A security inspection method and system

Abstract

This application provides a security inspection method and system, relating to the field of security inspection technology. The method includes: determining a first detection result for the target person based on a millimeter-wave scan image of the target person passing through a millimeter-wave security scanner; determining a target security zone corresponding to the first location information within the detection area of ​​the metal detector based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, and the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector; acquiring a second detection result corresponding to a detected item within the target security zone; correcting the first category information of the detected item using the second detection result to determine the target category information of the detected item; and determining whether the detected item is a contraband based on the target category information. This improves the accuracy of the millimeter-wave security scanner in detecting contraband.

Description

Technical Field

[0001] This application relates to the field of security inspection technology, and in particular to a security inspection method and system. Background Technology

[0002] In crowded public places and other locations with high security requirements, security checks are typically conducted on individuals entering these areas to ensure public safety, specifically to inspect them for prohibited items, particularly those made of metal. Millimeter-wave security scanners are one such technology used for these checks.

[0003] Millimeter-wave security scanners use millimeter waves to scan the human body surface to generate images that reveal items hidden under clothing. However, due to factors such as array configuration and target body shape, millimeter-wave security scanners can experience blind spots and incomplete target display. Furthermore, millimeter waves cannot penetrate the skin, making it difficult to detect very small contraband items or those concealed in areas like the mouth, resulting in insufficient accuracy in detecting contraband. Summary of the Invention

[0004] The purpose of this application is to provide a security inspection method and system to improve the accuracy of millimeter-wave security scanners in detecting contraband. The specific technical solution is as follows:

[0005] In a first aspect, embodiments of this application provide a security inspection method, the method comprising:

[0006] Based on the millimeter-wave scan image of the target person passing through the millimeter-wave security scanner, a first detection result of the target person is determined, wherein the first detection result includes the first location information of the detected items carried by the target person on the millimeter-wave scan image and the first category information of the detected items;

[0007] Based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, and the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, the target defense zone corresponding to the first position information in the detection area of ​​the metal detector gate is determined, wherein the detection area of ​​the metal detector gate can be divided into multiple defense zones;

[0008] Obtain a second detection result corresponding to the detected item in the target defense zone, wherein the second detection result includes at least one of the metal strength information corresponding to the detected item in the target defense zone and the second category information of the detected item;

[0009] The first category information of the detected items is corrected using the second detection result to determine the target category information of the detected items;

[0010] The detection of a contraband item is determined based on the target category information.

[0011] Optionally, the step of determining the target security zone corresponding to the first location information in the detection area of ​​the metal detector gate based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, and the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, includes:

[0012] Based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, the second location information corresponding to the first location information in the detection area of ​​the millimeter-wave security scanner is determined;

[0013] Based on the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, the third positional information corresponding to the second positional information in the detection area of ​​the metal detector gate is determined.

[0014] Based on the position of the third location information in the detection area of ​​the metal detector gate, the target defense zone corresponding to the first location information in the detection area of ​​the metal detector gate is determined;

[0015] or

[0016] The step of determining the target security zone corresponding to the first location information in the detection area of ​​the metal detector gate based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, and the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, includes:

[0017] Based on the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, and the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, the fourth position information corresponding to the preset position point in the detection area of ​​the metal detector gate in the millimeter-wave scan image is determined.

[0018] In the millimeter-wave scan image, based on the fourth position information, an image region that matches the detection area of ​​the metal detector gate is determined;

[0019] Based on the position of the first location information in the image area, the target defense zone corresponding to the first location information in the detection area of ​​the metal detector gate is determined.

[0020] Optionally, the method further includes:

[0021] Acquire human body detection images of the target person as they pass through the metal detection gate, captured by the image acquisition device;

[0022] The matching relationship between the human body detection image and the millimeter-wave scan image is determined based on the similarity between the body part features included in the human body detection image and the body part features included in the millimeter-wave scan image.

[0023] Based on the positional relationship between the image acquisition device and the metal detection gate, the correspondence between the human body detection image and the detection area of ​​the metal detection gate is determined;

[0024] Based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, the correspondence between the human body detection image and the detection area of ​​the metal detector gate, and the matching relationship, the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate is determined.

[0025] Optionally, the step of correcting the first category information of the detected item using the second detection result to determine the target category information of the detected item includes:

[0026] Based on the second category information of the detected items in the second test result, the material attention coefficient of the items is determined;

[0027] The first category information of the detected items is corrected using the material attention coefficient of the items to determine the target category information of the detected items;

[0028] or

[0029] A second adjustment parameter corresponding to the target defense zone is determined to adjust the second detection result, and the first category information of the detected item is corrected using the adjusted second detection result to determine the target category information of the detected item;

[0030] or

[0031] A first adjustment parameter corresponding to the first category information of the detected item is determined to adjust the first category information of the detected item, and the adjusted first category information is further corrected using the second detection result to determine the target category information of the detected item.

[0032] or

[0033] Determine the first adjustment parameter corresponding to the first category information of the detected item, and determine the second adjustment parameter corresponding to the target defense zone;

[0034] The first adjustment parameter is used to adjust the first category information of the detected items, and the second adjustment parameter is used to adjust the second detection result;

[0035] The adjusted first category information is further corrected using the adjusted second detection result to determine the target category information of the detected items.

[0036] Optionally, the second adjustment parameter includes the sensitivity corresponding to the target defense zone, and the step of determining the second adjustment parameter corresponding to the target defense zone includes:

[0037] Set the sensitivity corresponding to the target defense zone to the first sensitivity.

[0038] The method further includes:

[0039] Set the sensitivity corresponding to the non-target defense zone to the second sensitivity, wherein the first sensitivity is greater than the second sensitivity.

[0040] Optionally, the first adjustment parameter includes an item material attention coefficient and / or a category adjustment parameter, and the step of determining the first adjustment parameter corresponding to the first category information of the detected item includes:

[0041] Based on the second category information of the detected items in the second detection result and / or the item category information included in the pre-set millimeter wave whitelist, the category adjustment parameter is determined, wherein the category adjustment parameter represents the degree of attention paid to item category in the current security inspection scenario;

[0042] And / or,

[0043] Based on the second category information of the detected items in the second detection result, the material attention coefficient of the items is determined, wherein the material attention coefficient of the items represents the degree of attention to metal items of different materials.

[0044] Optionally, the first adjustment parameter includes an item material attention coefficient and a category adjustment parameter, the second adjustment parameter includes the sensitivity corresponding to the target defense zone, the first category information includes the first confidence level that the detected item belongs to the item category, and the second detection result includes the metal strength value corresponding to the detected item in the target defense zone.

[0045] The steps of determining a first adjustment parameter corresponding to the first category information of the detected item, and determining a second adjustment parameter corresponding to the target defense zone; adjusting the first category information of the detected item using the first adjustment parameter, and adjusting the second detection result using the second adjustment parameter; and further correcting the adjusted first category information using the adjusted second detection result to determine the target category information of the detected item include:

[0046] The overall confidence level P of the detected item belonging to the item category is calculated using the following formula.fin (c) , as target category information for the detected items:

[0047] P tin (c)=M0×α(c)×P mmw (c)+β(m)×A metal (m);

[0048] Where c is the item category number, α(c) is the category adjustment parameter corresponding to item category c, and P mmw (c) represents the first confidence level that the detected item belongs to item category c, M0 is the material attention coefficient corresponding to the detected item, m is the target zone number corresponding to the detection location of the detected item, β(m) is the sensitivity corresponding to target zone m, and A metal (m) represents the metal strength value corresponding to the target defense zone m.

[0049] Optionally, the step of obtaining the second detection result corresponding to the detected item in the target defense zone includes:

[0050] If the first detection result indicates that the detected item belongs to a suspicious category, or if the security level requirement of the current security inspection scenario is the target level and the first detection result indicates that the detected item belongs to a safe category, then obtain the second detection result corresponding to the detected item in the target security zone.

[0051] Secondly, this application provides a security inspection system, which includes: a millimeter-wave security scanner, a metal detection gate, and a processing device, wherein: the millimeter-wave security scanner and the metal detection gate are used to detect target personnel, the millimeter-wave security scanner and the metal detection gate are integrated into a single physical device, or the millimeter-wave security scanner and the metal detection gate are two independent physical devices;

[0052] The processing device is used to perform any of the methods described in the first aspect.

[0053] Optionally, the millimeter-wave security scanner and the metal detector gate are integrated into a single physical device;

[0054] The processing device is further configured to, when the first detection result meets preset acquisition conditions, control the metal detection gate to start working, so as to acquire a second detection result in the target zone corresponding to the detected item; or,

[0055] The processing device is also used to control the metal detection gate to work continuously, and to acquire a second detection result corresponding to the detected item in the target zone when the first detection result meets the preset acquisition conditions;

[0056] The preset acquisition conditions are used to define the conditions for triggering the acquisition of the second detection result.

[0057] Optionally, the millimeter-wave security scanner and the metal detector gate are two independent physical devices, and the system also includes a first access gate;

[0058] The processing device is further configured to, when the first detection result does not meet the preset acquisition conditions and the first detection result indicates that the detected item is not a dangerous category, control the first channel gate to allow the target person passing through the millimeter-wave security scanner to pass through; and when the first detection result meets the preset acquisition conditions, control the first channel gate to allow the target person passing through the millimeter-wave security scanner to enter the detection area of ​​the metal detection gate.

[0059] The preset acquisition conditions are used to define the conditions for triggering the acquisition of the second detection result.

[0060] Optionally, the system further includes a second channel gate;

[0061] The processing device is further configured to, if it is determined that the target person is carrying prohibited items, control the second gate to allow the target person passing through the metal detector gate to enter the manual security check area; or, if it is determined that the target person is not carrying prohibited items, control the second gate to allow the target person passing through the metal detector gate to pass.

[0062] Thirdly, embodiments of this application provide a processing apparatus, including:

[0063] Memory, used to store computer programs;

[0064] A processor, when executing a program stored in memory, implements any of the methods described above.

[0065] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements any of the methods described above.

[0066] Fifthly, embodiments of this application provide a computer program product containing instructions that, when run on a computer, cause the computer to perform any of the methods described above.

[0067] Beneficial effects of the embodiments in this application:

[0068] In the technical solution provided in this application embodiment, the processing device can determine a first detection result of the target person based on the millimeter-wave scan image of the target person passing through the millimeter-wave security scanner. The first detection result includes the first position information of the detected item carried by the target person on the millimeter-wave scan image and the first category information of the detected item. Based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, and the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, the device determines the target security zone corresponding to the first position information in the detection area of ​​the metal detector gate. The detection area of ​​the metal detector gate can be divided into multiple security zones. The device obtains a second detection result corresponding to the detected item in the target security zone. The second detection result includes at least one of the metal strength information and the second category information of the detected item in the target security zone. The first category information of the detected item is corrected using the second detection result to determine the target category information of the detected item. The device then determines whether the detected item is a contraband based on the target category information.

[0069] Because any metallic object will interfere with the magnetic field between metal detector gates when passing through them, metal detector gates can detect metallic objects more accurately than millimeter-wave security scanners. The second detection result obtained by using the metal detector gate to detect the object corrects the first category information obtained by the millimeter-wave security scanner. The resulting target category information more accurately reflects whether the detected object belongs to any metal category. Therefore, the processing device can more accurately determine whether the detected object is a prohibited metal item, improving the accuracy of the millimeter-wave security scanner in detecting prohibited metal items and overall prohibited items.

[0070] Of course, implementing any product or method of this application does not necessarily require achieving all of the advantages described above at the same time. Attached Figure Description

[0071] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other embodiments can be obtained based on these drawings.

[0072] Figure 1 This is a flowchart of a security inspection method provided in an embodiment of this application;

[0073] Figure 2 Based on Figure 1 A schematic diagram of the security zone division of the metal detection gate in the embodiment shown;

[0074] Figure 3 A flowchart illustrating a specific implementation of S102 provided in this application embodiment;

[0075] Figure 4 A flowchart illustrating another specific implementation of S102 provided in the embodiments of this application;

[0076] Figure 5 Based on Figure 2 A diagram showing the correspondence between the second location of the target personnel and the defense zone;

[0077] Figure 6 This is a schematic diagram of a security check process provided in an embodiment of this application;

[0078] Figure 7 This is a schematic diagram of a security inspection system provided in an embodiment of this application;

[0079] Figure 8 This is a schematic diagram of a security inspection device provided in an embodiment of this application;

[0080] Figure 9 This is a schematic diagram of a processing apparatus provided in an embodiment of this application. Detailed Implementation

[0081] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art based on this application are within the scope of protection of this application.

[0082] To improve the accuracy of millimeter-wave security scanners in detecting contraband, this application provides a security inspection method, system, apparatus, processing device, computer-readable storage medium, and computer program product. The security inspection method provided in this application is described below.

[0083] The security inspection method provided in this application is applied to a processing device. The processing device can be the backend processor of a millimeter-wave security scanner or metal detector gate, a processor in a security inspection system, a server, etc., and is not specifically limited here. For clarity, it will be referred to as a processing device below. Furthermore, the millimeter-wave security scanner and metal detector gate in this application can be integrated into one physical device or two independent physical devices, and is not specifically limited here.

[0084] like Figure 1 As shown, a security inspection method includes the following steps S101-S105.

[0085] S101, Based on the millimeter-wave scan image of the target person passing through the millimeter-wave security scanner, determine the first detection result of the target person.

[0086] The first detection result includes the first location information of the detected item carried by the target person on the millimeter-wave scan image and the first category information of the detected item.

[0087] S102, based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, and the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, determine the target security zone corresponding to the first position information in the detection area of ​​the metal detector gate.

[0088] The detection area of ​​the metal detector gate can be divided into multiple protection zones.

[0089] S103, Obtain the second detection result corresponding to the detected item in the target defense zone.

[0090] The second detection result includes at least one of the metal strength information corresponding to the detected item in the target defense zone and the second category information of the detected item.

[0091] S104, using the second detection result to correct the first category information of the detected item, and determine the target category information of the detected item.

[0092] S105, determine whether the detected item is a prohibited item based on the target category information.

[0093] In the technical solution provided in this application embodiment, the processing device can determine a first detection result of the target person based on the millimeter-wave scan image of the target person passing through the millimeter-wave security scanner. The first detection result includes the first position information of the detected item carried by the target person on the millimeter-wave scan image and the first category information of the detected item. Based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, and the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, the device determines the target security zone corresponding to the first position information in the detection area of ​​the metal detector gate. The detection area of ​​the metal detector gate can be divided into multiple security zones. The device obtains a second detection result corresponding to the detected item in the target security zone. The second detection result includes at least one of the metal strength information and the second category information of the detected item in the target security zone. The first category information of the detected item is corrected using the second detection result to determine the target category information of the detected item. The device then determines whether the detected item is a contraband based on the target category information.

[0094] Because any metallic object will interfere with the magnetic field between metal detector gates when passing through them, metal detector gates can detect metallic objects more accurately than millimeter-wave security scanners. The second detection result obtained by using the metal detector gate to detect the object corrects the first category information obtained by the millimeter-wave security scanner. The resulting target category information more accurately reflects whether the detected object belongs to any metal category. Therefore, the processing device can more accurately determine whether the detected object is a prohibited metal item, improving the accuracy of the millimeter-wave security scanner in detecting prohibited metal items and overall prohibited items.

[0095] To improve the accuracy of millimeter-wave security scanners in detecting prohibited items, in addition to using millimeter-wave security scanners to detect the target personnel and obtain the first detection result, metal detector gates can also be used to detect the target personnel and obtain the second detection result. The second detection result can then be used to correct the first detection result, so as to accurately determine whether the items carried by the target personnel are prohibited items.

[0096] The security checkpoint is equipped with millimeter-wave security scanners. When a person passes through the scanner, it scans them to obtain a millimeter-wave image. A processing device can then acquire this image. For example, the processing device can receive the millimeter-wave signal transmitted back from the scanner and generate a millimeter-wave image based on it. Alternatively, it can receive a millimeter-wave image from an electronic device used to generate the image, thus obtaining the millimeter-wave image of the person.

[0097] After acquiring the millimeter-wave scan image, the processing device can execute the aforementioned step S101, that is, determine the first detection result of the target person based on the millimeter-wave scan image of the target person passing through the millimeter-wave security scanner. The processing device can employ various methods to detect the millimeter-wave scan image. For example, the processing device can use a pre-trained first recognition network model to detect the millimeter-wave scan image and obtain the first detection result of the target person; the processing device can also use an image detection algorithm to detect the millimeter-wave scan image and obtain the first detection result of the target person, etc.

[0098] The first detection result may include the first location information of the detected item carried by the target person on the millimeter-wave scan image, and the first category information of the detected item.

[0099] The first position information of the detected item on the millimeter-wave scan image can be represented by the position information of the detection box of the detected item in the image, or by the coordinate range of the detected item in the image coordinate system of the millimeter-wave scan image.

[0100] Furthermore, when the first location information of the detected item is represented by the location information of the detection frame of the detected item in the image, the location information of the detection frame in the image can be represented by the vertex coordinates of one vertex of the detection frame, the width of the detection frame, and the height of the detection frame. For example, the location information of the detection frame can be represented as (x, y, w, h), where x and y are the horizontal and vertical coordinates in a fixed planar coordinate system in the detection area of ​​the millimeter-wave security scanner, respectively, w represents the width of the detection frame, and h represents the height of the detection frame. The location information of the detection frame in the image can also be represented by the vertex coordinates of the four vertices of the detection frame, etc., without specific limitations here.

[0101] The first category information of the detected items can include the first confidence level of the detected item belonging to each item category, which can be represented by a first confidence matrix. The item categories to which the detected items belong can include information on pre-defined prohibited items, such as knives and lighters, or information on items included in a pre-defined millimeter-wave whitelist. The whitelist includes non-prohibited items suspected of being prohibited items, or customized, specified types of items, such as belts and metal ornaments.

[0102] This allows for the pre-setting of item categories, which can be represented by numerical identifiers. For example, different item categories can correspond to numerical identifiers such as 0, 1, 2, ...

[0103] After determining the first detection result of the target person, the processing device can perform the above step S102, that is, based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, and the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, determine the target security zone corresponding to the first location information in the detection area of ​​the metal detector gate.

[0104] To improve the accuracy of detecting prohibited metal items, metal detectors are installed in the security checkpoint. The detection area of ​​these metal detectors can be divided into multiple zones. The number and size of these zones can be set according to actual needs. After determining the zones corresponding to the metal detector's detection area, these zones can be numbered. For example, in... Figure 2 In the schematic diagram of the detection area of ​​the metal detector gate, the detection area of ​​the metal detector gate corresponds to 15 zones numbered 1, 2, 3, 4, ..., 13, 14 and 15.

[0105] The positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate can be preset or obtained through image matching.

[0106] In one implementation, based on the dimensions of the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector, when the dimensions of the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector are the same, a positional correspondence between the areas at the same position in the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector can be directly established. When the dimensions of the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector are different, a target detection area with the same size and corresponding position as the detection area of ​​the smaller device can be segmented from the detection area of ​​the larger device, and the position in the target detection area can be correlated with the position in the detection area of ​​the smaller device to obtain the positional correspondence between the areas at the same position in the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector.

[0107] When the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate is preset, security personnel can provide standardized guidance on the posture of the target person passing through the millimeter-wave security scanner and the metal detector gate, so that the target person passes through the millimeter-wave security scanner and the metal detector gate in a standard posture. This ensures that the preset positional correspondence between the detection areas of the millimeter-wave security scanner and the detection areas of the metal detector gate can be used correctly to detect prohibited items.

[0108] In one implementation, the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate can be obtained through image matching. Specifically, the processing device can acquire the human body detection image of the target person passing through the metal detector gate collected by the image acquisition device, and match the human body detection image with the millimeter-wave scanning image to determine the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate.

[0109] For clarity, the specific method for determining the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate through image matching will be explained below.

[0110] By determining the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, the processing device can determine the target security zone corresponding to the first position information in the detection area of ​​the metal detector gate based on the correspondence.

[0111] Since the first position information of the detected item carried by the target person on the millimeter-wave scan image represents the image position of the detected item in the millimeter-wave scan image, and the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner represents the correspondence between each image position in the millimeter-wave scan image and the physical position of the detection area, the processing device can determine the physical position corresponding to the first position information in the detection area of ​​the millimeter-wave security scanner based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner; and since the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate represents the correspondence between each physical position in the detection area of ​​the millimeter-wave security scanner... The processing device determines the physical location of the first location information within the detection area of ​​the metal detector gate by establishing the correspondence between the first location information and the physical location of the first location information within the detection area of ​​the millimeter-wave security scanner, as well as the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate. Furthermore, the processing device can determine the target security zone corresponding to the first location information within the detection area of ​​the metal detector gate based on the zone division of the detection area. This target security zone can represent the security zone where the detected items carried by the target person are located when the target person passes through the metal detector gate.

[0112] For example, the processing device can determine the target zone corresponding to the first location information in the detection area of ​​the metal detector as the detection zone of the metal detector based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, and the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector. Figure 2 The defense zone numbered 7.

[0113] When a person passes through a metal detector gate, the gate can detect them. The processing device can obtain the metal detector gate detection result. For example, the processing device can acquire the metal detection signal obtained by the metal detector gate when detecting the person, reconstruct the metal detection signal to obtain metal detection data, and use a trained second recognition network model to detect the metal detection data to obtain the metal detector gate detection result.

[0114] Since the detection area of ​​the metal detector gate corresponds to multiple defense zones, the detection result of the metal detector gate includes the second detection result corresponding to the detected item in each defense zone. The metal detector gate detection result includes at least one of the metal strength information of each detected item carried by the target person and the second category information of the detected item; the second detection result corresponding to the detected item in each defense zone includes at least one of the metal strength information of the detected item in that defense zone and the second category information of the detected item.

[0115] The metal strength information includes the metal intensity of the detected item. The metal intensity of the detected item is positively correlated with the electromagnetic response intensity. Based on this metal strength information, the processing device can determine whether the detected item contains a metal object and distinguish the size, shape, and location of the metal object. The second category information may include the second confidence level that the detected item belongs to different metal materials. This second confidence level can be represented by a second confidence level matrix.

[0116] When the first location information of the detected item on the millimeter-wave scan image is determined to correspond to the target zone in the detection area of ​​the metal detector gate, the processing device can acquire a second detection result corresponding to the detected item in the target zone, wherein the second detection result includes at least one of the metal strength information corresponding to the detected item in the target zone and the second category information of the detected item.

[0117] It should be noted that, in this embodiment of the application, the order in which the processing device acquires the first detection result and the second detection result is not limited. The processing device may acquire the first detection result and the second detection result simultaneously, or it may acquire the first detection result first and then acquire the second detection result. For example, the processing device may acquire the first detection result first, and then acquire the second detection result if the first detection result meets the preset acquisition conditions; or it may acquire the second detection result first and then acquire the first detection result. All of these are reasonable and are not specifically limited here.

[0118] The aforementioned preset acquisition conditions are used to define the conditions for triggering the acquisition of the second detection result. These preset acquisition conditions can be reasonably configured according to actual detection needs. For example, the preset acquisition conditions may include: the first detection result indicates that the category of the detected item is suspicious, or the current security level requirement of the security inspection scenario is the target level and the first detection result indicates that the category of the detected item is safe, etc. These are all reasonable and are not specifically limited here.

[0119] After obtaining the second detection result corresponding to the detected item in the target defense zone, the processing device can execute the above step S104, that is, use the second detection result to correct the first category information of the detected item and determine the target category information of the detected item.

[0120] The target category information includes the overall confidence level of the detected items belonging to each item category.

[0121] In one implementation, when the first detection result is represented by a first confidence level and the second detection result includes second category information of the detected item and the second category information is represented by a second confidence level, the first confidence level and the second confidence level can be weighted and superimposed to obtain the target category information of the detected item.

[0122] In one implementation, when the first detection result is represented by a first confidence level and the second detection result includes the metal strength information of the detected item and the metal strength information is represented by a metal strength value, the first confidence level and the metal strength value can be weighted and superimposed to obtain the target category information of the detected item.

[0123] Since any metal object will interfere with the magnetic field between the metal detector gates when it passes through them, the metal detector gates can accurately detect metal objects. Therefore, the purpose of the processing device in correcting the first category information using the second detection result is to correct the reliability of the detected object belonging to each metal category.

[0124] Furthermore, the processing device can determine whether the detected item is a prohibited item based on the target category information, i.e., perform the above step S105.

[0125] A list of item categories can be pre-set, which includes information on the categories of prohibited items and those included in the millimeter-wave whitelist. The pre-set list of item categories is then stored in the processing device.

[0126] When the processing device determines the target type information of the detected item, it can determine whether the detected item is a prohibited item in the item category list based on the target category information.

[0127] In one implementation, when the confidence level of the target type information characterizing the category of the detected item as belonging to the category of prohibited items is greater than a first preset threshold, the processing device can determine that the detected item is a prohibited item.

[0128] Accordingly, if the confidence level that the target type information indicates the detected item belongs to a category included in the prohibited items is not greater than the first preset threshold, the processing device can determine that the detected item is not a prohibited item; or, if the confidence level that the target type information indicates the detected item belongs to a category included in the millimeter-wave whitelist is greater than the second preset threshold, the processing device can determine that the detected item is not a prohibited item. The first and second preset thresholds can be set according to actual inspection needs. For example, the first preset threshold could be 60% or 70%, and the second preset threshold could be 70% or 80%, all of which are reasonable and not specifically limited here.

[0129] In one implementation, the processing device can determine the maximum confidence level among the overall confidence levels of each item category corresponding to the detected item, and take the item category corresponding to the maximum confidence level as the item category to which the detected item belongs. If the determined item category to which the detected item belongs is an item category included in prohibited items, the processing device determines that the detected item is a prohibited item. If the determined item category to which the detected item belongs is not an item category included in prohibited items, the processing device determines that the detected item is not a prohibited item.

[0130] If the detected item is determined to be a prohibited item, the processing device can output a preset alarm message so that the staff can re-inspect the target person; correspondingly, if the detected item is not a prohibited item, the processing device can output a detection pass message to inform the staff that the target person is not carrying prohibited items and can be released.

[0131] As can be seen, in this embodiment, since any metal object will interfere with the magnetic field between the metal detector gates when passing through them, the metal detector gate can detect metal objects more accurately than the millimeter-wave security scanner. The second detection result obtained by using the metal detector gate to detect the object corrects the first category information obtained by the millimeter-wave security scanner. The resulting target category information can more accurately reflect whether the detected object belongs to each metal category. Therefore, the processing device can more accurately determine whether the detected object is a prohibited item of the metal category, improving the accuracy of the millimeter-wave security scanner in detecting prohibited items of the metal category and improving the overall accuracy of the millimeter-wave security scanner in detecting prohibited items.

[0132] As one embodiment of this application, the security inspection method provided in this application embodiment may further include:

[0133] The system acquires a human body detection image of the target person passing through the metal detector gate, captured by an image acquisition device; determines a matching relationship between the human body detection image and the millimeter-wave scan image based on the similarity between the body part features included in the human body detection image and the body part features included in the millimeter-wave scan image; determines a correspondence between the human body detection image and the detection area of ​​the metal detector gate based on the positional relationship between the image acquisition device and the metal detector gate; and determines a positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, the correspondence between the human body detection image and the detection area of ​​the metal detector gate, and the matching relationship.

[0134] During the process of a target person passing through the security check area, the target person's state when passing through the millimeter-wave security scanner and the metal detector gate may be different. For example, the target person may pass through the millimeter-wave security scanner from the middle area of ​​the detection area and the metal detector gate from the left half area of ​​the detection area. The target person's state when passing through the millimeter-wave security scanner and the metal detector gate is different.

[0135] In this case, in order to accurately and quickly determine the target security zone corresponding to the detection area of ​​the metal detector gate based on the first position information of the detected item carried by the target person in the millimeter-wave scanning image, the human body detection image of the target person passing through the metal detector gate can be matched with the millimeter-wave scanning image to establish the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate for the target person.

[0136] To acquire images of people passing through a metal detector gate, an image acquisition device can be installed at a location corresponding to the gate, such as at the gate's entrance. The positional relationship between the image acquisition device and the metal detector gate is determined based on their installation location.

[0137] The aforementioned image acquisition device can acquire images of the detection area of ​​the metal detector gate. When a target person passes through the metal detector gate, the image acquisition device can capture a human body detection image of the target person passing through the metal detector gate.

[0138] The processing device can acquire the human body detection image of the target person when passing through the metal detector gate, which is captured by the image acquisition device, and match the human body detection image with the millimeter-wave scan image of the target person when passing through the millimeter-wave security scanner.

[0139] The processing device can identify the body features of the target person in the human body detection image and the millimeter-wave scan image, such as the hands, feet and head of the target person, so as to determine the position of each body feature in the human body detection image and the millimeter-wave scan image respectively.

[0140] The processing device can determine the body part feature with the highest similarity between the body part features included in the human body detection image and the body part features included in the millimeter-wave scan image as the same body part feature based on the similarity between the body part features included in the millimeter-wave scan image and the body part features included in the human body detection image. The position of the human body detection image corresponding to the same body part feature and the position of the millimeter-wave scan image are used as the matching position, thereby establishing a matching relationship between the human body detection image and the millimeter-wave scan image based on the body part features.

[0141] The processing device can determine the transformation relationship between the image coordinate system of the image acquisition device and the world coordinate system of the metal detection gate based on the positional relationship between the image acquisition device and the metal detection gate, as well as the device parameters of the image acquisition device. Then, based on the transformation relationship, it can determine the correspondence between the image position of the human body detection image and the physical position of the metal detection gate detection area, thus obtaining the correspondence between the human body detection image and the metal detection gate detection area.

[0142] Since the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner represents the correspondence between the image positions in the millimeter-wave scan image and the physical positions in the detection area of ​​the millimeter-wave security scanner, and the matching relationship between the human body detection image and the millimeter-wave scan image represents the matching relationship between the image positions of body part features in the human body detection image and the image positions of those body part features in the millimeter-wave scan image, the processing device can determine the correspondence between the image positions of human body part features in the human body detection image and the physical positions of those human body part features in the detection area of ​​the millimeter-wave security scanner based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, thus obtaining the correspondence between the human body detection image and the detection area of ​​the millimeter-wave security scanner.

[0143] Furthermore, since the correspondence between the human body detection image and the metal detector gate's detection area represents the correspondence between the image position of the human body detection image and the physical position of the metal detector gate's detection area, the processing device can determine the physical position in the millimeter-wave security scanner's detection area and the physical position in the metal detector gate's detection area corresponding to the image position of the human body detection image, based on the correspondence between the human body detection image and the detection area of ​​the millimeter-wave security scanner, and establish the correspondence between the physical position in the millimeter-wave security scanner's detection area and the physical position in the metal detector gate's detection area corresponding to the same image position of the human body detection image, thereby determining the positional correspondence between the millimeter-wave security scanner's detection area and the metal detector gate's detection area.

[0144] As can be seen, in this embodiment of the application, the processing device can match the human body detection image of the target person captured by the image acquisition device when passing through the metal detector gate with the millimeter-wave scanning image, thereby quickly and accurately determining the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, so as to improve the accuracy of contraband detection in the subsequent process of using the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate to detect contraband.

[0145] As one implementation method of this application, such as Figure 3As shown, step S102 above, which is the step of determining the target security zone corresponding to the first location information in the detection area of ​​the metal detector gate based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, and the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, may include:

[0146] S301, based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, determine the second location information corresponding to the first location information in the detection area of ​​the millimeter-wave security scanner.

[0147] Millimeter-wave scan images are generated based on the millimeter-wave signals transmitted back from the millimeter-wave security scanner when scanning its detection area. Therefore, the image area of ​​the millimeter-wave scan image corresponds to the detection area of ​​the millimeter-wave security scanner, and there is a correspondence between each image position of the millimeter-wave scan image and each physical position in the detection area of ​​the millimeter-wave security scanner.

[0148] Since the first position information of the detected item carried by the target person on the millimeter-wave scan image represents the image position of the detected item in the millimeter-wave scan image, and the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner represents the correspondence between each image position in the millimeter-wave scan image and the physical position of the detection area of ​​the millimeter-wave security scanner, the processing device can determine the second position information corresponding to the first position information in the detection area of ​​the millimeter-wave security scanner based on the correspondence between the image area of ​​the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner. The second position information is the physical position of the detected item carried by the target person in the detection area of ​​the millimeter-wave security scanner when the target person passes through the millimeter-wave security scanner.

[0149] S302, based on the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, determine the third positional information corresponding to the second positional information in the detection area of ​​the metal detector gate.

[0150] As mentioned earlier, the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate represents the correspondence between the physical positions in the detection area of ​​the millimeter-wave security scanner and the physical positions in the detection area of ​​the metal detector gate. The second positional relationship characterizes the physical position of the detected items carried by the target person in the detection area of ​​the millimeter-wave security scanner when the target person passes through the millimeter-wave security scanner. Therefore, based on the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, the processing device can determine the physical position corresponding to the second position information in the detection area of ​​the metal detector gate, that is, obtain the third position information corresponding to the second position information in the detection area of ​​the metal detector gate.

[0151] The third location information is determined based on the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate. It refers to the physical location of the detected items carried by the target person within the detection area of ​​the metal detector gate when the target person passes through. Since the second location information represents the location information corresponding to the first location information within the detection area of ​​the millimeter-wave security scanner, and the third location information represents the location information corresponding to the second location information within the detection area of ​​the metal detector gate, the third location information essentially represents the location information corresponding to the first location information within the detection area of ​​the metal detector gate.

[0152] S303, based on the position of the third location information in the detection area of ​​the metal detector gate, determine the target defense zone corresponding to the first location information in the detection area of ​​the metal detector gate.

[0153] Since the detection area of ​​the metal detector gate is divided into multiple zones, and each physical location within the detection area corresponds to a zone, the processing device can determine the target zone corresponding to the third location information within the detection area of ​​the metal detector gate based on the location of the third location information within the detection area and the zone division of the detection area. Furthermore, since the third location information represents the location information corresponding to the first location information within the detection area of ​​the metal detector gate, the target zone corresponding to the third location information within the detection area of ​​the metal detector gate is the same as the target zone corresponding to the first location information within the detection area of ​​the metal detector gate.

[0154] As can be seen, in this embodiment of the application, the processing device can determine the second location information corresponding to the first location information in the detection area of ​​the millimeter-wave security scanner based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner; determine the third location information corresponding to the second location information in the detection area of ​​the metal detector based on the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate; and determine the target zone corresponding to the first location information in the detection area of ​​the metal detector gate based on the position of the third location information in the detection area of ​​the metal detector gate, thereby quickly and accurately determining the target zone corresponding to the first location information in the detection area of ​​the metal detector gate, and then directly obtaining the second detection result of the detected item corresponding to the target zone, and using the second detection result to correct the first detection result, thereby improving the detection efficiency and accuracy of the detected item.

[0155] As one implementation method of this application, such as Figure 4As shown, step S102 above, which is the step of determining the target security zone corresponding to the first location information in the detection area of ​​the metal detector gate based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, and the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, may include:

[0156] S401, based on the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, and the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, determine the fourth position information corresponding to the preset position point in the detection area of ​​the metal detector gate in the millimeter-wave scan image.

[0157] The preset location point is a landmark point within the detection area of ​​the metal detector gate, and the processing device can determine the location of the detection area based on this preset location point. This preset location point can be set according to actual detection needs. It can be a vertex of the detection area of ​​the metal detector gate, such as the upper left vertex or lower right vertex of the detection area; it can be the center of the detection area; or it can be the intersection of the centerline and the boundary of the detection area, such as the intersection of the vertical centerline and the upper boundary of the detection area, or the intersection of the horizontal centerline and the left boundary of the detection area. All of these are reasonable and are not specifically limited here.

[0158] Furthermore, the preset location point can be one or multiple, which is reasonable and will not be specifically limited here.

[0159] Since the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate represents the correspondence between the physical positions in the detection area of ​​the millimeter-wave security scanner and the physical positions in the detection area of ​​the metal detector gate, the processing device can determine the physical position of the preset position point in the detection area of ​​the metal detector gate within the detection area of ​​the millimeter-wave security scanner based on the positional correspondence between the detection areas of the millimeter-wave security scanner and the detection areas of the metal detector gate.

[0160] Furthermore, since the correspondence between the millimeter-wave scanned image and the detection area of ​​the millimeter-wave security scanner represents the correspondence between the image positions in the millimeter-wave scanned image and the physical positions of the detection area of ​​the millimeter-wave security scanner, the processing device can determine the image position corresponding to the preset position point in the millimeter-wave scanned image based on the correspondence between the millimeter-wave scanned image and the detection area of ​​the millimeter-wave security scanner, as well as the physical position of the preset position point in the detection area of ​​the millimeter-wave security scanner. That is, it can determine the fourth position information corresponding to the preset position point in the millimeter-wave scanned image.

[0161] S402, in the millimeter-wave scan image, based on the fourth position information, an image region matching the detection area of ​​the metal detection gate is determined.

[0162] Since the preset location point is a landmark within the detection area of ​​the metal detector gate, and the processing device can determine the location of the detection area of ​​the metal detector gate based on the preset location point, after determining the fourth location information corresponding to the preset location point in the millimeter-wave scan image, the processing device can determine the image position relationship between the preset location point and the detection area of ​​the metal detector gate based on the scaling ratio between the detection area of ​​the millimeter-wave security scanner and the image area of ​​the millimeter-wave scan image, as well as the physical positional relationship between the preset location point and the detection area of ​​the metal detector gate; and based on the fourth location information corresponding to the preset location point in the millimeter-wave scan image, and the image position relationship between the preset location point and the detection area of ​​the metal detector gate, the processing device can determine the image area in the millimeter-wave scan image whose positional relationship with the fourth location information corresponding to the preset location point in the millimeter-wave scan image satisfies the above image positional relationship, that is, obtain the image area that matches the detection area of ​​the metal detector gate.

[0163] The image region in the millimeter-wave scan image that matches the detection area of ​​the metal detector gate is the image region in the millimeter-wave scan image that is obtained by scaling the detection area of ​​the metal detector gate proportionally to the scaling ratio between the detection area of ​​the millimeter-wave security scanner and the image region of the millimeter-wave scan image.

[0164] In one implementation, the processing device can determine the fourth position information corresponding to a preset position point in the millimeter-wave scanning image within the detection area of ​​the metal detector gate, based on the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, as well as the correspondence between the millimeter-wave scanned image and the detection area of ​​the millimeter-wave security scanner. The processing device can then scale the detection area of ​​the metal detector gate proportionally according to the scaling ratio between the detection area of ​​the millimeter-wave security scanner and the image area of ​​the millimeter-wave scanned image, determining the proportionally scaled-down detection area size of the metal detector gate. Finally, in the millimeter-wave scanned image, using the fourth position information as a reference, it determines an image area of ​​the same size as the proportionally scaled-down detection area size of the metal detector gate.

[0165] For example, when the preset position point is the midpoint of the detection area of ​​the metal detector gate, the processing device can determine the fourth position information of the midpoint of the metal detector gate's detection area in the millimeter-wave scan image. Using this fourth position information as the center, it can determine the size of the metal detector gate's detection area after proportional scaling, and crop out an image area of ​​the aforementioned size from the millimeter-wave scan image. For instance, the processing device can use the fourth position point O(0,0) as the center and, based on the distance between this center and each boundary of the metal detector gate's detection area, determine an image area in the millimeter-wave scan image with the same length Lx and width Ly as the metal detector gate's detection area, thus obtaining an image area in the millimeter-wave scan image that matches the metal detector gate's detection area.

[0166] S403, based on the position of the first location information in the image area, determine the target defense zone corresponding to the first location information in the detection area of ​​the metal detector gate.

[0167] The processing device can determine the target zone corresponding to the first location information of the detected item carried by the target person in the image area that matches the metal detector gate, based on the first location information of the detected item in the image area that matches the metal detector gate. The target zone is the zone where the detected item is located when it passes through the metal detector gate.

[0168] In one implementation, the detection area of ​​the metal detector gate is divided into multiple zones. Similarly, the image area corresponding to the detection area of ​​the metal detector gate in the millimeter-wave scan image can also be divided into multiple zones. These multiple zones of the metal detector gate's detection area correspond to multiple zones of the image area corresponding to the detection area of ​​the metal detector gate in the millimeter-wave scan image. Based on the position of the first location information in the image area, the zone corresponding to the first location information in the image area can be determined. Furthermore, based on the correspondence between the zones in the metal detector gate's detection area and the zones in the image area, the target zone corresponding to the first location information in the metal detector gate's detection area can be determined.

[0169] In another implementation, since the image area matches the detection area of ​​the metal detector gate, the actual location of the metal detector gate's detection area corresponding to the first location information is determined based on the correspondence between the metal detector gate's detection area and the image area, as well as the location of the first location information in the image area. Then, based on the correspondence between the actual location of the detection area and the defense zone, the target defense zone corresponding to the first location information in the metal detector gate's detection area can be determined.

[0170] As can be seen, in this embodiment, the processing device can determine the fourth position information corresponding to a preset position point in the detection area of ​​the metal detector gate in the millimeter-wave scanning image, based on the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, as well as the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner. In the millimeter-wave scan image, based on the fourth position information, an image area matching the detection area of ​​the metal detector gate is determined. Based on the position of the first position information in the image area, the target detection zone corresponding to the first position information in the detection area of ​​the metal detector gate is quickly and accurately determined. Furthermore, the second detection result of the detected item corresponding to the target detection zone is directly obtained, and the first detection result is corrected using the second detection result, thereby improving the detection efficiency and accuracy of the detected item.

[0171] As one embodiment of this application, step S104, which involves correcting the first category information of the detected item using the second detection result to determine the target category information of the detected item, can include the following four methods:

[0172] Method 1: Based on the second category information of the detected items in the second detection result, determine the material attention coefficient; use the material attention coefficient to correct the first category information of the detected items to determine the target category information of the detected items. The material attention coefficient reflects the degree of attention paid to different items during security checks and affects the judgment result of whether the relevant items are identified as prohibited items. The attention coefficients corresponding to different items can be reasonably set according to security check requirements.

[0173] For example, when the second detection result includes second category information of the detected items, the processing device can determine the item material attention coefficient based on the second category information. The item material attention coefficient characterizes the degree of attention paid to metal items of different materials.

[0174] The material of an item can affect the accuracy of the identification of prohibited items, and different metals can also influence the accuracy of the identification. For example, items made of gold or silver are usually jewelry, and the target person may wear gold or silver jewelry, which is generally not considered prohibited. Therefore, security checks can appropriately reduce the focus on items made of this metal. On the other hand, items made of steel or iron are usually prohibited items such as knives, and security checks can appropriately increase the focus on items made of this metal.

[0175] Therefore, the processing device can determine the material concern coefficient of the detected item based on the second category information of the detected item.

[0176] For each zone, if the second category of information includes the second confidence level of the detected item belonging to different metal materials, the processing device can determine the maximum confidence level of the second confidence level of the detected item in that zone, determine the metal material corresponding to the maximum confidence level as the item material corresponding to the detected item, and determine the item material attention coefficient.

[0177] The material attention coefficient for items can be set according to the actual detection needs of the current security check scenario. If the prohibited items that may be found in the current security check scenario are mostly made of a certain material, a higher material attention coefficient can be set for that material, and a lower material attention coefficient can be set for other materials. For example, if the current security check scenario is a train station, a higher material attention coefficient can be set for materials such as iron or steel, and a lower material attention coefficient can be set for other materials such as gold.

[0178] After determining the material concern coefficient corresponding to the detected item, the processing device can use the material concern coefficient corresponding to the detected item to correct the first category information of the detected item and obtain the target category information of the detected item.

[0179] In one implementation, when the first category information is represented by the first confidence level of the detected item belonging to each item category, and the target category information is represented by the comprehensive confidence level of the detected item belonging to each item category, the processing device calculates the product of the item material attention coefficient and the first confidence level after determining the item material attention coefficient corresponding to the detected item to obtain the comprehensive confidence level.

[0180] In one implementation, when the first category information is represented by the first confidence level of the detected item belonging to each item category, and the target category information is represented by the comprehensive confidence level of the detected item belonging to each item category, the processing device, after determining the item material attention coefficient corresponding to the detected item, uses the item material attention coefficient to adjust the confidence level corresponding to the item type belonging to the item material in the first confidence level to obtain the comprehensive confidence level.

[0181] Method 2: Determine the second adjustment parameter corresponding to the target defense zone to adjust the second detection result, and use the adjusted second detection result to correct the first category information of the detected item, thereby determining the target category information of the detected item.

[0182] The second adjustment parameter may include the sensitivity and adjustment weight of the zone, as well as other parameters that adjust the detection capability of the zone. These are all reasonable and will not be specifically limited here.

[0183] The likelihood of detecting contraband varies across different zones of a metal detector gate, influenced by factors such as body shape, areas where contraband can be concealed, pedestrian habits, and the gate's zone division rules. For zones with a high probability of detecting contraband, the second adjustment parameter can be appropriately increased; conversely, for zones with a low probability of detecting contraband, the second adjustment parameter can be appropriately decreased.

[0184] The processing device can determine the second adjustment parameters of each defense zone corresponding to the detection area of ​​the metal detector gate based on the preset region of interest of the human body, the positional correspondence between the preset region of interest and the detection area of ​​the metal detector gate, wherein the preset region of interest can be determined based on the body parts of the person being detected.

[0185] Because certain parts of the human body are covered by limbs, such as the armpits, head, or groin, these areas are easy to conceal contraband, and millimeter-wave security scanners have a low accuracy rate in detecting these areas.

[0186] When inspecting these areas, the metal detector can appropriately increase its focus on these areas to make it easier to detect contraband hidden there. This involves designating these areas as preset regions of interest (ROIs), and determining the corresponding zones based on human anatomy, pedestrian habits, and the metal detector's zone divisions. When determining the second adjustment parameters for each zone of the metal detector, the processing device can set the second adjustment parameters of zones corresponding to the preset ROIs to a larger value; for other zones not corresponding to the preset ROIs, the second adjustment parameters can be set to a smaller value. The processing device can then use these determined second adjustment parameters to adjust the first confidence level corresponding to the detected item.

[0187] The processing device can determine the second adjustment parameter corresponding to the target defense zone and adjust the second detection result using the second adjustment parameter corresponding to the target defense zone.

[0188] In one implementation, the second adjustment parameter includes the sensitivity corresponding to the target defense zone. The processing device can set the sensitivity corresponding to the target defense zone to a first sensitivity and set the sensitivity corresponding to non-target defense zones to a second sensitivity. The first sensitivity is greater than the second sensitivity.

[0189] In one implementation, the processing device can determine a preset region of interest (ROI) within the detection area of ​​the millimeter-wave security scanner based on the body posture information of the target person in the millimeter-wave scanned image. The body posture information of the target person may include height, shoulder width, and arm span; the preset ROI may include the head, armpits, groin, and feet.

[0190] Based on the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, the corresponding security zone of the preset region of interest (ROI) within the detection area of ​​the metal detector gate is determined. The second adjustment parameter of the security zone corresponding to the preset ROI is set to a larger second adjustment parameter, while the second adjustment parameter of the security zones outside the preset ROI is set to a smaller second adjustment parameter. By combining the human body information from the millimeter-wave scan image and the security zone information of the metal detector gate, the sensitivity of each security zone of the metal detector gate is adjusted, resulting in a matrix of second adjustment parameters corresponding to each security zone.

[0191] For example, such as Figure 5 As shown, the detection area enclosed by rectangle 501 corresponds to the second position of the target personnel's armpit. The area enclosed by rectangle 501 corresponds to four defense zones: defense zone 4, defense zone 5, defense zone 7, and defense zone 8. The processing device can set a relatively large second adjustment parameter for each of these four defense zones. After determining the second adjustment parameter for each defense zone, the processing device can determine the second adjustment parameter corresponding to the target defense zone.

[0192] After determining the second adjustment parameter corresponding to the target defense zone, the processing device can use the second adjustment parameter corresponding to the target defense zone to adjust the second detection result.

[0193] In one implementation, when the second adjustment parameter is sensitivity, the second detection result is represented by the metal strength information of the detected item. The processing device can calculate the product of the sensitivity of the target defense zone and the metal strength information of the detected item corresponding to the target defense zone to obtain the adjusted metal strength information of the detected item corresponding to the target defense zone, that is, the adjusted second detection result.

[0194] For example, the second adjustment parameter is sensitivity, and the target defense zone is as follows: Figure 2 The defense zone 5 shown has a sensitivity of β(5), and the metal strength information of the detected item corresponding to defense zone 5 is A. metal (5) The processing device calculates the product β(5)×A of the sensitivity of zone 5 and the metal strength information of the detected item corresponding to zone 5. metal (5) Obtain the metal strength information of the detected items corresponding to the adjusted target defense zone, which is the adjusted second detection result.

[0195] In one implementation, when the second adjustment parameter is the material attention coefficient, the material attention coefficient can be determined based on the second category information of the item detected in the second detection result. The material attention coefficient is then used to adjust the second detection result to obtain the adjusted second detection result.

[0196] In one implementation, when the second adjustment parameter is the category adjustment parameter corresponding to the first category information of the detected item, the second detection result can be adjusted using the category adjustment parameter, and the first category information can be corrected using the adjusted second detection result to determine the target category information of the detected item.

[0197] The processing device can use the adjusted second detection result to correct the first category information of the detected item and determine the target category information of the detected item.

[0198] In one implementation, when the first category information is represented by the first confidence level of the detected item belonging to each item category, and the adjusted second detection information is the metal strength information of the detected item corresponding to the adjusted target defense zone, the first confidence level and the adjusted second detection information (metal strength information of the detected item corresponding to the adjusted target defense zone) can be weighted and summed to obtain the target category information of the detected item.

[0199] The subsequent control device can execute step S103 to adjust the first confidence level using the updated metal strength value to obtain a comprehensive confidence level.

[0200] Method 3: Determine the first adjustment parameter corresponding to the first category information of the detected item, adjust the first category information of the detected item, and use the second detection result to revise the adjusted first category information again to determine the target category information of the detected item.

[0201] The processing device can determine the first adjustment parameter corresponding to the first category information of the detected item, and adjust the first category information of the detected item using the first adjustment parameter corresponding to the first category information. Then, it can use the detection result of the metal detector gate to correct the detection result of the millimeter-wave security scanner again to ensure the accuracy and reliability of the jointly output detection result.

[0202] To further improve the accuracy of prohibited item detection results, the processing device can also adjust the first category information using a first adjustment parameter to enhance its accuracy. The first adjustment parameter can be determined based on the level of attention given to each item category in the current security check scenario. For item categories with high attention in the current security check scenario, such as prohibited items, a higher first adjustment parameter can be set; for item categories with low attention in the current security check scenario, a lower first adjustment parameter can be set. The first adjustment parameter may include an item material attention coefficient and / or a category adjustment parameter, where the category adjustment parameter characterizes the level of attention given to the item category in the current security check scenario, and can be represented in matrix form. For example, for Nc item categories numbered 1, 2, ..., Nc, these item categories correspond to category adjustment parameters α1, α2, ..., αc, respectively.Nc These category adjustment parameters can be represented in matrix form as follows:

[0203] [α1, α2, ..., α] Nc ].

[0204] The processing device can acquire the first adjustment parameters corresponding to each item category in the current security inspection scenario, which are manually set by security personnel. Based on the acquired first adjustment parameters, it adjusts the first category information corresponding to the detected item.

[0205] In one implementation, where the first category information is represented by a first confidence level that the detected item belongs to each item category, and the first adjustment parameter may include a category adjustment parameter, the processing device can calculate the product of the category adjustment parameter and the first confidence level to obtain the adjusted first category information.

[0206] In one implementation, when the first category information is represented by the first confidence level of the detected item belonging to each item category, and the first adjustment parameter may include the category adjustment parameter and the item material attention coefficient, the processing device can obtain the confidence adjustment parameter matrix based on the item material attention coefficient and the category adjustment parameter, and calculate the product of the confidence adjustment parameter matrix and the first confidence level to obtain the adjusted first category information.

[0207] The subsequent processing device can use the second detection result to further correct the adjusted first category information and determine the target category information of the detected items.

[0208] In one implementation, when the first category information is represented by the first confidence level of the detected item belonging to each item category, and the second detection result is represented by the metal strength information of the detected item, the weighted sum of the second detection result and the adjusted first category information can be calculated to obtain the second detection result relative to the adjusted first category information.

[0209] Method 4: Determine the first adjustment parameter corresponding to the first category information of the detected item, and determine the second adjustment parameter corresponding to the target defense zone; adjust the first category information of the detected item using the first adjustment parameter, and adjust the second detection result using the second adjustment parameter; use the adjusted second detection result to further correct the adjusted first category information, and determine the target category information of the detected item.

[0210] The processing device can determine a first adjustment parameter corresponding to the first category information of the detected item, and a second adjustment parameter corresponding to the target zone; adjust the first category information of the detected item using the first adjustment parameter, and adjust the second detection result using the second adjustment parameter; further correct the adjusted first category information using the adjusted second detection result to determine the target category information of the detected item. By separately correcting the detection results of the millimeter-wave security scanner and the metal detector gate, and then jointly determining the category information of the detected item based on the two corrected detection results, the accuracy and reliability of the final detection result are further improved.

[0211] The process involves determining the first adjustment parameter corresponding to the first category information of the detected item, and determining the second adjustment parameter corresponding to the target defense zone; adjusting the first category information of the detected item using the first adjustment parameter, and adjusting the second detection result using the second adjustment parameter; as described above, this can be implemented using one or more of the above methods, and will not be repeated here.

[0212] In one implementation, when the first category information is represented by the first confidence level of the detected item belonging to each item category, and the second detection result is represented by the metal strength information of the detected item, the weighted sum of the adjusted second detection result and the adjusted first category information can be calculated to determine the target category information of the detected item.

[0213] As can be seen, in the embodiments of this application, the first category information of the detected items can be corrected using the second detection result in a variety of ways to determine the target category information of the detected items. This has high scenario adaptability and can improve the accuracy of the detection results of prohibited items according to the needs of different security inspection scenarios.

[0214] As one embodiment of this application, the second adjustment parameter includes the sensitivity corresponding to the target defense zone. Accordingly, the step of determining the second adjustment parameter corresponding to the target defense zone may include:

[0215] Set the sensitivity corresponding to the target defense zone to the first sensitivity.

[0216] The security inspection method provided in this application embodiment may further include: setting the sensitivity corresponding to a non-target zone to a second sensitivity. Wherein, the first sensitivity is greater than the second sensitivity.

[0217] Since the target zone is the zone where the detected item is located, in order to improve the accuracy of the detection results for the detected item, when the second adjustment parameter is the sensitivity corresponding to the target zone, the processing device can set the sensitivity corresponding to the target zone to the first sensitivity and set the sensitivity corresponding to the non-target zone to the second sensitivity. The first sensitivity is greater than the second sensitivity.

[0218] As can be seen, in the embodiments of this application, the processing device can improve the sensitivity of the target zone corresponding to the detection area of ​​the metal detector gate, reduce the probability of the target zone missing the detection of contraband, and thus improve the accuracy of the millimeter-wave security inspection device in detecting contraband.

[0219] As one embodiment of this application, the aforementioned first adjustment parameter includes an article material attention coefficient and / or a category adjustment parameter. Accordingly, the step of determining the first adjustment parameter corresponding to the first category information of the detected article may include:

[0220] Based on the second category information of the detected items in the second detection result and / or the item category information included in the pre-set millimeter wave whitelist, the category adjustment parameter is determined, wherein the category adjustment parameter represents the degree of attention paid to item category in the current security inspection scenario;

[0221] And / or,

[0222] Based on the second category information of the detected items in the second detection result, the material attention coefficient of the items is determined, wherein the material attention coefficient of the items represents the degree of attention to metal items of different materials.

[0223] The first adjustment parameters may include the item material attention factor and / or category adjustment parameters.

[0224] The processing device can determine category adjustment parameters based on the second category information of the detected items in the second detection result and / or the item category information included in the pre-set millimeter wave whitelist. The category adjustment parameters represent the degree of attention paid to item categories in the current security inspection scenario.

[0225] In one implementation, upon obtaining a second detection result, the processing device can determine category adjustment parameters based on the second category information of the detected items in the second detection result and / or the item category information included in a pre-set millimeter-wave whitelist. That is, the processing device can, based on the item category information included in the pre-set millimeter-wave whitelist in the second category information of the detected items, reduce the focus on item categories included in the millimeter-wave whitelist and increase the focus on item categories outside the pre-set focus on specific categories.

[0226] The processing device can determine the material attention coefficient of the item based on the second category information of the item detected in the second detection result, wherein the material attention coefficient represents the degree of attention to metal items of different materials.

[0227] In one implementation, the processing device detects the second category information of the item in the second detection result, and can determine the item material attention coefficient based on the detected second category information. The implementation method for determining the item material attention coefficient has been explained above and will not be repeated here.

[0228] As one embodiment of this application, the first adjustment parameter includes an item material attention coefficient and a category adjustment parameter, the second adjustment parameter includes the sensitivity corresponding to the target defense zone, the first category information includes a first confidence level that the detected item belongs to the item category, and the second detection result includes the metal strength value corresponding to the detected item in the target defense zone.

[0229] The steps described above—determining the first adjustment parameter corresponding to the first category information of the detected item, and determining the second adjustment parameter corresponding to the target defense zone; adjusting the first category information of the detected item using the first adjustment parameter, and adjusting the second detection result using the second adjustment parameter; and further correcting the adjusted first category information using the adjusted second detection result to determine the target category information of the detected item—may include:

[0230] The overall confidence level P of the detected item belonging to the item category is calculated using the following formula. fin (c) , as target category information for the detected items:

[0231] P fin (c)=M0×α(c)×P mmw (c)+β(m)×A metal (m);

[0232] Where c is the item category number, α(c) is the category adjustment parameter corresponding to item category c, and P mmw (c) represents the first confidence level that the detected item belongs to item category c, M0 is the material attention coefficient corresponding to the detected item, m is the target zone number corresponding to the detection location of the detected item, β(m) is the sensitivity corresponding to target zone m, and A metal (m) represents the metal strength value corresponding to the target defense zone m.

[0233] Confidence P of detecting items belonging to different item categories mmw (The probabilities of all samples are between 0 and 1, and the total confidence level is 1), then the first confidence matrix is ​​P.mmw [P1, P2, ... P] Nc ], where 1 to Nc correspond to different sample category numbers, i.e., c belongs to [1, 2, ..., Nc]. M0×α is the confidence adjustment parameter matrix, [M0α1, M0α2, ..., M0α... Nc ]. m belongs to [1, 2, ..., M], where M is the number of detection zones of the metal detector gate.

[0234] The above formula adjusts the first confidence level of the detected item belonging to item category c using category adjustment parameters and item material attention coefficient to obtain the adjusted first confidence level. It also adjusts the metal strength value corresponding to the target defense zone m using the sensitivity to obtain the adjusted metal strength value. Finally, it corrects the adjusted first confidence level using the adjusted metal strength value to obtain the comprehensive confidence level of the item category.

[0235] For each detected item in the first test result, the first test result is corrected using the above formula and the second test result to obtain the overall confidence level P of the detected item belonging to the given item category. fin (c) serves as the target category information for the detected item.

[0236] As can be seen, in this embodiment, the processing device uses the item material attention coefficient, category adjustment parameters, zone sensitivity, the metal strength value corresponding to the target zone, and a first confidence level to calculate the comprehensive confidence level. The obtained comprehensive confidence level can accurately reflect whether the detected item belongs to the metal material items that are of greater concern in the current security inspection scenario, whether the detected item belongs to the prohibited item category with a high degree of concern in the current security inspection scenario, and whether the detected item in the preset area of ​​interest is a metal item. In this way, based on this comprehensive confidence level, it is possible to more accurately determine whether the detected item is a prohibited item of the metal category, further improving the accuracy of the millimeter-wave security scanner in detecting prohibited items of the metal category, and thus further improving the accuracy of the millimeter-wave security scanner in detecting prohibited items.

[0237] By applying the technical solution provided in this application, the information detected by the millimeter-wave security scanner and the metal detector gate can be complementaryly fused. For example, the shape and location information of the object in the millimeter-wave scan image generated by the millimeter-wave security scanner can be fused with the metal material and size detected by the metal detector gate. This improves the accuracy of the millimeter-wave security scanner in detecting contraband. For the problems of detection blind spots, incomplete target display, and the inability of millimeter waves to penetrate the skin, which make it difficult to identify very small contraband or contraband hidden in areas such as the mouth, the metal detector gate can detect contraband more accurately in these situations based on its own magnetic field characteristics. The millimeter-wave security scanner has a higher comprehensiveness of detection than the metal detector gate, and can detect items other than metal objects. This solution, by fusing the millimeter-wave security scanner and the metal detector gate, effectively improves the accuracy of contraband detection and enhances the detection effect of the active millimeter-wave security scanner in blind spots and when the target is incomplete.

[0238] As one embodiment of this application, step S103 above, namely the step of obtaining the second detection result corresponding to the detected item in the target defense zone, may include:

[0239] If the first detection result indicates that the detected item belongs to a suspicious category, or if the security level requirement of the current security inspection scenario is the target level and the first detection result indicates that the detected item belongs to a safe category, then obtain the second detection result corresponding to the detected item in the target security zone.

[0240] After obtaining the first detection result, if the first detection result indicates that the category of the detected item is suspicious, the processing device can obtain the second detection result corresponding to the detected item in the target defense zone, so as to use the second detection result to correct the first detection result and determine whether the detected item is a contraband.

[0241] For scenarios with high security requirements, such as museums, exhibition halls, or airports, even if the first detection result indicates that the detected item belongs to the safe category, it is still necessary to obtain a second detection result from the metal detector gate. Combining the dual detection results from the millimeter-wave security scanner and the metal detector gate, it is necessary to determine whether the detected item is a prohibited item.

[0242] Based on this, the security level requirements of the security inspection scenario can be classified in advance. According to the security requirements from high to low, multiple security levels can be divided. After obtaining the first detection result, the processing device can obtain the second detection result corresponding to the detected item in the target security zone if the security level requirement of the current security inspection scenario is the target level and the first detection result indicates that the category of the detected item is a safe category. This allows the processing device to use the second detection result to correct the first detection result and further determine whether the detected item is a contraband.

[0243] The target level mentioned above should be set reasonably according to the actual testing needs. For example, it can be the highest safety level corresponding to the safety level requirement, or it can be any safety level other than the lowest safety level among the various safety levels corresponding to the safety level requirement. All of these are reasonable and no specific limitation is made here.

[0244] As can be seen, in this embodiment of the application, when the first detection result indicates that the category of the detected item is suspicious, or when the security level requirement of the current security inspection scenario is the target level and the first detection result indicates that the category of the detected item is safe, a second detection result corresponding to the detected item in the target area is obtained. In this way, the processing device can continue to obtain the second detection result corresponding to the detected item in the target area under the above circumstances, and use the second detection result to correct the first detection result, thereby improving the accuracy of detecting contraband.

[0245] The following is combined with Figure 6 The security check flowchart shown illustrates the security check methods with examples.

[0246] Figure 6 In this process, the target personnel pass through the active millimeter-wave security scanner at the entrance, and the processing device can obtain the initial inspection result of the millimeter-wave security scanner inspection of the target personnel, that is, the first inspection result.

[0247] The detected items may belong to one of the following categories: hazardous, suspicious, or safe.

[0248] Based on the initial inspection results, the processing device can determine the category of the detected item as dangerous if the first confidence level of the detected item being a prohibited item is greater than the preset confidence threshold of the millimeter-wave security scanner; if the first confidence level of the detected item being a category included in the millimeter-wave whitelist is greater than the preset confidence threshold, the processing device can determine the category of the detected item as safe if the first detection result indicates that the detected item belongs to neither a dangerous nor a safe category; and if the first detection result indicates that the detected item belongs to neither a dangerous nor a safe category, the processing device can determine the category of the detected item as suspicious.

[0249] The preset reliability threshold can be set according to the security level requirements of the current scenario. If a higher security level is required, a lower preset reliability threshold can be set to reduce the false negative rate; conversely, if a lower security level is required, a higher preset reliability threshold can be set to reduce the false positive rate and prevent false positives from affecting security check efficiency. For example, the preset reliability threshold can be set to 90%.

[0250] In one implementation, some item categories may be falsely identified as contraband because their materials are similar to those of prohibited items. To reduce the false detection rate, a millimeter-wave whitelist of item categories can be set up, adding these item categories with similar materials to prohibited items to the millimeter-wave whitelist.

[0251] If the initial detection result indicates that the detected item belongs to a dangerous category, the millimeter-wave security scanner can directly issue a warning, alerting security personnel that the person is carrying prohibited items. In this case, security personnel can directly guide the person to the manual security check area for a manual inspection. After the manual inspection eliminates prohibited items and suspicious items that could easily lead to false detection by the millimeter-wave security scanner, security personnel will guide the person through passage 3 between the manual security check area and the entrance of the millimeter-wave security scanner for a second inspection.

[0252] If the first detection result indicates that the detected item belongs to a suspicious category, the processing device can determine that the first detection result corresponding to the target person meets the preset acquisition conditions, and it is necessary to use a metal detector gate to detect the target person again.

[0253] If the initial detection result indicates that the detected item belongs to the safe category, the processing device can further determine whether the security level requirement of the current security check scenario meets the target level. The target level indicates a higher security level requirement for the current security check scenario; unless the initial detection result indicates that the detected item belongs to the dangerous category, a metal detector gate is required to further check the target person.

[0254] If the security level requirement of the current security inspection scenario is the target level and the first detection result indicates that the category of the detected item is a safe category, the processing device can determine that the first detection result corresponding to the target person meets the preset acquisition conditions, and it is necessary to use a metal detector gate to inspect the target person again.

[0255] In situations where the current security level requirement is not the target level (i.e., the security level requirement is low) and the first detection result indicates that the detected item belongs to the safe category, or when the millimeter-wave device does not detect a large item, the processing device can determine that the first detection result for the target person does not meet the preset detection conditions. Therefore, it is unnecessary to use a metal detector gate to check the target person. The processing device can then allow the target person to exit through channel 1, achieving rapid security checks and improving throughput efficiency.

[0256] If the first detection result corresponding to the target person meets the preset acquisition conditions, the target person needs to pass through a metal detector gate for security screening. The processing device can acquire the metal detector gate detection result obtained from the metal detector gate detection of the target person. The processing device executes subsequent steps S102-S105, using the second detection result of the detected item to correct the first category information, thereby obtaining the target category information of the detected item. Based on the target category information, it determines whether the detected item carried by the target person is a prohibited item, that is, it performs data fusion decision to determine whether the target person is carrying prohibited items.

[0257] If the items carried by the target person are found to be prohibited, the processing device determines the target person's final detection result as a final danger. Security personnel can guide the target person to the manual security check area for a manual search. After the manual search eliminates the target person from carrying prohibited items and suspicious items that could easily cause false detections by the millimeter-wave security scanner, security personnel will guide the target person through passage 3 between the manual security check area and the millimeter-wave security scanner entrance for a second search. Before the second search, any suspicious items must be removed from the person, and the above detection process must be repeated.

[0258] If the items carried by the target personnel are not contraband, the processing device determines that the target personnel's final test result is safe and allows the target personnel to leave through channel 2 from the exit.

[0259] As can be seen, the two preset fusion detection conditions are set according to the security level requirements of the current scenario. Under higher security requirements, all target personnel are detected as much as possible to improve the accuracy of security checks and thus ensure the safety of people and items in the current scenario. Under lower security requirements, metal detectors are used for fusion detection only when target personnel are carrying suspicious items, to improve the efficiency of security checks and avoid congestion at the entrance of the current scenario.

[0260] In accordance with the above-described security inspection method, this application also provides a security inspection system.

[0261] like Figure 7As shown, a security inspection system includes: a millimeter-wave security scanner 701, a metal detection gate 702, and a processing device (not shown in the figure). Wherein:

[0262] The millimeter-wave security scanner 701 and the metal detector gate 702 are used to detect target personnel. The millimeter-wave security scanner 701 and the metal detector gate 702 are integrated into one physical device, or the millimeter-wave security scanner 701 and the metal detector gate 702 are two independent physical devices.

[0263] The processing device is used to execute the security inspection method provided in the embodiments of this application.

[0264] A security inspection system includes a millimeter-wave security scanner 701, a metal detection gate 702, and a processing device, wherein the millimeter-wave security scanner 701 and the metal detection gate 702 are integrated into a single physical device.

[0265] A security inspection system includes a millimeter-wave security scanner 701, a metal detector gate 702, and a processing device, wherein the millimeter-wave security scanner 701 and the metal detector gate 702 are integrated as two independent physical devices.

[0266] In the technical solution provided in this application embodiment, the processing device can determine a first detection result of the target person based on the millimeter-wave scan image of the target person passing through the millimeter-wave security scanner. The first detection result includes the first position information of the detected item carried by the target person on the millimeter-wave scan image and the first category information of the detected item. Based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, and the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, the device determines the target security zone corresponding to the first position information in the detection area of ​​the metal detector gate. The detection area of ​​the metal detector gate can be divided into multiple security zones. The device obtains a second detection result corresponding to the detected item in the target security zone. The second detection result includes at least one of the metal strength information and the second category information of the detected item in the target security zone. The first category information of the detected item is corrected using the second detection result to determine the target category information of the detected item. The device then determines whether the detected item is a contraband based on the target category information.

[0267] Because metallic objects interfere with the magnetic field between metal detector gates when passing through them, metal detector gates can detect metallic objects more accurately than millimeter-wave security scanners. The second detection result obtained by using the metal detector gate to detect the object corrects the first category information obtained by the millimeter-wave security scanner. The resulting target category information more accurately reflects whether the detected object belongs to any metal category. Therefore, the processing device can more accurately determine whether the detected object is a prohibited metal item, improving the accuracy of the millimeter-wave security scanner in detecting prohibited metal items and overall prohibited items.

[0268] As one embodiment of this application, the millimeter-wave security scanner and the metal detector gate are integrated into a single physical device;

[0269] The processing device is further configured to, when the first detection result meets preset acquisition conditions, control the metal detection gate to start operating, so as to acquire a second detection result corresponding to the detected item in the target zone; or,

[0270] The processing device is also used to control the metal detector gate to work continuously, and to acquire a second detection result corresponding to the detected item in the target zone when the first detection result meets the preset acquisition conditions; wherein, the preset acquisition conditions are used to define the conditions for triggering the acquisition of the second detection result.

[0271] Millimeter-wave security scanners and metal detector gates can be integrated into a single physical device. In this case, the millimeter-wave security scanner operates continuously, while the metal detector gate can either start operating only after the first inspection result of the millimeter-wave security scanner on the target person meets the preset acquisition conditions, or it can operate continuously.

[0272] When the metal detector gate is not continuously operating, the millimeter-wave security scanner of the physical device can detect the target person as they pass through, obtaining the initial detection result.

[0273] The processing device analyzes the first detection result. If the first detection result meets the preset acquisition conditions, the processing device can control the metal detection gate to start detecting the target person and obtain the second detection result of the target person.

[0274] The aforementioned preset acquisition conditions are used to define the conditions for triggering the acquisition of the second detection result. These preset acquisition conditions can be reasonably configured according to actual detection needs. For example, the preset acquisition conditions may include: the first detection result indicates that the category of the detected item is suspicious, or the current security level requirement of the security inspection scenario is the target level and the first detection result indicates that the category of the detected item is safe, etc. These are all reasonable and are not specifically limited here.

[0275] The processing device can determine the first location information of the detected item carried by the target person on the millimeter-wave scan image. Based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, and the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, it determines the target security zone corresponding to the first location information in the detection area of ​​the metal detector gate, and obtains the second detection result corresponding to the detected item in the target security zone. Based on the first detection result and the second detection result, it determines whether the detected item is a contraband.

[0276] When the metal detector gate is not continuously operating, as a target person passes through the physical device, the millimeter-wave security scanner of the physical device can detect the target person and obtain the first detection result. Then, the metal detector gate begins to detect the target person and obtains the second detection result.

[0277] The processing device can analyze the first detection result. If the first detection result meets the preset acquisition conditions, it can determine the first position information of the detected item carried by the target person on the millimeter-wave scanning image. Based on the correspondence between the millimeter-wave scanning image and the detection area of ​​the millimeter-wave security scanner, and the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, it can determine the target security zone corresponding to the first position information in the detection area of ​​the metal detector gate, and acquire the second detection result corresponding to the detected item in the target security zone. Based on the first detection result and the second detection result, it can determine whether the detected item is a contraband.

[0278] As can be seen, in this embodiment, the millimeter-wave security scanner and the metal detector gate can be integrated into a single physical device. When the first detection result meets preset acquisition conditions, the processing device controls the metal detector gate to start operating to obtain a second detection result corresponding to the detected item in the target area; or, it controls the metal detector gate to continue operating and, when the first detection result meets preset acquisition conditions, obtains a second detection result corresponding to the detected item in the target area. The operating mode of the metal detector gate in the physical device can be flexibly set according to actual detection needs, adapting to various application scenarios and improving the efficiency of contraband detection.

[0279] In one embodiment of this application, the millimeter-wave security scanner and the metal detector gate are two independent physical devices, such as... Figure 7 As shown in the embodiment of this application, a security inspection system may further include a first access gate 703;

[0280] The processing device is further configured to, when the first detection result does not meet the preset acquisition conditions and the first detection result indicates that the detected item is not a dangerous category, control the first channel gate to allow the target person passing through the millimeter-wave security scanner to pass through; and when the first detection result meets the preset acquisition conditions, control the first channel gate to allow the target person passing through the millimeter-wave security scanner to enter the detection area of ​​the metal detector gate; wherein, the preset acquisition conditions are used to define the conditions for triggering the acquisition of the second detection result.

[0281] The millimeter-wave security scanner and the metal detector gate are two independent physical devices. When the first detection result does not meet the preset acquisition conditions and the first detection result indicates that the detected item is not of a dangerous category, the processing device controls the first channel gate 703 to allow the target person passing through the millimeter-wave security scanner 701 to pass. When the first detection result meets the preset acquisition conditions, the processing device controls the first channel gate 703 to allow the target person passing through the millimeter-wave security scanner 701 to enter the detection area of ​​the metal detector gate 702, and executes the security inspection method described in any of the above method embodiments to detect whether the detected item carried by the target person is a prohibited item. The preset acquisition conditions are used to define the conditions for triggering the acquisition of the second detection result.

[0282] As can be seen, in this embodiment of the application, a first channel gate can be set up in the security inspection system to facilitate the diversion of personnel based on the first detection result, thereby improving the efficiency and accuracy of security inspection results.

[0283] As one implementation method of this application, such as Figure 7 As shown in the embodiment of this application, a security inspection system may further include a second access gate 704.

[0284] The processing device is also configured to, if it is determined that the target person is carrying prohibited items, control the second gate to allow the target person passing through the metal detector gate to enter the manual security check area; or, if it is determined that the target person is not carrying prohibited items, control the second gate to allow the target person passing through the metal detector gate to pass.

[0285] The processing device can control the second gate 704 to allow the target person to enter the manual security check area if it determines that the target person is carrying prohibited items; or, if it determines that the target person is not carrying prohibited items, control the second gate 704 to allow the target person to pass through the metal detector 702.

[0286] As can be seen, in this embodiment of the application, a second channel gate can be set up in the security inspection system to facilitate the diversion of personnel based on the detection results of the items carried by each person, thereby improving the passage efficiency and the security inspection efficiency and the accuracy of the security inspection results.

[0287] In one implementation, such as Figure 7 As shown, a rotating gate can also be installed at the entrance of the millimeter-wave security scanner 701, and a capture camera can be matched with it. Figure 7 (Not shown in the image) This ensures that only one person passes through at a time, avoiding errors in the detection results caused by multiple target individuals passing through the millimeter-wave security scanner simultaneously.

[0288] In one implementation, an image acquisition device (optical camera) can be installed at the entrance of the metal detection gate 702. Figure 7 (Not shown in the image). The image acquisition device is used to capture images of the target person as they pass through the metal detector gate. The body posture information of the target person in the image can be matched with the body posture information of the target person in the millimeter-wave scan image to determine the area where the same body part of the target person is located as the corresponding area, thereby obtaining the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate.

[0289] In one implementation, both the millimeter-wave security scanner 701 and the metal detector gate 702 are equipped with capture cameras. Before passing through the gate, people must have their faces scanned and then pass through the gate. The facial images captured by the capture cameras at the entrances of the millimeter-wave security scanner 701 and the metal detector gate 702 are used to match and verify the information of the target person, ensuring that the detected person is the same person.

[0290] In addition, based on the security inspection system provided in this application embodiment, first-inspection personnel and second-inspection personnel can be set up at the entrance of the millimeter-wave security gate and the manual security inspection area respectively, so as to make efficient use of the security inspection system provided in this application embodiment and improve security inspection efficiency.

[0291] In accordance with the above-described security inspection method, this application also provides a security inspection device.

[0292] like Figure 8 As shown, a security inspection device includes:

[0293] The acquisition module 801 is used to determine a first detection result of the target person based on the millimeter-wave scan image of the target person passing through the millimeter-wave security scanner, wherein the first detection result includes a first position information of the detected items carried by the target person on the millimeter-wave scan image and a first category information of the detected items;

[0294] The first determining module 802 is used to determine the target security zone corresponding to the first location information in the detection area of ​​the metal detector gate based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, and the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, wherein the detection area of ​​the metal detector gate can be divided into multiple security zones.

[0295] The second determining module 803 is used to obtain a second detection result in the target defense zone corresponding to the detected item, wherein the second detection result includes at least one of the metal strength information in the target defense zone corresponding to the detected item and the second category information of the detected item;

[0296] Correction module 804 is used to correct the first category information of the detected item using the second detection result, and to determine the target category information of the detected item.

[0297] The third determining module 805 is used to determine whether the detected item is a prohibited item based on the target category information.

[0298] In the technical solution provided in this application embodiment, the processing device can determine a first detection result of the target person based on the millimeter-wave scan image of the target person passing through the millimeter-wave security scanner. The first detection result includes the first position information of the detected item carried by the target person on the millimeter-wave scan image and the first category information of the detected item. Based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, and the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, the device determines the target security zone corresponding to the first position information in the detection area of ​​the metal detector gate. The detection area of ​​the metal detector gate can be divided into multiple security zones. The device obtains a second detection result corresponding to the detected item in the target security zone. The second detection result includes at least one of the metal strength information and the second category information of the detected item in the target security zone. The first category information of the detected item is corrected using the second detection result to determine the target category information of the detected item. The device then determines whether the detected item is a contraband based on the target category information.

[0299] Because metallic objects interfere with the magnetic field between metal detector gates when passing through them, metal detector gates can detect metallic objects more accurately than millimeter-wave security scanners. The second detection result obtained by using the metal detector gate to detect the object corrects the first category information obtained by the millimeter-wave security scanner. The resulting target category information more accurately reflects whether the detected object belongs to any metal category. Therefore, the processing device can more accurately determine whether the detected object is a prohibited metal item, improving the accuracy of the millimeter-wave security scanner in detecting prohibited metal items and overall prohibited items.

[0300] As one embodiment of this application, the first determining module 802 can be specifically used for:

[0301] Based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, the second location information corresponding to the first location information in the detection area of ​​the millimeter-wave security scanner is determined;

[0302] Based on the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, the third positional information corresponding to the second positional information in the detection area of ​​the metal detector gate is determined.

[0303] Based on the position of the third location information in the detection area of ​​the metal detector gate, the target defense zone corresponding to the first location information in the detection area of ​​the metal detector gate is determined;

[0304] or

[0305] Based on the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, and the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, the fourth position information corresponding to the preset position point in the detection area of ​​the metal detector gate in the millimeter-wave scan image is determined.

[0306] In the millimeter-wave scan image, based on the fourth position information, an image region that matches the detection area of ​​the metal detector gate is determined;

[0307] Based on the position of the first location information in the image area, the target defense zone corresponding to the first location information in the detection area of ​​the metal detector gate is determined.

[0308] As one embodiment of this application, this application also provides a security inspection device that further includes:

[0309] Human body image acquisition module, used to acquire human body detection images of the target person as they pass through the metal detection gate, captured by the image acquisition device;

[0310] The matching relationship determination module is used to determine the matching relationship between the human body detection image and the millimeter-wave scan image based on the similarity between the body part features included in the human body detection image and the body part features included in the millimeter-wave scan image;

[0311] The first correspondence determination module is used to determine the correspondence between the human body detection image and the detection area of ​​the metal detection gate based on the positional relationship between the image acquisition device and the metal detection gate;

[0312] The second correspondence determination module is used to determine the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, the correspondence between the human body detection image and the detection area of ​​the metal detector gate, and the matching relationship.

[0313] As one embodiment of this application, the correction module 804 is specifically used for:

[0314] Based on the second category information of the detected items in the second test result, the material attention coefficient of the items is determined;

[0315] The first category information of the detected items is corrected using the material attention coefficient of the items to determine the target category information of the detected items;

[0316] or

[0317] A second adjustment parameter corresponding to the target defense zone is determined to adjust the second detection result, and the first category information of the detected item is corrected using the adjusted second detection result to determine the target category information of the detected item;

[0318] or

[0319] A first adjustment parameter corresponding to the first category information of the detected item is determined to adjust the first category information of the detected item, and the adjusted first category information is further corrected using the second detection result to determine the target category information of the detected item.

[0320] or

[0321] Determine the first adjustment parameter corresponding to the first category information of the detected item, and determine the second adjustment parameter corresponding to the target defense zone;

[0322] The first adjustment parameter is used to adjust the first category information of the detected items, and the second adjustment parameter is used to adjust the second detection result;

[0323] The adjusted first category information is further corrected using the adjusted second detection result to determine the target category information of the detected items.

[0324] As one embodiment of this application, the second adjustment parameter includes the sensitivity corresponding to the target defense zone, and the correction module 804 is specifically used for:

[0325] Set the sensitivity corresponding to the target defense zone to the first sensitivity.

[0326] This application embodiment also provides a security inspection device that further includes:

[0327] The setting module is used to set the sensitivity corresponding to the non-target defense zone to a second sensitivity, wherein the first sensitivity is greater than the second sensitivity.

[0328] As one embodiment of this application, the correction module 804 is specifically used for:

[0329] Based on the second category information of the detected items in the second detection result and / or the item category information included in the pre-set millimeter wave whitelist, the category adjustment parameter is determined, wherein the category adjustment parameter represents the degree of attention to item category in the current security inspection scenario; and / or, based on the second category information of the detected items in the second detection result, the item material attention coefficient is determined, wherein the item material attention coefficient represents the degree of attention to metal items of different materials.

[0330] As one embodiment of this application, the first adjustment parameter includes an item material attention coefficient and a category adjustment parameter; the second adjustment parameter includes the sensitivity corresponding to the target defense zone; the first category information includes a first confidence level that the detected item belongs to the item category; and the second detection result includes the metal strength value corresponding to the detected item in the target defense zone; the correction module 804 is specifically used for:

[0331] The overall confidence level P of the detected item belonging to the item category is calculated using the following formula. fin (c) , as target category information for the detected items:

[0332] P fin (c)=M0×α(c)×P mmw (c)+β(m)×A metal (m);

[0333] Where c is the item category number, α(c) is the category adjustment parameter corresponding to item category c, and P mmw (c) represents the first confidence level that the detected item belongs to item category c, M0 is the material attention coefficient corresponding to the detected item, m is the target zone number corresponding to the detection location of the detected item, β(m) is the sensitivity corresponding to target zone m, and A metal (m) represents the metal strength value corresponding to the target defense zone m.

[0334] As one embodiment of this application, the second determining module 803 is specifically used for:

[0335] If the first detection result indicates that the detected item belongs to a suspicious category, or if the security level requirement of the current security inspection scenario is the target level and the first detection result indicates that the detected item belongs to a safe category, then obtain the second detection result corresponding to the detected item in the target security zone.

[0336] In the technical solutions provided in this application, the operations of obtaining, storing, using, processing, transmitting, providing, and disclosing the personal information of the target personnel are all carried out with the authorization of the target personnel.

[0337] It should be noted that the security inspection data in this application embodiment comes from a publicly available dataset.

[0338] This application also provides a processing device, such as... Figure 9 As shown, it includes:

[0339] Memory 901 is used to store computer programs;

[0340] The processor 902, when executing the program stored in the memory 901, implements the security inspection method in any of the above embodiments.

[0341] Furthermore, the aforementioned control device may also include a communication bus and / or a communication interface, with the processor 902, the communication interface, and the memory 901 communicating with each other via the communication bus.

[0342] The communication bus mentioned in the control device above can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not indicate that there is only one bus or one type of bus.

[0343] The communication interface is used for communication between the aforementioned control device and other devices.

[0344] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0345] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0346] In another embodiment provided in this application, a computer-readable storage medium is also provided, which stores a computer program that, when executed by a processor, implements the steps of any of the security inspection methods described above.

[0347] In another embodiment provided in this application, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to perform any of the security check methods described above.

[0348] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a solid-state drive (SSD), etc.

[0349] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0350] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system, apparatus, processing apparatus, computer-readable storage medium, and computer program product embodiments are basically similar to the method embodiments, and therefore the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0351] The above description is merely a preferred embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application are included within the scope of protection of this application.

Claims

1. A security inspection method, characterized in that, The method includes: Based on the millimeter-wave scan image of the target person passing through the millimeter-wave security scanner, a first detection result of the target person is determined, wherein the first detection result includes the first location information of the detected items carried by the target person on the millimeter-wave scan image and the first category information of the detected items; Based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, and the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, the target defense zone corresponding to the first position information in the detection area of ​​the metal detector gate is determined, wherein the detection area of ​​the metal detector gate can be divided into multiple defense zones; Obtain a second detection result corresponding to the detected item in the target defense zone, wherein the second detection result includes at least one of the metal strength information corresponding to the detected item in the target defense zone and the second category information of the detected item; The first category information of the detected items is corrected using the second detection result to determine the target category information of the detected items; The detection of a contraband item is determined based on the target category information.

2. The method according to claim 1, characterized in that, The step of determining the target security zone corresponding to the first location information in the detection area of ​​the metal detector gate based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, and the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, includes: Based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, the second location information corresponding to the first location information in the detection area of ​​the millimeter-wave security scanner is determined; Based on the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, the third positional information corresponding to the second positional information in the detection area of ​​the metal detector gate is determined. Based on the position of the third location information in the detection area of ​​the metal detector gate, the target defense zone corresponding to the first location information in the detection area of ​​the metal detector gate is determined; or The step of determining the target security zone corresponding to the first location information in the detection area of ​​the metal detector gate based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, and the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, includes: Based on the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate, and the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, the fourth position information corresponding to the preset position point in the detection area of ​​the metal detector gate in the millimeter-wave scan image is determined. In the millimeter-wave scan image, based on the fourth position information, an image region that matches the detection area of ​​the metal detector gate is determined; Based on the position of the first location information in the image area, the target defense zone corresponding to the first location information in the detection area of ​​the metal detector gate is determined.

3. The method according to claim 2, characterized in that, The method further includes: Acquire human body detection images of the target person as they pass through the metal detection gate, captured by the image acquisition device; The matching relationship between the human body detection image and the millimeter-wave scan image is determined based on the similarity between the body part features included in the human body detection image and the body part features included in the millimeter-wave scan image. Based on the positional relationship between the image acquisition device and the metal detection gate, the correspondence between the human body detection image and the detection area of ​​the metal detection gate is determined; Based on the correspondence between the millimeter-wave scan image and the detection area of ​​the millimeter-wave security scanner, the correspondence between the human body detection image and the detection area of ​​the metal detector gate, and the matching relationship, the positional correspondence between the detection area of ​​the millimeter-wave security scanner and the detection area of ​​the metal detector gate is determined.

4. The method according to claim 1, characterized in that, The step of correcting the first category information of the detected item using the second detection result to determine the target category information of the detected item includes: Based on the second category information of the detected items in the second test result, the material attention coefficient of the items is determined; The first category information of the detected items is corrected using the material attention coefficient of the items to determine the target category information of the detected items; or A second adjustment parameter corresponding to the target defense zone is determined to adjust the second detection result, and the first category information of the detected item is corrected using the adjusted second detection result to determine the target category information of the detected item; or A first adjustment parameter corresponding to the first category information of the detected item is determined to adjust the first category information of the detected item, and the adjusted first category information is further corrected using the second detection result to determine the target category information of the detected item. or Determine the first adjustment parameter corresponding to the first category information of the detected item, and determine the second adjustment parameter corresponding to the target defense zone; The first adjustment parameter is used to adjust the first category information of the detected items, and the second adjustment parameter is used to adjust the second detection result; The adjusted first category information is further corrected using the adjusted second detection result to determine the target category information of the detected items.

5. The method according to claim 4, characterized in that, The second adjustment parameter includes the sensitivity corresponding to the target defense zone, and the step of determining the second adjustment parameter corresponding to the target defense zone includes: Set the sensitivity corresponding to the target defense zone to the first sensitivity. The method further includes: Set the sensitivity corresponding to the non-target defense zone to the second sensitivity, wherein the first sensitivity is greater than the second sensitivity.

6. The method according to claim 4, characterized in that, The first adjustment parameter includes an item material attention coefficient and / or a category adjustment parameter. The step of determining the first adjustment parameter corresponding to the first category information of the detected item includes: Based on the second category information of the detected items in the second detection result and / or the item category information included in the pre-set millimeter wave whitelist, the category adjustment parameter is determined, wherein the category adjustment parameter represents the degree of attention paid to item category in the current security inspection scenario; And / or, Based on the second category information of the detected items in the second detection result, the material attention coefficient of the items is determined, wherein the material attention coefficient of the items represents the degree of attention to metal items of different materials.

7. The method according to claim 4, characterized in that, The first adjustment parameter includes the material attention coefficient and the category adjustment parameter; the second adjustment parameter includes the sensitivity corresponding to the target defense zone; the first category information includes the first confidence level that the detected item belongs to the item category; and the second detection result includes the metal strength value corresponding to the detected item in the target defense zone. The first adjustment parameter corresponding to the first category information of the detected item is determined, and the second adjustment parameter corresponding to the target defense zone is determined; the first adjustment parameter is used to adjust the first category information of the detected item, and the second adjustment parameter is used to adjust the second detection result; The step of further refining the adjusted first category information using the adjusted second detection result to determine the target category information of the detected items includes: The comprehensive confidence P that the detected item belongs to the item category is calculated according to the following formula fin (c) as the target category information of the detected item: P fin (c)=M0×α(c)×P mmw (c)+β(m)×A metal (m); Where c is the item category number, α(c) is the category adjustment parameter corresponding to item category c, and P mmw (c) represents the first confidence level that the detected item belongs to item category c, M0 is the material attention coefficient corresponding to the detected item, m is the target zone number corresponding to the detection location of the detected item, β(m) is the sensitivity corresponding to target zone m, and A metal (m) represents the metal strength value corresponding to the target defense zone m.

8. The method according to any one of claims 1-7, characterized in that, The step of obtaining the second detection result corresponding to the detected item in the target defense zone includes: If the first detection result indicates that the detected item belongs to a suspicious category, or if the security level requirement of the current security inspection scenario is the target level and the first detection result indicates that the detected item belongs to a safe category, then obtain the second detection result corresponding to the detected item in the target security zone.

9. A security inspection system, characterized in that, The system includes: a millimeter-wave security scanner, a metal detection gate, and a processing device, wherein: the millimeter-wave security scanner and the metal detection gate are used to detect target personnel respectively; the millimeter-wave security scanner and the metal detection gate are integrated into a single physical device, or the millimeter-wave security scanner and the metal detection gate are two independent physical devices; The processing device is used to perform the method according to any one of claims 1-8.

10. The system according to claim 9, characterized in that, The millimeter-wave security scanner and the metal detector gate are integrated into a single physical device; The processing device is further configured to, when the first detection result meets preset acquisition conditions, control the metal detection gate to start working, so as to acquire a second detection result in the target zone corresponding to the detected item; or, The processing device is also used to control the metal detection gate to work continuously, and to acquire a second detection result corresponding to the detected item in the target zone when the first detection result meets the preset acquisition conditions; The preset acquisition conditions are used to define the conditions for triggering the acquisition of the second detection result.

11. The system according to claim 9, characterized in that, The millimeter-wave security scanner and the metal detector gate are two independent physical devices, and the system also includes a first channel gate; The processing device is further configured to control the first channel gate to allow the target personnel passing through the millimeter-wave security scanner when the first detection result does not meet the preset acquisition conditions and the first detection result indicates that the detected item is not a dangerous category; If the first detection result meets the preset acquisition conditions, the first channel gate is controlled to allow the target person passing through the millimeter-wave security scanner to enter the detection area of ​​the metal detection gate; The preset acquisition conditions are used to define the conditions for triggering the acquisition of the second detection result.

12. The system according to any one of claims 9-11, characterized in that, The system also includes a second channel gate; The processing device is further configured to, if it is determined that the target person is carrying prohibited items, control the second gate to allow the target person passing through the metal detector gate to enter the manual security check area; or, if it is determined that the target person is not carrying prohibited items, control the second gate to allow the target person passing through the metal detector gate to pass.

Images