KR101884575B1 - Scanning system for security - Google Patents

Abstract

A security search system comprises: a security search apparatus; article database storing hazard article characteristic data; and an article information server providing the hazard article characteristic data to a security terminal through a communication network. The security search apparatus includes: a security search device which scans at least one targeted object to obtain transmitted image data; and the security terminal for extracting characteristic data of the targeted object from the transmitted image data and comparing the characteristic data of the targeted object with the pre-stored hazard article characteristic data to determine an article to be inspected.

Description

{SCANNING SYSTEM FOR SECURITY}

The following description relates to a secure retrieval system.

Generally, in an airport, a port or various meeting places, a metal detector is installed so as to detect metallic materials such as firearms and prohibit the bringing in of dangerous materials. The metal detector is characterized in that the generated magnetic flux lines are changed depending on the presence or absence of a metallic material To detect firearms or other metallic dangerous materials.

The metal detector is widely used as a portable metal detector to detect the presence of metal by passengers or security personnel and to detect the metal possessed by the passengers through the gate structure It is true.

However, since the above-described metal detector alone can not fundamentally detect plastic bombs, which are not metallic ones, it is difficult to prepare for a sudden change in terrorism. To this end, a large number of officers or security officers must search baggage individually There was a problem.

In order to solve this problem, the search system using X-RAY which can read real-time about the baggage passing through the search bag has been introduced, so that the search time of baggage can be reduced. X-RAY search equipment installed at airports and ports and searching illegal items included in baggage can be read in real time, and search equipment having recording function is also introduced and used.

However, this X-RAY scanner scans consecutively moving luggage at a speed of more than 12m per minute and identifies the images to be visualized by security personnel. Thus, as the workload of the security detection personnel increases, the retrieval read rate may be lowered. In addition, according to the skill level of the security search personnel, there may be a difference in the reading ability of the articles. In addition, the X-RAY scanner requires a long time to retrieve accurate information from security scanners when the goods are brought in by manual confirmation of the scanned image, and it takes a long time to search, the congestion in the airport office is increased, there was.

The object of the embodiment is to provide a security search apparatus which automatically performs a security search operation through an article database when a goods are brought into a security search apparatus to increase the retrieval read rate to prevent security search failure in advance, And to reduce the time required for the search.

It is also an object of the present invention to provide a security retrieval system that can automatically read articles passing through a security retrieval apparatus by utilizing pattern recognition, machine learning, and three-dimensional modeling technology.

It is another object of the present invention to provide a security retrieval system for constructing an integrated database of images of articles passing through a security retrieval apparatus disposed at a plurality of locations, thereby enhancing product retrieval efficiency of each security retrieval apparatus.

It is another object of the present invention to provide a security search system capable of generating an alarm upon retrieving a forgery product, thereby reducing human error according to the proficiency level of the security search agent.

A security search system according to an embodiment will be described.

The security search system includes a security search unit that scans at least one object to obtain transmitted image data, and a feature extraction unit that extracts feature data of the object from the transmitted image data and compares feature data of the object with pre- A security search apparatus including a security terminal for identifying an article, an article database storing the risk article characteristic data, and an article information server providing the risk article characteristic data to the secure terminal through a communication network.

According to one aspect of the present invention, the plurality of security search devices are provided, and at least one of the plurality of security search devices is provided at a plurality of different positions, and the plurality of security search devices interoperate with the goods information server to share the goods characteristic data .

According to one aspect of the present invention, the secure terminal includes a receiving unit that receives the transmitted image data, a preprocessor that preprocesses the transmitted image data to generate two-dimensional shape data, a three-dimensional modeling unit that generates three- Dimensional shape data and three-dimensional modeling data, a feature extraction unit for extracting features necessary for distinguishing a risky item from the risky product feature data of the object including the two-dimensional shape data and the three-dimensional modeling data, A pattern recognition unit for recognizing the pattern with respect to the feature extracted by the pattern recognition unit, and a determination unit for determining the damage item using the result recognized by the pattern recognition unit.

According to one aspect of the present invention, the pre-processing unit may perform at least one of pre-processing such as binarization, sessionization, and noise removal on the transmission image data.

According to an aspect of the present invention, the secure terminal may further include a color attaching unit that adds color to the transmitted image data by discriminating a substance of each object.

According to one aspect, the color adder may add a different color to each of the organic, inorganic, mixture and undetectable items.

According to one aspect of the present invention, the secure terminal includes a matching unit for matching the item characteristic data including the type of the determined risk item to the transmitted image data, and a matching unit for displaying the transmitted image data of the matched object And may further include a display unit.

According to one aspect, when the transmitted image data in which the hazardous goods characteristic data is matched is displayed, the display unit may further display an alarm in an audible or visual form.

According to one aspect of the present invention, an image data base interlocked with the article information server and storing an original of the transmitted image data, and an image data base interlocked with the article information server, And a statistical database for storing the article generation data including at least one of country information, waypoint information, owner information, and statistical data.

According to an aspect of the invention, the product information server may further include an information classifying unit for classifying all data transmitted from the secure terminal into a predetermined category, and a statistic unit for generating the statistical data from the classified information.

According to an aspect of the present invention, the article information server may include a server communication unit for connecting to the communication network through a wired / wireless interface, and a service providing unit for providing a goods information management service through the communication network.

According to the embodiment, when the goods are brought into the security search device, the security search device automatically performs the security search operation through the article database to increase the search read rate to prevent security search failure in advance, It is possible to reduce the time required for the search.

In addition, pattern recognition, machine learning, and three-dimensional modeling techniques can be utilized to automatically read articles passing through the security search device.

In addition, an integrated database of images of articles passing through the security search apparatuses arranged at a plurality of positions can be constructed, thereby enhancing the efficiency of retrieving articles of each security search apparatus.

In addition, an alarm may be generated when a product is searched for, thereby reducing human error caused by the skill of the security searcher.

FIG. 1 schematically shows a security search system according to an embodiment.
2 is a block diagram of a security search system according to an embodiment.
3 is a diagram illustrating an exemplary screen of a secure terminal on which a risky product according to an exemplary embodiment is displayed.

Hereinafter, embodiments will be described in detail with reference to exemplary drawings. It should be noted that, in adding reference numerals to the constituent elements of the drawings, the same constituent elements are denoted by the same reference symbols as possible even if they are shown in different drawings. In the following description of the embodiments, detailed description of known functions and configurations incorporated herein will be omitted when it may make the best of an understanding clear.

In describing the components of the embodiment, terms such as first, second, A, B, (a), and (b) may be used. These terms are intended to distinguish the constituent elements from other constituent elements, and the terms do not limit the nature, order or order of the constituent elements. When a component is described as being "connected", "coupled", or "connected" to another component, the component may be directly connected or connected to the other component, Quot; may be "connected," "coupled," or "connected. &Quot;

The method according to an embodiment may be implemented in the form of a program command that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions to be recorded on the medium may be those specially designed and configured for the embodiments, or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

FIG. 1 schematically shows a security search system according to an embodiment.

Referring to FIG. 1, the security search system 1 is used to prevent entry of goods in a specific place such as an airport, a port, a customs house, a conference hall, and the like. For example, the security retrieval system 1 is deployed at an airport and used to prevent entry of a prohibited article, i.e., a dangerous article, into the aircraft. The security search system (1) emits X-rays to the baggage to generate a transmission image, and discriminates each object on the transmission image through comparison analysis with the article characteristic data of the goods information server.

The security search system 1 includes a security search apparatus 10, an article information server 20, an article data base 30, a statistical data base 40 and an image data base 50.

A plurality of security search apparatuses 10 are provided and at least one or more security search apparatuses 10 are provided at a plurality of different positions. The plurality of security search apparatuses 10 may be connected to the goods information server 20 via a communication network and receive the goods characteristic data stored in the goods database 30. [ The security search apparatus 10 scans a target object to obtain transmitted image data, extracts feature data of the target object from the transmitted image data, and compares the received feature data with the received risk item characteristic data to determine an item to be analyzed.

Here, the communication network may be a data communication network including LAN, WAN, UWB, WiFi, 3G, 4G, and 5G, or may be a separate network for security purposes.

The security search system 1 can interoperate with the security search apparatus 10 in which the article information server 20 is provided at a plurality of locations and can share information on the determined harmful articles in each security search apparatus 10 It is possible to facilitate the product discrimination in each security search device 10. In addition, the security search system 1 can effectively reduce errors in product identification in order to lower the reliance of the security search staff, by determining the goods for the security search apparatus 10. [ Here, the damage item characteristic data may include material data of the hazardous material, two-dimensional shape data, three-dimensional shape data, kind, and pattern data.

2 is a block diagram of a security search system according to an embodiment.

Referring to FIG. 2, the security search apparatus 10 includes a security search station 11 and a secure terminal 12.

The security search band 11 scans at least one object to acquire a transmission image. For example, the security searching unit 11 can radiate X-rays to a target object, that is, scan an article to obtain a transmitted image. The security search base 11 moves the object inside at a speed of 12 m per minute, and scans the moving object inside to acquire the respective transmitted images. The security search base 11 can acquire the transmitted images in the vertical and horizontal directions of the object when scanning the object.

In the embodiment, the security search band 11 radiates an X-ray to scan a target object. However, the present invention is not limited to this, and the security search band 11 may transmit a transmitted image of an article such as a method using a muon, Any method that can be obtained may be possible.

The security terminal 12 extracts the feature data of the object from the transmitted image data, and compares the feature data of the object with the pre-stored risk item feature data to determine an article to be inspected. The security terminal 12 can communicate the characteristic data of the object with the article information server 20 and store it in the article data base 30. The security terminal 12 matches the characteristic data of the determined harmful article with the transmission image data, Or in the form of a moving picture. For example, the secure terminal 12 may be a kind of computing device including a PC, a laptop, a smart phone, a tablet PC, and the like.

The security terminal 12 includes a receiving unit 121, a communication unit 122, a preprocessor 123, a color attaching unit 124, a three-dimensional modeling unit 125, a feature extracting unit 126, a pattern recognizing unit 127, A determination unit 128, a matching unit 129, and a display unit 130. [

The receiving unit 121 receives the transmitted image data. For example, the receiving unit 121 is connected to the security search unit 11 through a wired or wireless data communication method to receive transmitted image data. The receiving unit 121 receives the transmission image data in real time at the same time.

The communication unit 122 connects with the article information server 20, which will be described later, through a communication network. For example, the communication unit 122 may be connected to the commodity information server 20 through a wired or wireless interface to the communication network.

The color adder 124 adds color to the material of each object to the transmitted image data. For example, the color attaching unit 124 extracts each image that varies depending on the density characteristic with respect to the object on the transmission image data, and adds different colors to the organic substance, the inorganic substance, the mixture of the organic substance and the inorganic substance, do. The color addition section 124 can generate the material data of the object by adding the color to the object.

The preprocessing unit 123 preprocesses the received transmission image data. The preprocessing unit 123 performs one or more preprocessing operations on the received transmission image, such as binarization, sessionization, and noise removal. The preprocessing unit 123 can generate two-dimensional shape data for a specific object from the preprocessed transmission image data.

The three-dimensional modeling unit 125 may generate three-dimensional modeling data from the transmitted image data. For example, the three-dimensional modeling unit 125 can generate three-dimensional modeling data of each object through the structural analysis by mapping vertical and horizontal shapes based on vertical and horizontal transmission image data corresponding to a specific object in the transmission image data have. In addition, the three-dimensional modeling unit 125 may generate the three-dimensional modeling data through the preprocessed two-dimensional shape data.

The feature extraction unit 126 extracts features necessary for product identification in the risk item data of the object including the two-dimensional shape data and the three-dimensional modeling data. For example, the feature extraction unit 126 extracts features from the vertical and horizontal shapes of the shape of the object in the two-dimensional shape data. The feature extraction unit 126 extracts features of the three-dimensional shape from the three-dimensional modeling data.

The security search system 1 may be able to distinguish the dangerous goods through only the two-dimensional shape data and the three-dimensional modeling data. However, the present invention is not limited thereto, and the feature extraction unit 126 may be able to identify a target object by further extracting a characteristic of a substance of the target object from the material data of the target object.

The pattern recognition unit 127 mechanically learns the pattern of the damage object characteristic data of the object and recognizes the pattern with respect to the feature extracted by the feature extraction unit 126. [ For example, the pattern recognition unit 127 may classify the pattern of the object goods characteristic data of the object into generalized linear models (GLM), decision trees, random forests, and gradient boosting machines A machine learning method is performed by a method of at least one of a map learning method including a gradient boosting machine (GBM) or an autonomous learning method such as clustering and anomaly detection.

Here, machine learning techniques have already been studied extensively and various publicly available software can be used. Therefore, detailed description of the machine learning technique used in the pattern recognition unit 127 will be omitted.

The pattern recognition unit 127 stores the learned pattern data in the article database 30. The pattern recognition unit 127 calculates a degree of importance of a pattern of a feature to be learned and creates a tree structure for the attribute to store the pattern data. The pattern recognizing unit 127 recognizes the pattern of the new object by the tree structure of the learned pattern data previously stored in the article database 30, when the item characteristic data of the new object is input.

The discrimination unit 128 discriminates a hazardous article using the result recognized by the pattern recognition unit. For example, the determination unit 128 determines the type of the article by comparing the pattern data of the article stored in the risky article database 30 with the pattern of the article recognized by the pattern recognition unit, Is determined.

The matching unit 129 matches the discriminated risk item characteristic data with the transmission image data. For example, the matching unit 129 matches the type and shape pattern of the article identified by the determination unit with the transmission image data. The matching unit 129 may add color to the transmitted image data by matching the material data of the object, and may match any one or more of the two-dimensional shape data, the three-dimensional modeling data, or the type of the article.

In addition, the matching unit 129 stores the determined harmful article characteristic data in the article data base 30 in cooperation with the article information server 20. [ In addition, the matching unit 129 matches and stores pre-stored risk item characteristic data with respect to the risk item characteristic data determined as the same article.

The matching unit 129 matches the risk item generation data including at least one of the type of the item, the scan time, the scan position, the importing country, the stopover, and the bearer for the identified risk item characteristic data, Lt; / RTI > The matching unit 129 may also be capable of storing the original image data of the transmission image data in the image database 50 through the communication unit 122 in correspondence with the article characteristic data.

By matching the object identified as the article for the article characteristic data, the article occurrence data and the transmitted image data for the matching section 129 with the same or similar articles for the purpose of storing, 40, and 50 may be provided to increase the efficiency of the security search system.

The display unit 130 displays a transmitted image in which the hazardous article characteristic data is matched. For example, the display unit 130 may include a display panel. The display unit 130 displays the dangerous goods information on the transmitted image of the target object matched with the hazardous material, and displays an alarm. The display unit 130 displays the dangerous goods information and displays the alarm so that the security detection agent can recognize the dangerous goods information. It is also possible to provide a separate flashing light, a siren or a speaker in the security search box 11 to facilitate recognition of the security search staff, and the display unit 130 may display an alarm through a flashing light, a siren or a speaker .

3 is a diagram illustrating an exemplary screen of a secure terminal on which a risky product according to an exemplary embodiment is displayed.

Referring to FIG. 3, the display unit 130 displays a border for the transmitted image data to which each color of the material is added, and displays the highlighted image. The shape of the frame is not limited, and can be highlighted and displayed such as a rectangle, a circle, an ellipse, and the like.

The display unit 130 can enlarge or reduce the displayed transmitted image data, and display the hazardous goods information for each object. For example, the display unit can display the hazardous goods information in the form of a window. In addition, the display unit 130 may generate an event for storing the article characteristic data in the article database 30 via the article information server with respect to the object whose risk information is not matched.

The security terminal 12 can identify the target object by comparing the similarity with the stored risk item information by extracting the characteristic of the target object through pattern recognition. The security terminal 12 can easily identify the hazardous object of the hazardous object by matching the dangerous goods information with the object through the machine learning technique and accumulating information on the dangerous goods information and the similarity determination.

2, the article information server 20 interlocks the security search apparatus 10 with the article data base 30, the statistical data base 40, and the image data base 50, Generation data, and transmission image original data. For example, the goods information server 20 may provide the security terminal 12 with the goods characteristic information stored in the goods item database 30.

In the embodiment, the article information server 20 is illustrated as being linked with the article data base 30, the statistical data base 40 and the image data base 50. However, But it is also possible to construct each server corresponding to each of the databases 30, 40 and 50 so that the respective servers cooperate with each other. It may also be possible to partially build a server of some databases.

The article information server 20 may include a server communication unit 21, a service providing unit 22, an information classification unit 23, and a statistics unit 24.

The server communication unit 21 is connected to the security terminal 12 by connecting to the communication network through a wired or wireless interface. For example, the server communication unit 21 can connect to the communication network through the communication method corresponding to the communication unit 122. [

The service providing unit 22 provides an article information management service through a communication network. The service providing unit 22 performs periodic or aperiodic security program / version update for secure communication with the secure terminal 12. The service providing unit 22 may provide the security terminal 12 with the article characteristic data for the security terminal 12 in cooperation with the article database.

In addition, the service providing unit 22 may provide the display unit 130 with a hazardous goods registration event for the new hazardous goods. The service providing unit 22 may provide the display unit 130 with an item registration event in the form of a window or an event page.

The information classification unit 23 classifies the item characteristic data, the item generation data, and the transmission image data transmitted from each security terminal. The information classification section 23 stores the classified risk item characteristic data, the risk item generation data, and the transmission image in the respective databases 30, 40 and 50. For example, the information classification unit 23 may store, in the article database 30, the item property data of the item including the item data of the item to be damaged, the two-dimensional shape data, the three-dimensional shape data, the type, and the learned pattern data. In addition, the information classification unit 23 stores in the statistical database 40 the hazardous article occurrence data including the hazardous article occurrence time, the risky article occurrence position, the type of the generated article, the carry-in country information, the stopover position information, and the holder information . In addition, the information classification unit 23 may store the transmitted image data in the image database 50.

The security retrieval system 1 continuously accumulates the item characteristic data, the item generation data and the transmission image data transmitted through the security terminal 12 in the respective databases 30, 40 and 50, Or when a similar object is retrieved in the security retrieval zone 11, it is possible to identify the item to be disposed through the accumulated risk item characteristic data.

The statistics section 24 can generate statistical data from the classified data. In other words, the statistic unit 24 stores statistics on the basis of the item occurrence time stored in the statistical database 40, the risky article occurrence location, the type of the generated goods, the import country information, the stopover information and the holder information The statistical data can be stored in the statistical database 40. [0050] FIG.

Also, the statistic unit 24 can generate statistical data of the time at which the specific risk item was scanned. The statistical unit 24 stores the generated statistical information in the statistical database 40. Based on the statistical data of the time when the specific risk item occurs most frequently, Therefore, it is possible to improve the discrimination speed of the defective article of the security search system 1 by providing it to the service providing unit 22 so that the dangerous goods characteristic data of the specific hazardous goods can be discriminated first. In addition, the statistics unit 24 can determine the priority of the hazardous article characteristic data of the specific hazardous product according to the hazardous article generating position, the bringing-in country information, the waypoint information, and the holder information, The discrimination speed can be improved.

The goods database 30 stores the hazardous article characteristic data, the statistical database 40 stores the hazardous article generation data, and the image database 50 stores the transmitted image data. The product database 30, the statistical database 40 and the image database 50 may be a RAM, a ROM, an HDD, an SSD, a MICRO SSD, a USB storage device, a CD, a DVD, a floppy disk, and the like.

According to the embodiment, when the goods are brought into the security search device, the security search device automatically performs the security search operation through the article database to increase the search read rate to prevent security search failure in advance, It is possible to reduce the time required for the search.

In addition, pattern recognition, machine learning, and three-dimensional modeling techniques can be utilized to automatically read articles passing through the security search device.

In addition, an integrated database of images of articles passing through the security search apparatuses arranged at a plurality of positions can be constructed, thereby enhancing the efficiency of retrieving articles of each security search apparatus.

In addition, an alarm may be generated when a product is searched for, thereby reducing human error caused by the skill of the security searcher.

Although the preferred embodiments of the present invention have been disclosed for illustrative purposes, those skilled in the art will appreciate that various modifications, additions and substitutions are possible, without departing from the scope and spirit of the invention as disclosed in the accompanying claims. For example, it is contemplated that the techniques described may be performed in a different order than the described methods, and / or that components of the described structures, devices, and the like may be combined or combined in other ways than the described methods, Appropriate results can be achieved even if they are replaced or replaced.

1: security search system 10: security search apparatus
11: security search base 12: security terminal
121: Receiver 122:
123: preprocessing unit 124:
125: three-dimensional modeling unit 126: feature extraction unit
127: pattern recognition unit 128:
129: matching unit 130: display unit
20: Goods information server 21: Server communication part
22: service providing unit 23: information classification unit
24: Statistical section 30: Commodity database
40: statistical database 50: image database

Claims (11)

A security searching unit that scans at least one object to acquire transmitted image data; and a determination unit that extracts feature data of the object from the transmitted image data and compares the characteristic data of the object with pre- A security search apparatus including a secure terminal;
An article database storing the risk item characteristic data; And
An article information server for providing the risk item characteristic data to the secure terminal through a communication network;
Lt; / RTI >
The secure terminal includes:
A receiving unit for receiving the transmitted image data;
A preprocessing unit for performing preprocessing on the transmission image data to generate two-dimensional shape data;
A three-dimensional modeling unit for generating three-dimensional modeling data from the transmitted image data;
A feature extracting unit for extracting features necessary for distinguishing the weighed item from the weighed item characteristic data of the object including the two-dimensional shape data and the three-dimensional modeling data;
A pattern recognition unit for mechanically learning a pattern of the hazardous article characteristic data of the object and recognizing a pattern of the characteristic extracted by the characteristic extraction unit; And
A discrimination unit for discriminating a hazardous article using the result recognized by the pattern recognition unit;
The security search system comprising:
The method according to claim 1,
The security search apparatuses are provided in plural, and at least one or more security search apparatuses are provided at a plurality of different positions,
Wherein the plurality of security search apparatuses can share the dangerous goods characteristic data in cooperation with the goods information server.
delete The method according to claim 1,
The pre-
Processing the at least one of the binary image, the session image, and the noise image to the transmission image data.
The method according to claim 1,
The secure terminal includes:
A color adding unit for adding color to the transmitted image data by discriminating the substance of each object;
Further comprising:
6. The method of claim 5,
The color-
A security search system that adds different colors to organic, inorganic, mixtures and undetectable items.
The method according to claim 1,
The secure terminal includes:
A matching unit for matching the item characteristic data including the type of the determined risk item to the transmitted image data; And
A display unit for displaying the transmitted image data of the object matched with the weighed item characteristic data;
Further comprising:
8. The method of claim 7,
The display unit includes:
And further displaying warnings in an audible or visual form when the transmitted image data in which the hazardous goods characteristic data is matched is displayed.
The method according to claim 1,
An image data base interlocked with the article information server and storing an original of the transmitted image data; And
And stores the generated item occurrence data including at least one of a dangerous article occurrence time, a risky article occurrence position, a type of the generated article, a carry-in country information, a waypoint information, a holder information, Statistical database;
Further comprising:
10. The method of claim 9,
The article information server comprises:
An information classifying unit classifying all data transmitted from the secure terminal into a predetermined category; And
A statistic unit for generating the statistical data from the classified information;
Further comprising:
The method according to claim 1,
The article information server comprises:
A server communication unit for connecting to the communication network through a wired / wireless interface; And
A service providing unit for providing a hazardous goods information management service through the communication network;
The security search system comprising:

Images