obtaining each gray value of each target object corresponding to each to-be-analyzed parcel belonging to each detection time period, and analyzing the area of each apparent gray value region corresponding to each target object of each to-be-analyzed parcel belonging to each detection time period.
[0023] Summarizing the area of each apparent gray value region corresponding to each target object of each to-be-analyzed parcel belonging to each detection time period, and then obtaining the total area of the apparent gray value region corresponding to each target object of each to-be-analyzed parcel belonging to each detection time period. Wherein h represents the number of each to-be-analyzed parcel, h = 1, 2, …, g, and f represents the number of each target object, m = 1, 2, …, l.
[0024] Obtaining the surface area corresponding to each target object of each to-be-analyzed parcel belonging to each detection time period.
[0025] Analyzing the similarity evaluation index of each target object of each to-be-analyzed parcel belonging to each detection time period and the specified dangerous article. Wherein e is a natural constant, and χ1 represents a preset correction factor corresponding to the apparent gray similarity.
[0026] As a preferred scheme, the specific analysis method of the specified dangerous article corresponding to each target object of each to-be-analyzed parcel belonging to each detection time period is as follows: obtaining the initial similarity coefficient of each target object of each to-be-analyzed parcel belonging to each detection time period and each dangerous article, and then screening the target dangerous article corresponding to the maximum initial similarity coefficient, and taking it as the specified dangerous article, and then obtaining the specified dangerous article corresponding to each target object of each to-be-analyzed parcel belonging to each detection time period.
[0027] As a preferred scheme, the specific judgment method of whether each to-be-analyzed parcel belonging to each detection time period belongs to a dangerous article, and then obtaining each target to-be-processed dangerous parcel belonging to each detection time period is as follows: comparing the similarity evaluation index of each target object of each to-be-analyzed parcel belonging to each detection time period and the specified dangerous article with a predefined similarity evaluation index threshold value, if the similarity evaluation index of a target object of a to-be-analyzed parcel belonging to a detection time period and the specified dangerous article is greater than or equal to the similarity evaluation index threshold value, then marking the to-be-analyzed parcel as a target to-be-processed dangerous parcel, and then obtaining each target to-be-processed dangerous parcel belonging to each detection time period.
[0028] As a preferred scheme, the specific analysis method of the expected arrival time point corresponding to each target dangerous parcel belonging to each detection time period is as follows: a two-dimensional rectangular coordinate system is established with the long side of the security conveyor as the x-axis and the short side of the security conveyor as the y-axis, and then the coordinates of the center point corresponding to each target dangerous parcel belonging to each detection time period are obtained.
[0029] A point on the boundary corresponding to the conveying terminal of the security conveyor is randomly selected and marked as a reference point, and then the x-axis coordinate value corresponding to the reference point is obtained.
[0030] The x-axis coordinate value in the center point coordinate corresponding to each target dangerous parcel belonging to each detection time period is subtracted by the x-axis coordinate value corresponding to the reference point, and then the x-axis deviation value of each target dangerous parcel belonging to each detection time period from the conveying terminal of the security conveyor is obtained, and the x-axis deviation value is taken as the expected conveying distance corresponding to each target dangerous parcel belonging to each detection time period.
[0031] The expected conveying distance corresponding to each target dangerous parcel belonging to each detection time period is divided by the conveying speed of the conveying belt stored in the cloud database, and then the expected conveying duration corresponding to each target dangerous parcel belonging to each detection time period is obtained.
[0032] According to the expected conveying duration corresponding to each target dangerous parcel belonging to each detection time period and the current time point, the expected arrival time point corresponding to each target dangerous parcel belonging to each detection time period is obtained.
[0033] As a preferred scheme, the specific analysis method of the face image and the whole-process moving video of each associated person corresponding to each target dangerous parcel belonging to each detection time period is as follows: the whole-process video of each target dangerous parcel belonging to each detection time period is obtained from the monitoring video, and then the first carrying person, the final placing person and the midway contact person corresponding to each target dangerous parcel belonging to each detection time period are obtained from the monitoring video, and are marked as each associated person, and then the face image of each associated person is obtained, and the whole-process moving video corresponding to each associated person is obtained, and then the face image and the whole-process moving video of each associated person corresponding to each target dangerous parcel belonging to each detection time period are obtained.
[0034] Compared with the prior art, the embodiments of the present application have at least the following advantages or beneficial effects: (1) the present application detects the conveying parcel in the conveying parcel detection, and then obtains the image of the conveying parcel, which lays a foundation for the subsequent similarity assessment of the conveying parcel and the dangerous goods.
[0035] (2) The application analyzes the similarity of the delivery package and dangerous goods in the delivery package analysis, and analyzes the apparent material quality of the internal goods of the package by using the apparent gray value of the package, which makes up for the defect that the apparent material quality of the internal goods of the package is not paid attention to in the prior art, and thus improves the accuracy of airport explosive detection to a certain extent, reduces the incidence of aircraft accidents, and is beneficial to the long-term sustainable development of the airport related industry.
[0036] (3) The application determines the danger of the delivery package in the delivery package danger judgment, and thus provides data support for the subsequent processing of dangerous goods.
[0037] (4) In the dangerous goods processing, the dangerous goods are automatically pushed into the explosion-proof area, which reduces the maintenance cost of airport security to a certain extent, improves the related income of the airport, overcomes the defect of the subjective initiative of the staff, avoids the phenomenon that the explosive goods are not suitable for processing, and reduces the harm to the life safety of the staff.
[0038] (5) In the dangerous goods related personnel analysis, the dangerous package is tracked from the monitoring video, which makes up for the defect that the staff traces the source through monitoring in the prior art, and thus shortens the trace time of the dangerous package, so as to reduce the waste of resources when the related personnel are investigated, improve the value of the dangerous package analysis, and thus reduce the reputation influence of the related airlines. BRIEF DESCRIPTION OF DRAWINGS
[0039] The application will be further described by using the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the application. For ordinary skilled persons in the art, other drawings can be obtained without creative labor on the basis of the following drawings.
[0040] Figure 1 The method flowchart of the application. DETAILED DESCRIPTION
[0041] The technical solutions in the embodiments of the application will be described clearly and completely in combination with the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, not all. Based on the embodiments in the application, all other embodiments obtained by ordinary skilled persons in the art without creative labor belong to the protection scope of the application.
[0042] Referring to Figure 1 The application provides a three-dimensional comparison and analysis method for explosive information, which comprises the following steps: S1. Delivery package detection: when the package is conveyed by the security conveyor, each delivery package is detected, and thus the images of each delivery package in each detection time period are obtained.
[0043] The application detects the conveying package in conveying package detection, and further obtains an image of the conveying package, thereby laying a foundation for subsequent similarity evaluation of the conveying package and dangerous goods.
[0044] S2. Conveying package analysis: according to the image of each conveying package in each detection time period, the initial similarity coefficients corresponding to each object and each dangerous good of each conveying package in each detection time period are analyzed, and each conveying package to be analyzed in each detection time period is screened, and each target object corresponding thereto is obtained, so as to obtain each target dangerous good of each target object corresponding to each conveying package to be analyzed in each detection time period, and the similarity evaluation index of each target object of each conveying package to be analyzed in each detection time period and the specified dangerous good is analyzed.
[0045] In specific embodiments of the application, the initial similarity coefficients corresponding to each object and each dangerous good of each conveying package in each detection time period are analyzed in the following specific method: according to the image of each conveying package in each detection time period, the contour of each object corresponding to each conveying package in each detection time period is obtained.
[0046] The image corresponding to each dangerous good is extracted from the cloud database, and the contour corresponding to each dangerous good is obtained.
[0047] The contour of each object corresponding to each conveying package in each detection time period is compared with the contour corresponding to each dangerous good, and the overlapping volume of each object corresponding to each conveying package and each dangerous good in each detection time period is obtained. Where i represents the number of each detection time period, i = 1, 2,..., n, m represents the number of each conveying package, m = 1, 2,..., l, p represents the number of each object, p = 1, 2,..., q, and k represents the number of each dangerous good, k = 1, 2,..., j.
[0048] The image corresponding to each dangerous good is extracted from the cloud database, and the size information corresponding to each dangerous good is obtained, and the length, width, height and volume of each dangerous good are marked as C' k , K' k , G' k and V' k , respectively.
[0049] The initial similarity coefficients corresponding to each object and each dangerous good of each conveying package in each detection time period are analyzed Where is the size similarity coefficient corresponding to the pth object of the mth conveying package in the ith detection time period and the kth dangerous good, and γ1 and γ2 represent the preset object and dangerous good corresponding size similarity and contour similarity proportion factor, respectively.
[0050] In specific embodiments of the present application, the size similarity coefficient of each object of each delivery package in each detection time period corresponding to each dangerous article is analyzed as follows: size information of each object of each delivery package in each detection time period is obtained according to images of each delivery package in each detection time period, wherein the size information includes length width height and volume
[0051] The size similarity coefficient of each object of each delivery package in each detection time period corresponding to each dangerous article is analyzed wherein
[0052] wherein λ1, λ2, λ3, λ4 represent the weight influence factor corresponding to the preset length similarity, width similarity, height similarity and volume similarity, respectively.
[0053] In specific embodiments of the present application, the similarity evaluation index of each target object of each to-be-analyzed package in each detection time period corresponding to a specified dangerous article is analyzed as follows: the specified dangerous article corresponding to each target object of each to-be-analyzed package in each detection time period is analyzed according to the initial similarity coefficient of each object of each delivery package in each detection time period corresponding to each dangerous article.
[0054] The apparent image corresponding to each dangerous article is extracted from the cloud database, and each gray value is randomly selected therefrom, and then the apparent reference gray value range corresponding to each dangerous article is constructed according to the gray values, so as to obtain the apparent reference gray value range of the specified dangerous article corresponding to each target object of each to-be-analyzed package in each detection time period.
[0055] Each gray value of each target object of each to-be-analyzed package in each detection time period is obtained, and the area of each apparent gray value region of each target object of each to-be-analyzed package in each detection time period is analyzed.
[0056] It should be noted that each gray value of each target object of each to-be-analyzed package in each detection time period is compared with the apparent reference gray value range of the specified dangerous article, if a certain gray value is within the apparent reference gray value range, the gray value is marked as an apparent gray value, the region corresponding to the apparent gray value is marked as an apparent gray value region, and the area of the apparent gray value region is obtained, and then the area of each apparent gray value region of each target object of each to-be-analyzed package in each detection time period is obtained
[0057] The area of each target object corresponding to each apparent gray value region of each parcel to be analyzed in each detection time period is summarized to obtain the total area of the apparent gray value region corresponding to each target object of each parcel to be analyzed in each detection time period Wherein h represents the number of each parcel to be analyzed, h=1, 2,..., g, and f represents the number of each target object, m=1, 2,..., l.
[0058] The surface area corresponding to each target object of each parcel to be analyzed in each detection time period is obtained
[0059] The similarity evaluation index of each target object of each parcel to be analyzed in each detection time period with a specified dangerous article is analyzed Wherein e is a natural constant, and χ1 represents a preset correction factor corresponding to the apparent gray similarity.
[0060] In specific embodiments of the present application, the specific analysis method of the specified dangerous article corresponding to each target object of each parcel to be analyzed in each detection time period is as follows: the initial similarity coefficient of each target object of each parcel to be analyzed in each detection time period with each target dangerous article is obtained according to the initial similarity coefficient of each object corresponding to each parcel in each detection time period with each dangerous article, then the target dangerous article corresponding to the maximum initial similarity coefficient is screened and selected as the specified dangerous article, and thus the specified dangerous article corresponding to each target object of each parcel to be analyzed in each detection time period is obtained.
[0061] It should be noted that the specific method for screening each parcel to be analyzed in each detection time period and obtaining the corresponding target object, and thus obtaining the target dangerous article corresponding to each target object of each parcel to be analyzed in each detection time period, is as follows: the initial similarity coefficient of each object corresponding to each parcel in each detection time period with each dangerous article is compared with a preset initial similarity coefficient threshold value, if the initial similarity coefficient of a certain object corresponding to a certain parcel in a certain detection time period with a certain dangerous article is greater than or equal to the initial similarity coefficient threshold value, then the parcel is marked as a parcel to be analyzed, the object is marked as a target object, and the dangerous article is marked as a target dangerous article, and thus the target dangerous article corresponding to each target object of each parcel to be analyzed in each detection time period is obtained.
[0062] In the present application, the similarity of the parcel and the dangerous article is analyzed in the parcel analysis, and the apparent material of the parcel is analyzed by using the apparent gray value of the parcel, which makes up for the defect that the apparent material of the parcel is not paid enough attention to in the prior art, and thus the accuracy of the airport explosive detection is improved to a certain extent, the incidence of aircraft accidents is reduced, and the long-term sustainable development of the airport related industry is facilitated.
[0063] S3. Judgment of dangerousness of the delivery package: judging whether each parcel to be analyzed belonging to each detection time period is dangerous goods, and further obtaining each target dangerous parcel to be processed belonging to each detection time period.
[0064] In specific embodiments of the present application, the judgment of whether each parcel to be analyzed belonging to each detection time period is dangerous goods, and further obtaining each target dangerous parcel to be processed belonging to each detection time period, the specific judgment method is: comparing the similarity evaluation index of each target object of each parcel to be analyzed belonging to each detection time period with the similarity evaluation index of the specified dangerous goods with the pre-defined similarity evaluation index threshold, if the similarity evaluation index of a certain target object of a certain parcel to be analyzed belonging to a certain detection time period is greater than or equal to the similarity evaluation index threshold, then the parcel to be analyzed is marked as a target dangerous parcel to be processed, and further obtaining each target dangerous parcel to be processed belonging to each detection time period.
[0065] In the judgment of dangerousness of the delivery package, the present application judges the dangerousness of the delivery package, and further provides data support for the subsequent processing of dangerous goods.
[0066] S4. Dangerous goods processing: analyzing the predicted arrival time point corresponding to each target dangerous parcel to be processed belonging to each detection time period, and pushing each target dangerous parcel to be processed belonging to each detection time period to the corresponding explosion-proof area at the predicted arrival time point.
[0067] In specific embodiments of the present application, the predicted arrival time point corresponding to each target dangerous parcel to be processed belonging to each detection time period, the specific analysis method is: establishing a two-dimensional rectangular coordinate system with the long side of the security conveyor as the x-axis and the short side of the security conveyor as the y-axis, and further obtaining the coordinates of the center point corresponding to each target dangerous parcel to be processed belonging to each detection time period.
[0068] Randomly selecting a point on the boundary corresponding to the conveying terminal of the security conveyor, and marking it as a reference point, and further obtaining the x-axis coordinate value corresponding to the reference point.
[0069] Subtracting the x-axis coordinate value corresponding to the reference point from the x-axis coordinate value in the center point coordinates corresponding to each target dangerous parcel to be processed belonging to each detection time period, and further obtaining the x-axis deviation value of each target dangerous parcel to be processed belonging to each detection time period from the conveying terminal of the security conveyor, and taking it as the predicted conveying distance corresponding to each target dangerous parcel to be processed belonging to each detection time period.
[0070] Dividing the predicted conveying distance corresponding to each target dangerous parcel to be processed belonging to each detection time period by the conveying speed of the conveyor belt stored in the cloud database, and further obtaining the predicted conveying duration corresponding to each target dangerous parcel to be processed belonging to each detection time period.
[0071] The expected arrival time points of the target dangerous parcels in each detection time period are obtained according to the expected conveying time corresponding to the target dangerous parcels in each detection time period and the current time point.
[0072] It should be noted that the expected arrival time points of the target dangerous parcels in each detection time period are obtained by adding the expected conveying time corresponding to the target dangerous parcels in each detection time period to the current time point.
[0073] The dangerous goods are automatically pushed into the explosion-proof area in the processing of the dangerous goods, which reduces the maintenance cost of airport security, improves the related income of the airport to a certain extent, overcomes the defects of the subjective initiative of the staff, avoids the phenomenon that the explosive goods are not suitable for processing, and reduces the harm to the life safety of the staff.
[0074] S5. Dangerous goods associated personnel analysis: analyzing the face images and whole-process moving videos of the associated personnel corresponding to the target dangerous parcels in each detection time period according to the target dangerous parcels in each detection time period.
[0075] In the specific embodiments of the present application, the specific analysis method of the face images and whole-process moving videos of the associated personnel corresponding to the target dangerous parcels in each detection time period is as follows: obtaining the whole-process videos of the target dangerous parcels in each detection time period from the monitoring videos, then obtaining the first carrying personnel, the final placing personnel and the midway contact personnel corresponding to the target dangerous parcels in each detection time period from the monitoring videos, marking them as the associated personnel, obtaining the face images of the associated personnel, obtaining the whole-process moving videos corresponding to the associated personnel, and then obtaining the face images and whole-process moving videos of the associated personnel corresponding to the target dangerous parcels in each detection time period.
[0076] It should be noted that the whole-process videos of the target dangerous parcels in each detection time period are obtained by performing feature recognition on the target dangerous parcels in each detection time period from the monitoring videos.
[0077] In the dangerous goods associated personnel analysis, the dangerous parcels are tracked from the monitoring videos, which makes up for the defects of the staff tracing through monitoring in the prior art, shortens the tracing time of the dangerous parcels, reduces the waste of resources when the related personnel are investigated, improves the value of the dangerous parcel analysis, and reduces the reputation influence of the related airlines.
[0078] S6. Associated personnel related information processing: displaying the face images and whole-process moving videos of the associated personnel corresponding to the target dangerous parcels in each detection time period.
[0079] The above merely illustrates and explains the structure of the present application. Those skilled in the art can make various modifications or supplements to the described specific embodiments or replace them with similar ways, as long as they do not deviate from the structure of the present application or exceed the scope defined by the present application, which shall belong to the protection scope of the present application.
Claims
1. A three-dimensional comparative analysis method for explosive information, characterized in that, include: S1. Package Inspection: During the transport of packages on the security conveyor, each package is inspected to obtain images of each package within each inspection time period. S2. Parcel Delivery Analysis: Based on the images of each parcel delivered within each detection time period, analyze the initial similarity coefficients between each object and each hazardous material in each parcel delivered within each detection time period. Then, filter each parcel to be analyzed within each detection time period and obtain its corresponding target objects. This allows for the acquisition of each target hazardous material corresponding to each target object in each parcel to be analyzed within each detection time period. Finally, analyze the similarity evaluation index between each target object in each parcel to be analyzed within each detection time period and the specified hazardous material. S3. Hazard assessment of transported packages: Determine whether each package to be analyzed in each detection time period is a dangerous item, and thus obtain each target dangerous package to be processed in each detection time period; S4. Hazardous Materials Handling: Analyze the estimated arrival time of each target hazardous package to be handled in each detection time period, and push each target hazardous package to be handled in each detection time period to the corresponding explosion-proof area at the estimated arrival time. S5. Analysis of personnel associated with dangerous goods: Based on the dangerous packages to be processed for each target in each detection time period, analyze the facial images and full-process video of the personnel associated with each dangerous package to be processed for each target in each detection time period; S6. Processing of related personnel information: Displaying the facial images and full-process video of the relevant personnel associated with each target hazardous package to be processed in each detection time period; Step S2 includes: Based on the initial similarity coefficient, the designated dangerous goods corresponding to each target object of each package to be parsed are analyzed; the appearance image corresponding to each dangerous goods is extracted from the cloud database, and each gray value is randomly selected from it to construct the appearance reference gray value range corresponding to each dangerous goods, thereby obtaining the appearance reference gray value range corresponding to each target object of each package to be parsed in each detection time period. The gray values of each target object in each package to be parsed are compared with the apparent reference gray value range of the corresponding designated dangerous goods. If a gray value is within the apparent reference gray value range, the gray value is marked as an apparent gray value, and the area corresponding to the apparent gray value is obtained and marked as an apparent gray value area. The area of the apparent gray value area is then obtained, thereby obtaining the area of each apparent gray value area corresponding to each target object in each package to be parsed. The areas of the apparent grayscale regions corresponding to each target object in each package to be parsed are summed to obtain the total area of the apparent grayscale regions corresponding to each target object in each package to be parsed. ,in This represents the number of each package to be parsed. , This is represented by the number of each target object. ; Obtain the surface area of each target object in the package to be parsed. ; Analyze the similarity assessment index between each target object in each package to be analyzed and the specified hazardous materials. ,in It is a natural constant. This represents the correction factor corresponding to the preset apparent grayscale similarity.
2. The method for three-dimensional comparison and analysis of explosive information according to claim 1, characterized in that: The specific method for analyzing the initial similarity coefficients between each object in each delivery package and each hazardous material within each detection time period is as follows: Based on the images of each transported package within each detection time period, obtain the outline of each object corresponding to each transported package within each detection time period; Extract images of each hazardous item from the cloud database and obtain the outline of each hazardous item. The contours of each object in each transported package within each inspection time period are compared with the contours of each hazardous material to obtain the overlapping volume of each object and each hazardous material within each inspection time period. ,in This is represented by the number of each detection time period. , This indicates the number of each delivered package. , This is represented by the number of each object. , This is indicated by the code for each hazardous material. ; Images of each hazardous material are extracted from the cloud database, and their corresponding size information is obtained. The length, width, height, and volume of each hazardous material are then labeled as follows: , , and ; Analyze the initial similarity coefficients between each object in each transported package and each hazardous material within each detection time period. ,in For the first The detection time period belongs to the [number]th [time period]. The first package delivered The object and the first The size similarity coefficient of each dangerous item , These represent the pre-defined proportion factors for objects and hazardous materials that are similar in size and outline.
3. The method for three-dimensional comparison and analysis of explosive information according to claim 2, characterized in that: The specific analysis method for the dimensional similarity coefficient between each object in each transported package and each hazardous material within each inspection time period is as follows: Based on the images of each transported package within each detection time period, obtain the size information of each object corresponding to each transported package within each detection time period, where the size information includes length. ,width ,high and volume ; Analyze the dimensional similarity coefficients between each object in each transported package and each hazardous material within each detection time period. ,in , , , ,in , , , These represent the preset weighted influence factors for similar length, similar width, similar height, and similar volume, respectively.
4. The method for three-dimensional comparison and analysis of explosive information according to claim 1, characterized in that: The specific analysis method for the designated dangerous items corresponding to each target object of each package to be analyzed is as follows: based on the initial similarity coefficient between each object and each dangerous item of each transported package in each detection time period, the initial similarity coefficient between each target object and each target dangerous item of each package to be analyzed in each detection time period is obtained, and then the target dangerous item corresponding to the maximum initial similarity coefficient is selected and used as the designated dangerous item, thereby obtaining the designated dangerous items corresponding to each target object of each package to be analyzed in each detection time period.
5. The method for three-dimensional comparison and analysis of explosive information according to claim 1, characterized in that: The specific method for determining whether each package to be analyzed in each detection time period belongs to dangerous goods, and thus obtaining each target dangerous package to be processed in each detection time period, is as follows: the similarity evaluation index of each target object of each package to be analyzed in each detection time period and the designated dangerous goods is compared with a predefined similarity evaluation index threshold. If the similarity evaluation index of a target object of a package to be analyzed in a certain detection time period and the designated dangerous goods is greater than or equal to the similarity evaluation index threshold, then the package to be analyzed is marked as a target dangerous package to be processed, and thus the target dangerous packages to be processed in each detection time period are obtained.
6. The method for three-dimensional comparison and analysis of explosive information according to claim 1, characterized in that: The specific analysis method for the estimated arrival time of each target hazardous package to be processed within each detection time period is as follows: A two-dimensional rectangular coordinate system is established with the long side of the security inspection conveyor as the x-axis and the short side of the security inspection conveyor as the y-axis, so as to obtain the coordinates of the center point of each target dangerous package to be processed in each inspection time period; A point is randomly selected on the boundary corresponding to the conveying terminal of the security inspection conveyor, and it is marked as a reference point. Then, the x-axis coordinate value corresponding to the reference point is obtained. Subtract the x-axis coordinate value of the reference point from the x-axis coordinate value of the center point of each target dangerous package to be processed in each detection time period to obtain the x-axis deviation value between each target dangerous package to be processed in each detection time period and the conveying terminal of the security inspection conveyor, and use it as the expected conveying distance of each target dangerous package to be processed in each detection time period. Divide the estimated transport distance of each target hazardous package to be processed in each detection time period by the transport speed of the conveyor belt stored in the cloud database to obtain the estimated transport time of each target hazardous package to be processed in each detection time period. Based on the estimated delivery time of each target hazardous package to be processed in each detection time period and the current time, the estimated arrival time of each target hazardous package to be processed in each detection time period is obtained.
7. The method for three-dimensional comparison and analysis of explosive information according to claim 1, characterized in that: The specific analysis method for the facial images and full-process video of each associated person corresponding to each target hazardous package to be processed in each detection time period is as follows: the full-process video of each target hazardous package to be processed in each detection time period is obtained from the monitoring video, and then the first person carrying, the final person placing, and the person who came into contact with the target hazardous package to be processed in each detection time period are obtained from the monitoring video and marked as associated persons, then the facial images of each associated person are obtained, and the full-process video of each associated person is obtained, thus obtaining the facial images and full-process video of each associated person corresponding to each target hazardous package to be processed in each detection time period.
Citation Information
Patent Citations
Security check method and device based on person-package association
CN111323835A