Subway security inspection mode configuration management method

文档序号:1903647 发布日期:2021-11-30 浏览:29次 中文

阅读说明:本技术 一种地铁安检模式配置管理方法 (Subway security inspection mode configuration management method ) 是由 邓秋雄 严军 蒲磊 赵玲 张利 连天友 于 2021-08-31 设计创作,主要内容包括:本发明公开了一种地铁安检模式配置管理方法,所述地铁安检分级模式管理方法应用于地铁安检判图系统,所述地铁安检判图系统包括智能分析仪、图像处理装置、集中判图子系统、开包终端和安检监管平台,所述地铁安检分级模式管理方法包括:对地铁安检中的违禁品进行归类处理以及对安检分级模式进行等级配置。本发明通过对违禁品分类并对其智能识图的置信度预设置来针对不同时间段或者针对不同情境下保证不同的通行效率,具有重要的意义。(The invention discloses a subway security check mode configuration management method, which is applied to a subway security check judgment map system, wherein the subway security check judgment map system comprises an intelligent analyzer, an image processing device, a centralized judgment map subsystem, a package opening terminal and a security check supervision platform, and the subway security check hierarchical mode management method comprises the following steps: and classifying contraband in subway security inspection and carrying out level configuration on a security inspection grading mode. The invention guarantees different passing efficiencies for different time periods or under different situations by classifying contraband and presetting the confidence coefficient of the intelligent identification map, and has important significance.)

1. A subway security inspection mode configuration management method is characterized by comprising the following steps:

s1: acquiring current image data, wherein the current image data comprises a current X-ray imaging picture, a current corresponding contraband class mark of the current X-ray imaging picture and a current corresponding confidence coefficient of contraband;

s2: and matching the current contraband class mark and the confidence corresponding to the current contraband in the current image data with the contraband class configuration and the confidence threshold configuration in the security inspection modes of different grades, and directly pushing the current image data to a package opening terminal or a centralized image judging subsystem according to the matching result.

2. The management method according to claim 1,

the execution process of contraband class configuration and confidence threshold configuration in different levels of security inspection modes comprises the following steps:

SN, configure different grade security check modes, including:

the method comprises the following steps of configuring contraband categories and confidence thresholds in different levels of security inspection modes:

in a first-level security inspection mode: the method comprises the following steps of configuring contraband categories in an X-ray imaging graph and confidence coefficient threshold values corresponding to different contraband categories in the X-ray imaging graph, and under a secondary security inspection mode: the method comprises the following steps that contraband class configuration in an X-ray imaging graph and confidence coefficient threshold configuration corresponding to different contraband classes in the X-ray imaging graph are carried out in a three-level security inspection mode: the method comprises the following steps of configuring contraband categories in an X-ray imaging graph and configuring confidence threshold values corresponding to different contraband categories in the X-ray imaging graph, …, under an N-level security inspection mode: the method comprises the steps of configuring contraband categories in an X-ray imaging graph and configuring confidence coefficient threshold values corresponding to different contraband categories in the X-ray imaging graph.

3. The management method according to claim 1, wherein the management method is applied to an image processing device of a subway security inspection centralized judging map system, and the subway security inspection centralized judging map system comprises an intelligent analyzer, an image processing device, a centralized judging map subsystem, a package opening terminal and a security inspection supervision platform;

the current image data is derived from an intelligent analyzer;

the contraband category configuration and the confidence coefficient threshold configuration are from a security inspection supervision platform or an image processing device;

and the security inspection supervision platform feeds back the contraband class configuration and the confidence coefficient threshold configuration to the image processing device.

4. The management method according to claim 1, wherein the different levels of security inspection modes include a primary security inspection mode, a secondary security inspection mode, and a tertiary security inspection mode:

the first-level security inspection mode directly pushes contraband class configuration of the unpacking terminal to comprise guns, control cutters and water cup liquid, confidence thresholds of the guns, the control cutters and the water cup liquid are 80% in a peak leveling period and 90% in a peak height period;

the secondary security inspection mode directly pushes the contraband class configuration of the unpacking terminal to comprise guns, control cutters and water cup liquid, the confidence threshold values of the guns and the control cutters are 80% in the peak-off period and 90% in the peak-up period, the confidence threshold values of the water cup liquid are 90% in the peak-off period and 99% in the peak-up period;

the third-level security inspection mode is characterized in that the contraband class configuration directly pushed to the package opening terminal comprises guns and control cutters, the confidence coefficient threshold values of the guns and the control cutters are 80% in the peak leveling period and 90% in the peak period, and water cups are not classified as the contraband class directly pushed to the package opening terminal.

5. The management method according to claim 1, wherein the different-level security check mode, when executing a first-level security check mode:

if the current image data contains contraband class marks and the current corresponding contraband class marks conform to the contraband class configuration directly pushed to the package opening terminal, and the confidence level of the current corresponding contraband is greater than or equal to the confidence level threshold configuration corresponding to the contraband, pushing the current image data to the package opening terminal; if the confidence level of the current corresponding contraband is smaller than the confidence level threshold configuration corresponding to the contraband, pushing the current image data to the centralized image judgment subsystem;

if the current image data contains contraband class marks and the current corresponding contraband class marks do not conform to the contraband class configuration of the direct-pushing package opening terminal, pushing the current image data to the centralized image judging subsystem;

and if no contraband class mark exists in the current image data, pushing the current image data to the centralized image judging subsystem.

6. The management method according to claim 1, wherein the different-level security check mode, when executing the secondary security check mode:

if the current image data contains contraband class marks and the current corresponding contraband class marks conform to the contraband class configuration directly pushed to the package opening terminal, and the confidence level of the current corresponding contraband is greater than or equal to the confidence level threshold configuration corresponding to the contraband, pushing the current image data to the package opening terminal; if the confidence level of the current corresponding contraband is smaller than the confidence level threshold configuration corresponding to the contraband, pushing the current image data to the centralized image judgment subsystem;

if the current image data contains contraband class marks and the current corresponding contraband class marks do not conform to the contraband class configuration of the direct-pushing package opening terminal, pushing the current image data to the centralized image judging subsystem;

and if no contraband class mark exists in the current image data, the current image data is not pushed to the centralized image judging subsystem.

7. The management method according to claim 1, wherein the different-level security check mode, when executing a three-level security check mode:

if the current image data contains contraband class marks and the current corresponding contraband class marks conform to the contraband class configuration directly pushed to the package opening terminal, and the confidence level of the current corresponding contraband is greater than or equal to the confidence level threshold configuration corresponding to the contraband, pushing the current image data to the package opening terminal; if the confidence level of the current corresponding contraband is smaller than the confidence level threshold configuration corresponding to the contraband, pushing the current image data to the centralized image judgment subsystem;

if the current image data contains contraband class marks and the current corresponding contraband class marks do not conform to the contraband class configuration of the direct-pushing package opening terminal, pushing the current image data to the centralized image judging subsystem;

and if no contraband class mark exists in the current image data, the current image data is not pushed to the centralized image judging subsystem.

8. The management method according to claim 1, wherein the security classification mode supports a security classification mode configuration of a whole network line, a specified site, and a specified security point.

9. The management method according to claim 1, wherein the configuration of the security classification mode supports manual configuration and automatic configuration, the effective condition of the automatic configuration is automatically adjusted according to the total passenger flow number of the security door of the security inspection point, and the manual configuration further comprises setting default configuration.

10. The method of claim 1, wherein the security classification mode is further configured to configure alarm levels for different categories of contraband, and the different alarm levels are displayed by alarm lamps of different colors.

Technical Field

The invention relates to the technical field of security inspection, in particular to a subway security inspection mode configuration management method.

Background

Most of security inspection centralized judging image system security inspection services at present generally only have one working mode, one working mode cannot deal with changeable service situations, the deficiency of working mode change can be made up only by increasing cost, and the utilization rate of security inspection resources and the working efficiency are not high. In addition, some security inspection centralized image judging systems exist, different working modes of the system can be set manually, and the working modes of the image judging systems are adjusted manually mainly according to information such as the number of backlog image judging tasks, the time for distributing the image judging tasks and the like. Or, the operator of the graph determining system may also manually arrange the adjustment of the working mode according to the historical change rule of the service state, the experience in the process of developing security check work, and the like. However, the workload of the map judging personnel and the field security check personnel cannot be effectively balanced, and the traffic efficiency and the riding safety cannot be effectively adjusted.

A security inspection centralized image judging system and an automatic working mode switching control method thereof are disclosed in the patent with the application number of CN202011367948.2, and the method mainly collects real-time security inspection service data in the security inspection process; predicting the judging task pressure level of the safety inspection centralized judging system based on the change trend of the real-time safety inspection service data along with time; and controlling the security inspection centralized image judging system to automatically switch to an image judging working mode matched with the image judging task pressure level.

Disclosure of Invention

The invention aims to provide a subway security check mode configuration management method, which can be used for guaranteeing different passing efficiencies for different time periods or different situations by classifying contraband and presetting confidence coefficients of intelligent identification maps of the contraband so as to solve the problems in the background art.

A subway security inspection mode configuration management method comprises the following steps:

s1: acquiring current image data, wherein the current image data comprises a current X-ray imaging picture, a current corresponding contraband class mark of the current X-ray imaging picture and a current corresponding confidence coefficient of contraband;

s2: and matching the current contraband class mark and the confidence corresponding to the current contraband in the current image data with the contraband class configuration and the confidence threshold configuration in the security inspection modes of different grades, and directly pushing the current image data to a package opening terminal or a centralized image judging subsystem according to the matching result.

Further, the execution process of the contraband class configuration and the confidence threshold configuration in the security inspection modes with different levels is as follows:

SN, configure different grade security check modes, including:

the method comprises the following steps of configuring contraband categories and confidence thresholds in different levels of security inspection modes:

in a first-level security inspection mode: the method comprises the following steps of configuring contraband categories in an X-ray imaging graph and confidence coefficient threshold values corresponding to different contraband categories in the X-ray imaging graph, and under a secondary security inspection mode: the method comprises the following steps that contraband class configuration in an X-ray imaging graph and confidence coefficient threshold configuration corresponding to different contraband classes in the X-ray imaging graph are carried out in a three-level security inspection mode: the method comprises the following steps of configuring contraband categories in an X-ray imaging graph and configuring confidence threshold values corresponding to different contraband categories in the X-ray imaging graph, …, under an N-level security inspection mode: the method comprises the steps of configuring contraband categories in an X-ray imaging graph and configuring confidence coefficient threshold values corresponding to different contraband categories in the X-ray imaging graph.

Furthermore, the management method is applied to an image processing device of a subway security inspection centralized image judging system, wherein the subway security inspection centralized image judging system comprises an intelligent analyzer, an image processing device, a centralized image judging subsystem, a package opening terminal and a security inspection supervision platform;

the current image data is derived from an intelligent analyzer,

the contraband category configuration and the confidence coefficient threshold configuration are from a security inspection supervision platform or an image processing device;

and the security inspection supervision platform feeds back the contraband class configuration and the confidence coefficient threshold configuration to the image processing device.

Further, the different-level security check modes include a first-level security check mode, a second-level security check mode and a third-level security check mode:

the first-level security inspection mode directly pushes contraband class configuration of the unpacking terminal to comprise guns, control cutters and water cup liquid, confidence thresholds of the guns, the control cutters and the water cup liquid are 80% in a peak leveling period and 90% in a peak height period;

the secondary security inspection mode directly pushes the contraband class configuration of the unpacking terminal to comprise guns, control cutters and water cup liquid, the confidence threshold values of the guns and the control cutters are 80% in the peak-off period and 90% in the peak-up period, the confidence threshold values of the water cup liquid are 90% in the peak-off period and 99% in the peak-up period;

the third-level security inspection mode is characterized in that the contraband class configuration directly pushed to the package opening terminal comprises guns and control cutters, the confidence coefficient threshold values of the guns and the control cutters are 80% in the peak leveling period and 90% in the peak period, and water cups are not classified as the contraband class directly pushed to the package opening terminal.

Further, the different-level security check mode, when executing the first-level security check mode:

if the current image data contains contraband class marks and the current corresponding contraband class marks accord with the contraband classes directly pushed to the package opening terminal, and the confidence level of the current corresponding contraband is larger than or equal to the confidence level threshold value corresponding to the contraband, pushing the current image data to the package opening terminal; if the confidence level of the corresponding contraband is smaller than the confidence level threshold value corresponding to the contraband, pushing the current image data to the centralized image judgment subsystem;

if the current image data contains contraband class marks and the current corresponding contraband class marks do not accord with the contraband class of the direct-pushing package opening terminal, pushing the current image data to the centralized image judging subsystem;

and if no contraband class mark exists in the current image data, pushing the current image data to the centralized image judging subsystem.

Further, the different-level security check mode, when executing the secondary security check mode:

if the current image data contains contraband class marks and the current corresponding contraband class marks accord with the contraband classes directly pushed to the package opening terminal, and the confidence level of the current corresponding contraband is larger than or equal to the confidence level threshold value corresponding to the contraband, pushing the current image data to the package opening terminal; if the confidence level of the corresponding contraband is smaller than the confidence level threshold value corresponding to the contraband, pushing the current image data to the centralized image judgment subsystem;

if the current image data contains contraband class marks and the current corresponding contraband class marks do not accord with the contraband class of the direct-pushing package opening terminal, pushing the current image data to the centralized image judging subsystem;

and if no contraband class mark exists in the current image data, the current image data is not pushed to the centralized image judging subsystem.

Further, the different-level security inspection mode, when executing the three-level security inspection mode:

if the current image data contains contraband class marks and the current corresponding contraband class marks accord with the contraband classes directly pushed to the package opening terminal, and the confidence level of the current corresponding contraband is larger than or equal to the confidence level threshold value corresponding to the contraband, pushing the current image data to the package opening terminal; if the confidence level of the corresponding contraband is smaller than the confidence level threshold value corresponding to the contraband, pushing the current image data to the centralized image judgment subsystem;

if the current image data contains contraband class marks and the current corresponding contraband class marks do not accord with the contraband class of the direct-pushing package opening terminal, pushing the current image data to the centralized image judging subsystem;

and if no contraband class mark exists in the current image data, the current image data is not pushed to the centralized image judging subsystem.

Further, the security inspection hierarchical mode supports the configuration of the security inspection hierarchical mode of the whole network line, the designated station and the designated security inspection point.

Further, the configuration of the security inspection hierarchical mode supports manual configuration and automatic configuration, the effective condition of the automatic configuration is automatically adjusted according to the total passenger flow number of the security inspection door of the security inspection point, the manual configuration also comprises default configuration, the automatic configuration is effective mainly and automatically adjusted according to the total passenger flow number of the security inspection door of the security inspection point (hereinafter referred to as the passenger flow number of the security inspection door), and when the passenger flow number of the security inspection door is less than 30 persons/minute (namely 1800 persons/hour), the primary security inspection mode is started; when the passenger flow of the security inspection door is more than or equal to 30 persons/minute (1800 persons/hour), and the passenger flow of the security inspection door is less than 35 persons/minute (2100 persons/hour), starting a secondary security inspection mode; and when the passenger flow of the security inspection door is more than or equal to 35 persons/minute (namely 2100 persons/hour), starting a three-level security inspection mode.

Furthermore, alarm levels of different contraband categories can be configured, the security inspection grading mode configures the alarm levels of the different contraband categories, and the different alarm levels are displayed through alarm lamps with different colors.

The invention has the following beneficial effects: different levels of security inspection modes are configured according to the categories of the contraband and the confidence degrees corresponding to the contraband of different categories, the traffic efficiency and the riding safety are balanced under different times or different conditions, and the working efficiency is improved.

Drawings

FIG. 1 is a flow chart of the operation of the present invention;

FIG. 2 is a matching flow chart of the present invention.

Detailed Description

The present disclosure will be further explained below with reference to specific examples in order to facilitate understanding of the technical contents of the present disclosure by those skilled in the art.

The specific technical scheme is as follows: as shown in fig. 1 and 2, a method for managing a subway security inspection mode configuration includes: s1: acquiring current image data, wherein the current image data comprises a current X-ray imaging picture, a current corresponding contraband class mark of the current X-ray imaging picture and a current corresponding confidence coefficient of contraband; s2: and matching the current contraband class mark and the confidence corresponding to the current contraband in the current image data with the contraband class configuration and the confidence threshold configuration in the security inspection modes of different grades, and directly pushing the current image data to a package opening terminal or a centralized image judging subsystem according to the matching result.

The method also comprises the execution processes of contraband class configuration and confidence threshold configuration under different levels of security inspection modes, which are as follows: SN, configuring security inspection modes of different levels, configuring contraband categories and confidence thresholds in the security inspection modes of different levels as follows:

in a first-level security inspection mode: the method comprises the following steps of configuring contraband categories in an X-ray imaging graph and confidence coefficient threshold values corresponding to different contraband categories in the X-ray imaging graph, and under a secondary security inspection mode: the method comprises the following steps that contraband class configuration in an X-ray imaging graph and confidence coefficient threshold configuration corresponding to different contraband classes in the X-ray imaging graph are carried out in a three-level security inspection mode: the method comprises the following steps of configuring contraband categories in an X-ray imaging graph and configuring confidence threshold values corresponding to different contraband categories in the X-ray imaging graph, …, under an N-level security inspection mode: the method comprises the steps of configuring contraband categories in an X-ray imaging graph and configuring confidence coefficient threshold values corresponding to different contraband categories in the X-ray imaging graph.

The management method is applied to an image processing device of a subway security inspection centralized image judging system, wherein the subway security inspection centralized image judging system comprises an intelligent analyzer, an image processing device, a centralized image judging subsystem, a package opening terminal and a security inspection supervision platform; the current image data is from an intelligent analyzer, and the contraband class configuration and the confidence coefficient threshold configuration are from a security inspection supervision platform or an image processing device; and the security inspection supervision platform feeds back the contraband class configuration and the confidence coefficient threshold value configuration to the image processing device.

The different-grade security check modes comprise a first-grade security check mode, a second-grade security check mode and a third-grade security check mode:

the first-level security inspection mode directly pushes contraband class configuration of the unpacking terminal to comprise guns, control cutters and water cup liquid, confidence thresholds of the guns, the control cutters and the water cup liquid are 80% in a peak leveling period and 90% in a peak height period;

the secondary security inspection mode directly pushes the contraband class configuration of the unpacking terminal to comprise guns, control cutters and water cup liquid, the confidence threshold values of the guns and the control cutters are 80% in the peak-off period and 90% in the peak-up period, the confidence threshold values of the water cup liquid are 90% in the peak-off period and 99% in the peak-up period;

the third-level security inspection mode is characterized in that the contraband class configuration directly pushed to the package opening terminal comprises guns and control cutters, the confidence coefficient threshold values of the guns and the control cutters are 80% in the peak leveling period and 90% in the peak period, and water cups are not classified as the contraband class directly pushed to the package opening terminal.

The different-level security inspection mode is characterized in that when the first-level security inspection mode is executed:

if the current image data contains contraband class marks and the current corresponding contraband class marks accord with the contraband classes directly pushed to the package opening terminal, and the confidence level of the current corresponding contraband is larger than or equal to the confidence level threshold value corresponding to the contraband, pushing the current image data to the package opening terminal; if the confidence level of the corresponding contraband is smaller than the confidence level threshold value corresponding to the contraband, pushing the current image data to the centralized image judgment subsystem;

if the current image data contains contraband class marks and the current corresponding contraband class marks do not accord with the contraband class of the direct-pushing package opening terminal, pushing the current image data to the centralized image judging subsystem;

and if no contraband class mark exists in the current image data, pushing the current image data to the centralized image judging subsystem.

The different-level security inspection mode is characterized in that when the secondary security inspection mode is executed:

if the current image data contains contraband class marks and the current corresponding contraband class marks accord with the contraband classes directly pushed to the package opening terminal, and the confidence level of the current corresponding contraband is larger than or equal to the confidence level threshold value corresponding to the contraband, pushing the current image data to the package opening terminal; if the confidence level of the corresponding contraband is smaller than the confidence level threshold value corresponding to the contraband, pushing the current image data to the centralized image judgment subsystem;

if the current image data contains contraband class marks and the current corresponding contraband class marks do not accord with the contraband class of the direct-pushing package opening terminal, pushing the current image data to the centralized image judging subsystem;

and if no contraband class mark exists in the current image data, the current image data is not pushed to the centralized image judging subsystem.

The different-grade security inspection mode is characterized in that when a three-grade security inspection mode is executed:

if the current image data contains contraband class marks and the current corresponding contraband class marks accord with the contraband classes directly pushed to the package opening terminal, and the confidence level of the current corresponding contraband is larger than or equal to the confidence level threshold value corresponding to the contraband, pushing the current image data to the package opening terminal; if the confidence level of the corresponding contraband is smaller than the confidence level threshold value corresponding to the contraband, pushing the current image data to the centralized image judgment subsystem;

if the current image data contains contraband class marks and the current corresponding contraband class marks do not accord with the contraband class of the direct-pushing package opening terminal, pushing the current image data to the centralized image judging subsystem;

and if no contraband class mark exists in the current image data, the current image data is not pushed to the centralized image judging subsystem.

The security inspection hierarchical mode supports the configuration of a security inspection hierarchical mode of a whole network line, an appointed station and an appointed security inspection point.

The configuration of the security inspection grading mode supports manual configuration and automatic configuration, the effective condition of the automatic configuration is automatically adjusted according to the total passenger flow number of the security inspection door of a security inspection point, and the manual configuration further comprises default configuration.

The alarm levels of different contraband categories can be configured, the security inspection grading mode configures the alarm levels of different contraband categories, and the different alarm levels are displayed through alarm lamps with different colors.

Example (b): firstly, contraband existing in subway security inspection is classified, the specific classification condition is shown in the following table, and different alarm grades are displayed through alarm lamps with different colors, wherein a red alarm lamp is 1 grade, an orange alarm lamp is 2 grade, and a yellow alarm lamp is 3 grade.

Secondly, the security inspection hierarchical mode management method supports the security inspection hierarchical mode configuration of managing a whole network, an appointed line, an appointed station and an appointed security inspection point, can be configured according to the difference of the peak periods of different subway station points, divides the security inspection hierarchy into 3 hierarchical modes of a first-level security inspection mode, a second-level security inspection mode and a third-level security inspection mode, and defines three security inspection modes of different levels according to the following conditions: the primary security check mode mainly aims at ensuring the riding safety to be prior, the secondary security check mode mainly aims at the appropriate balance of the passing efficiency and the riding safety, the tertiary security check mode mainly aims at ensuring the passing efficiency of passengers to be prior, and for example, the tertiary security check mode is usually adopted in the rush hour to ensure the co-operation efficiency of the passengers. And finally, configuring the contents of the primary security inspection mode, the secondary security inspection mode and the tertiary security inspection mode, wherein the configuration details are as follows: the method supports the configuration of alarm levels for different contraband categories, supports the configuration of whether the packet terminal is directly pushed or not for different contraband categories, namely, the intelligent recognition graph judges that the contraband is directly pushed or not, supports the configuration of confidence coefficients of the packet terminal is directly pushed or not for different contraband categories, namely, the intelligent recognition graph judges that the contraband is, and the packet terminal is directly pushed only if the confidence coefficient is greater than or equal to a specified threshold value. The confidence degree of directly pushing the packet terminal at the peak time is configured for different contraband categories, namely, the threshold value of the confidence degree can be further increased in order to improve the passing efficiency. When the contraband does not meet the conditions of the direct-push package opening terminal, whether the package X-ray imaging graph is pushed or not is configured for different contraband categories (including no contraband) or not is supported, for example, when a first-level security inspection mode is configured, the intelligent analyzer does not recognize the contraband pushing artificial judgment graph, and when a second-level security inspection mode and a third-level security inspection mode are configured, the intelligent analyzer does not recognize the contraband pushing artificial judgment graph. The first-level security inspection mode, the second-level security inspection mode and the third-level security inspection mode support 2 modes of manual configuration validation and automatic configuration validation when configuration is in effect, automatic configuration validation is mainly carried out according to the total passenger flow number (hereinafter referred to as 'security inspection door passenger flow') of a security inspection door of a security inspection point, and when the passenger flow of the security inspection door is less than 30 persons/minute (namely 1800 persons/hour), the first-level security inspection mode is started; when the passenger flow of the security inspection door is more than or equal to 30 persons/minute (1800 persons/hour), and the passenger flow of the security inspection door is less than 35 persons/minute (2100 persons/hour), starting a secondary security inspection mode; and when the passenger flow of the security inspection door is more than or equal to 35 persons/minute (namely 2100 persons/hour), starting a three-level security inspection mode.

When the primary security check mode is executed:

the image processing device is set to be in a primary security inspection mode, the intelligent analyzer outputs current image data, the current image data comprise a current X-ray imaging picture, a current gun, control knife or water cup liquid (contraband) category mark corresponding to the current X-ray imaging picture, confidence coefficients of the current gun, control knife or water cup liquid (contraband) corresponding to the current X-ray imaging picture and the current gun, control knife or water cup liquid (contraband), the confidence coefficients of the current gun, control knife or water cup liquid (contraband) are configured in a confidence coefficient threshold value of more than or equal to 80% in a flat peak period, and when the peak period is configured in a confidence coefficient threshold value of more than or equal to 90%, the current image data are directly pushed to the unpacking terminal; when the confidence coefficient of the liquid (contraband) of the gun, the controlled knife or the water cup corresponding to the current image is smaller than the preset confidence coefficient threshold value configuration, pushing the current image data containing the liquid of the gun, the controlled knife or the water cup to a centralized judging image subsystem;

when the intelligent analyzer does not identify contraband, the current picture data is pushed to the centralized judging picture subsystem;

when the secondary security check mode is executed:

the gun and the control knife are configured to be required to directly push the contraband category of the unpacking terminal, the image processing device is set to be in a primary security inspection mode, the intelligent analyzer outputs current picture data, the current picture data comprise a current X-ray imaging picture, a current corresponding gun and control knife (contraband) category mark thereof and confidence degrees of the current corresponding gun and control knife (contraband), the confidence degrees of the current corresponding gun and control knife (contraband) are configured at a confidence degree threshold value which is more than or equal to 80% in a peak-off period, and when the peak-off period is more than or equal to 90% in the confidence degree threshold value configuration, the current picture data are directly pushed to the unpacking terminal; when the confidence coefficient of the current corresponding gun and controlled tool (contraband) is smaller than the preset confidence coefficient threshold value configuration, the current picture data containing the gun and the controlled tool is pushed to the centralized judging picture subsystem;

the cup liquid is also configured to be contraband needing to be pushed to the package opening terminal, but the confidence of the cup liquid in the current picture data needs to be configured at a confidence threshold value which is greater than or equal to 90% in a flat peak period, and when the confidence threshold value is greater than or equal to 99% in a high peak period, the current picture data containing the cup liquid is pushed to the package opening terminal; when the confidence of the water cup liquid in the current picture data is smaller than the preset confidence threshold configuration, pushing the current picture data containing the water cup liquid to a centralized image judging subsystem;

the intelligent analyzer does not identify contraband, and the current picture data does not need to be pushed to the centralized judging picture subsystem;

when the three-level security inspection mode is executed:

the gun and the control knife are configured to be required to directly push the contraband category of the unpacking terminal, the image processing device is set to be in a primary security inspection mode, the intelligent analyzer outputs current picture data, the current picture data comprise a current X-ray imaging picture, a current corresponding gun and control knife (contraband) category mark thereof and confidence degrees of the current corresponding gun and control knife (contraband), the confidence degrees of the current corresponding gun and control knife (contraband) are configured at a confidence degree threshold value which is more than or equal to 80% in a peak-off period, and when the peak-off period is more than or equal to 90% in the confidence degree threshold value configuration, the current picture data are directly pushed to the unpacking terminal; when the confidence coefficient of the current corresponding gun and controlled tool (contraband) is smaller than the preset confidence coefficient threshold value configuration, the current picture data containing the gun and the controlled tool is pushed to the centralized judging picture subsystem;

the cup liquid is not configured to be contraband needing to be pushed directly to the packet opening terminal, and the current picture data containing the cup liquid are all pushed to the centralized judging picture subsystem;

the intelligent analyzer does not identify contraband, and current picture data does not need to be pushed to the centralized judging picture subsystem.

The key data indexes of the primary security inspection mode, the secondary security inspection mode and the tertiary security inspection mode are compared as shown in the following table:

in addition, the primary security inspection mode, the secondary security inspection mode and the tertiary security inspection mode are also provided with default configurations, and the modified time is recorded after manual configuration adjustment and the functions of resetting to the default configurations are achieved.

The default initial values of the primary security inspection mode are as follows:

the default initial values of the secondary security inspection mode are as follows:

the default initial values of the three-level security inspection mode are as follows:

the principle and embodiments of the present invention are described in detail by using specific examples, which are only used to help understanding the core technical content of the present invention, and are not used to limit the protection scope of the present invention, and the technical solution of the present invention is not limited to the specific embodiments described above. Based on the above embodiments of the present invention, those skilled in the art should make any improvements and modifications to the present invention without departing from the principle of the present invention, and therefore, the present invention should fall into the protection scope of the present invention.

16页详细技术资料下载
上一篇:一种医用注射器针头装配设备
下一篇:一种激光设备箱

网友询问留言

已有0条留言

还没有人留言评论。精彩留言会获得点赞!

精彩留言,会给你点赞!