Article security inspection method, device, computer equipment and storage medium

Abstract

The present application relates to a method, device, computer equipment and storage medium for item security inspection. The method includes: obtaining an image to be inspected corresponding to the item to be inspected, the image to be inspected carries binding identity information and a transport sequence identifier; inputting the image to be inspected into a trained target item detection model, performing item category detection on the image to be inspected through the trained target item detection model, and obtaining an item category detection result; obtaining a security inspection result according to the item category in the item category detection result; when the security inspection result is that there is a target prohibited item, generating an interception prompt according to the target prohibited item and the binding identity information, and determining the target prompt terminal according to the transport sequence identifier; and pushing the interception prompt to the target prompt terminal. The use of this method can improve security inspection efficiency and enhance security inspection results.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method, apparatus, computer device and storage medium for checking the safety of an item. Background Art

[0002] With the development of computer technology, security inspection technology has emerged. Security inspection technology refers to the security inspection of items to detect whether there are prohibited items and deal with them. Security inspection technology can be applied to logistics transfer sites, airports and other areas.

[0003] In traditional technology, security inspection is mainly carried out by using a terminal to scan the object to be inspected and displaying a security inspection image corresponding to the object to be inspected, so that the staff at the terminal can make a judgment based on the security inspection image, obtain the judgment result and then feed it back for processing.

[0004] However, the traditional method relies on staff to make judgments based on the displayed security inspection images and then process them, which has the problems of low security inspection efficiency and poor security inspection effect. Summary of the invention

[0005] Based on this, it is necessary to provide an article security inspection method, device, computer equipment and storage medium that can improve security inspection efficiency and enhance security inspection results in response to the above technical problems.

[0006] A method for inspecting the safety of an article, the method comprising:

[0007] Acquire an image to be inspected corresponding to the item to be inspected, the image to be inspected carrying binding identity information and a transport sequence identifier;

[0008] The image to be inspected is input into the trained target object detection model, and the trained target object detection model is used to detect the object category of the image to be inspected to obtain the security inspection result;

[0009] When the security check result shows that there are target prohibited items, an interception prompt is generated according to the target prohibited items and the bound identity information, and the target prompt terminal is determined according to the transport sequence identifier;

[0010] Push the interception prompt to the target prompt terminal.

[0011] In one embodiment, obtaining an image to be inspected corresponding to the object to be inspected includes:

[0012] Obtain the transport sequence identification corresponding to the item to be inspected and the first identity information of the item before the security inspection scan;

[0013] Perform security inspection scans on the items to be inspected, and obtain scanned images corresponding to the security inspection scans;

[0014] According to the corresponding transport sequence identifier, obtain the second identity information of the item after the security inspection scan;

[0015] The second identity information is verified according to the first identity information, the scanned image is bound to a code according to the identity verification result, and the corresponding transport sequence identifier is associated to obtain the image to be inspected.

[0016] In one embodiment, performing identity verification on the second identity information according to the first identity information, binding the scanned image with a code according to the identity verification result, and associating the corresponding transport sequence identifier to obtain the image to be inspected includes:

[0017] Comparing the first identity information with the second identity information;

[0018] When the second identity information corresponds to the first identity information, the second identity information and the scanned image are bound and associated with the corresponding transport sequence identifier to obtain the image to be inspected.

[0019] In one embodiment, the image to be inspected is input into a trained target object detection model, and the trained target object detection model is used to detect the object category of the image to be inspected. Before obtaining the security inspection result, the method further includes:

[0020] Obtain a training image set with object image annotations and an initial object detection model;

[0021] The initial object detection model is trained according to the training image set to obtain a trained target object detection model.

[0022] In one embodiment, the image to be inspected is input into a trained target object detection model, and the trained target object detection model is used to detect the object category of the image to be inspected. The security inspection result includes:

[0023] The image to be inspected is input into the trained target object detection model, and the trained target object detection model is used to detect the object category of the image to be inspected, thereby obtaining the object category detection result;

[0024] When the item category in the item category detection result includes prohibited items, obtaining the category confidence corresponding to the prohibited items;

[0025] When the category confidence is greater than or equal to the preset confidence threshold, the security check result is obtained as the presence of the target prohibited item;

[0026] When the category confidence is less than the confidence threshold, a security inspection image judgment page is generated and displayed according to the image to be inspected and the prohibited items, an image judgment result corresponding to the security inspection image judgment page is received, and a security inspection result is obtained according to the image judgment result.

[0027] In one embodiment, determining the target prompt terminal according to the transport sequence identifier includes:

[0028] Get the transport sequence information of each transport line;

[0029] According to the transport sequence information and the transport sequence identifier, the target transport line and the target prompt terminal on the target transport line corresponding to the sorting operation are determined.

[0030] In one embodiment, after the interception prompt is pushed to the target prompt terminal, the method further includes:

[0031] Receive an interception feedback result corresponding to the interception prompt;

[0032] When the interception feedback result is an interception error, the trained target object detection model is trained according to the object detection result in the interception feedback result and the image to be inspected for security.

[0033] An article safety inspection device, comprising:

[0034] An acquisition module is used to acquire an image to be inspected corresponding to the item to be inspected, the image to be inspected carrying binding identity information and a transport sequence identifier;

[0035] A detection module is used to input the image to be inspected into the trained target object detection model, and perform object category detection on the image to be inspected through the trained target object detection model to obtain the security inspection result;

[0036] A processing module, for generating an interception prompt according to the target prohibited object and the bound identity information when the security check result shows that the target prohibited object exists, and determining the target prompt terminal according to the transport sequence identifier;

[0037] The push module is used to push the interception prompt to the target prompt terminal.

[0038] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0039] Acquire an image to be inspected corresponding to the item to be inspected, the image to be inspected carrying binding identity information and a transport sequence identifier;

[0040] The image to be inspected is input into the trained target object detection model, and the trained target object detection model is used to detect the object category of the image to be inspected to obtain the security inspection result;

[0041] When the security check result shows that there are target prohibited items, an interception prompt is generated according to the target prohibited items and the bound identity information, and the target prompt terminal is determined according to the transport sequence identifier;

[0042] Push the interception prompt to the target prompt terminal.

[0043] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0044] Acquire an image to be inspected corresponding to the item to be inspected, the image to be inspected carrying binding identity information and a transport sequence identifier;

[0045] The image to be inspected is input into the trained target object detection model, and the trained target object detection model is used to detect the object category of the image to be inspected to obtain the security inspection result;

[0046] When the security check result shows that there are target prohibited items, an interception prompt is generated according to the target prohibited items and the bound identity information, and the target prompt terminal is determined according to the transport sequence identifier;

[0047] Push the interception prompt to the target prompt terminal.

[0048] The above-mentioned article security inspection method, device, computer equipment and storage medium, after obtaining the image to be inspected corresponding to the article to be inspected, first input the image to be inspected into the trained target object detection model, and perform object category detection on the image to be inspected through the trained target object detection model, so as to achieve accurate object category recognition of the image to be inspected, improve the security inspection effect, and the security inspection result. When the security inspection result is that there are target prohibited items, an interception prompt is generated according to the target prohibited items and the bound identity information, and the target prompt terminal is determined according to the transport sequence identifier, and the interception prompt is pushed to the target prompt terminal to complete the automated security inspection, which can improve the security inspection efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A schematic diagram of a safety inspection method in one embodiment;

[0050] Figure 2 An application environment diagram of a security inspection method in an embodiment;

[0051] Figure 3 is an application environment diagram of a security inspection method in another embodiment;

[0052] Figure 4 A schematic flow chart of a safety inspection method in another embodiment;

[0053] Figure 5 is a structural block diagram of a safety inspection device in one embodiment;

[0054] Figure 6 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0056] In one embodiment, Figure 1 As shown, a method for checking item security is provided. This embodiment uses the method applied to a terminal as an example for illustration. It is understandable that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0057] Step 102: Acquire an image corresponding to the item to be inspected, where the image carries binding identity information and a transport sequence identifier.

[0058] Among them, items to be inspected for security refer to items that need to be inspected for security. For example, items to be inspected for security may specifically refer to parcels in a logistics transfer yard. For another example, items to be inspected for security may specifically refer to luggage that needs to be inspected for security at an airport. An image to be inspected for security refers to an image obtained by scanning and corresponding to the item to be inspected for security. For example, an image to be inspected for security may specifically refer to an X-ray image obtained by scanning and corresponding to the item to be inspected for security. Bound identity information refers to identity information used to represent different items to be inspected for security, for example, bound identity information may specifically refer to a barcode corresponding to the item to be inspected for security, and the item to be inspected for security may specifically be a parcel. A transport sequence identifier refers to an identifier used to identify the current transport location of the item to be inspected for security, and the transport sequence identifier of each item to be inspected for security is unique.

[0059] Specifically, when a security check is required, the terminal will obtain an image corresponding to the item to be checked, and the image carries the binding identity information and the transport sequence identifier.

[0060] Step 104 , input the image to be inspected for security into the trained target object detection model, perform object category detection on the image to be inspected for security by the trained target object detection model, and obtain a security inspection result.

[0061] Among them, the trained target item detection model refers to a pre-trained model for detecting the item category of the security inspection image. For example, the trained target item detection model can be a model built based on the R-CNN (R-Convolutional Neural Networks) network. For another example, the trained target item detection model can be a model built based on the YOLO (You Only Look Once) network. In this embodiment, the target item detection model is not specifically limited, as long as the item category detection can be achieved. Item category detection refers to the detection of the item category present in the security inspection image. For example, common item categories include: lighters, mobile phones, milk powder, cosmetics, wine, heating stickers, dry batteries / lithium batteries, compressed gas tanks, knives, water / unknown liquids, powdered objects, tablets, guns, toy guns, iron shell lighters, laptops, button batteries, scissors, power banks, explosives, folding umbrellas, etc. The security inspection result refers to the result of the security inspection of the security inspection items, including the presence of target prohibited items and the absence of target prohibited items.

[0062] Specifically, after obtaining the image to be inspected for security, the terminal will input the image to be inspected for security into the trained target item detection model, and perform item category detection on the image to be inspected for security by the trained target item detection model, and obtain the item category in the image to be inspected for security and the category confidence corresponding to the item category, that is, the item category detection result, and further obtain the security inspection result according to the item category detection result. Furthermore, when performing item category detection on the image to be inspected for security by the trained target item detection model, the trained target item detection model will first perform target recognition on the image to be inspected for security, identify the area where the item may exist in the image to be inspected for security, and then perform category detection on the items in the area, and obtain the confidence that the item belongs to each preset item category, sort the confidence, and determine the item category corresponding to the item in the area and the confidence corresponding to the item category. Among them, the preset item category can be set as needed, for example, the preset item category can specifically be the common item category mentioned above.

[0063] It should be noted that, in order to improve the efficiency of security inspection, in this embodiment, the object category detection can also be performed in a parallel processing manner. For example, the parallel processing manner can specifically create multiple object category detection tasks at the same time, and use multiple trained target object detection models to perform object category detection on multiple images to be inspected for security inspection, respectively, to obtain object category detection results.

[0064] Furthermore, after obtaining the item category detection result, the terminal will preset prohibited items according to the item category comparison in the item category detection result to determine whether there are prohibited items in the item category. When there are no prohibited items in the item category, the security check result can be obtained that there are no target prohibited items. When there are prohibited items in the item category, the terminal needs to further obtain the category confidence corresponding to the prohibited item and obtain the security check result based on the category confidence.

[0065] Step 108: When the security check result shows that there is a target prohibited object, an interception prompt is generated according to the target prohibited object and the bound identity information, and the target prompt terminal is determined according to the transport sequence identifier.

[0066] Among them, the interception prompt refers to a prompt that instructs the staff to intercept the items to be inspected. The target prompt terminal refers to the terminal that the staff who can intercept the items to be inspected need to use. For example, the target prompt terminal can specifically refer to the branch line scanning and trunk loading scanning equipment in the logistics transfer yard.

[0067] Specifically, when the security check result shows that there are target prohibited items, it means that the items to be inspected need to be intercepted. The terminal will generate an interception prompt based on the target prohibited items and the bound identity information, and locate the current transportation position of the items to be inspected according to the transportation sequence identifier, and determine the target prompt terminal to which the interception prompt needs to be pushed based on the current transportation position.

[0068] Step 110: Push the interception prompt to the target prompt terminal.

[0069] Specifically, the terminal will push an interception prompt to the target prompt terminal, and the staff using the target prompt terminal will receive the interception prompt. When scanning the items to be inspected using the target prompt terminal, the staff will know through the interception prompt that the items to be inspected may contain target contraband, and intercept the items to be inspected. It should be noted that in addition to pushing the interception prompt to the target prompt terminal, the terminal can also control the preset business system to physically divert and intercept the items to be inspected including the target contraband, that is, by outputting control instructions to the preset business system, the items to be inspected including the target contraband are directly diverted to the preset exception handling area. Among them, the preset business system is used to control the sorting and diversion of the items to be inspected.

[0070] The above-mentioned method for security inspection of items, after obtaining the image to be inspected corresponding to the item to be inspected, first inputs the image to be inspected into a trained target item detection model, and performs item category detection on the image to be inspected through the trained target item detection model, so as to realize accurate item category recognition of the image to be inspected, improve the security inspection effect, and the security inspection result. When the security inspection result is that there are target prohibited items, an interception prompt is generated according to the target prohibited items and the bound identity information, and the target prompt terminal is determined according to the transport sequence identifier, and the interception prompt is pushed to the target prompt terminal to complete the automated security inspection, which can improve the security inspection efficiency.

[0071] In one embodiment, obtaining an image to be inspected corresponding to the object to be inspected includes:

[0072] Obtain the transport sequence identification corresponding to the item to be inspected and the first identity information of the item before the security inspection scan;

[0073] Perform security inspection scans on the items to be inspected, and obtain scanned images corresponding to the security inspection scans;

[0074] According to the corresponding transport sequence identifier, obtain the second identity information of the item after the security inspection scan;

[0075] The second identity information is verified according to the first identity information, the scanned image is bound to a code according to the identity verification result, and the corresponding transport sequence identifier is associated to obtain the image to be inspected.

[0076] Among them, the scanned image refers to the corresponding image obtained by scanning the item to be inspected for security. For example, the scanned image may specifically refer to the X-ray image corresponding to the item to be inspected for security obtained by scanning. The first identity information refers to the identity information obtained before the item is scanned for security inspection. For example, the first identity information may specifically refer to the barcode information used to identify the item to be inspected for security. The barcode information may specifically refer to the waybill information of the package, including one-dimensional barcode information and two-dimensional barcode information. The second identity information refers to the identity information obtained after the item is scanned for security inspection. For example, the second identity information may specifically refer to the barcode information used to identify the item to be inspected for security. The barcode information may specifically refer to the waybill information of the package, including one-dimensional barcode information and two-dimensional barcode information.

[0077] Specifically, when it is necessary to obtain the image to be inspected for security, the terminal will obtain the transport sequence identifier corresponding to the item to be inspected for security and the first identity information before the security inspection scan of the item, perform a security inspection scan on the item to be inspected, obtain a scanned image corresponding to the security inspection scan, establish a correspondence between the scanned image and the first identity information, cache the scanned image and the first identity information in a preset scanned image queue according to the correspondence, obtain the second identity information after the security inspection scan of the item, perform an identity verification on the second identity information according to the first identity information to confirm whether the second identity information is correct, and when the second identity information passes the verification, bind the second identity information and the scanned image, and associate the corresponding transport sequence identifier to obtain the image to be inspected for security. Whenever the second identity information fails to pass the verification, obtain other first identity information from the preset scanned image queue, perform an identity verification on the second identity information according to the other first identity information, obtain an identity verification result, bind the scanned image to a code according to the identity verification result, and associate the corresponding transport sequence identifier to obtain the image to be inspected for security.

[0078] For example, this embodiment is explained through an application scenario, in which the terminal is a security inspection machine. During the security inspection process, the items to be inspected are transmitted in real time on the conveyor belt, and will pass through the first scanning device, the terminal, and the second scanning device in sequence. When passing through the first scanning device, the first scanning device will obtain the first identity information corresponding to the items to be inspected and output it to the security inspection machine. When passing through the security inspection machine, the security inspection machine will scan the items to be inspected to obtain a scanned image, and obtain a scanned image queue based on the scanned image, the first identity information corresponding to the scanned image, and the transport sequence identifier, wherein the transport sequence identifier is collected by the collection device set on the conveyor belt. When passing through the second scanning device, the second scanning device will obtain the second identity information corresponding to the items to be inspected and output it to the security inspection machine. When the security inspection machine receives the second identity information output by the second scanning device, it indicates that a code binding verification is required. The security inspection machine will perform an identity verification on the second identity information based on the first identity information, perform code binding on the scanned image based on the identity verification result, and associate the corresponding transport sequence identifier to obtain the image to be inspected.

[0079] In this embodiment, by obtaining the transport sequence identification, the first identity information, the scanned image and the second identity information, the second identity information can be verified according to the first identity information, and the scanned image can be bound to the code after the identity verification, and associated with the corresponding transport sequence identification to obtain the image to be inspected for security.

[0080] In one embodiment, performing identity verification on the second identity information according to the first identity information, binding the scanned image with a code according to the identity verification result, and associating the corresponding transport sequence identifier to obtain the image to be inspected includes:

[0081] Comparing the first identity information with the second identity information;

[0082] When the second identity information corresponds to the first identity information, the second identity information and the scanned image are bound and associated with the corresponding transport sequence identifier to obtain the image to be inspected.

[0083] Specifically, the terminal will compare the first identity information with the second identity information. When the second identity information corresponds to the first identity information, it means that the identity verification has passed. The terminal will directly bind the second identity information and the scanned image, and associate the corresponding transportation sequence identifier to obtain the image to be inspected. When the second identity information does not correspond to the first identity information, it means that the identity verification has failed. The terminal will obtain other first identity information from the preset scanned image queue and continue to verify the second identity information based on the other first identity information.

[0084] Among them, the scanned image queue includes the scanned image and the first identity information corresponding to the scanned image. When the first identity information does not correspond to the second identity information, the terminal needs to further obtain other first identity information except the first identity information from the scanned image queue, compare the other first identity information and the second identity information, determine the matching first identity information corresponding to the second identity information, perform image code binding according to the corresponding relationship between the matching first identity information and the scanned image, and associate the corresponding transport sequence identifier to obtain the image to be inspected for security. Further, when the matching first identity information corresponds to the scanned image of the unbound identity information, it means that the second identity information at this time is available and the identity verification passes. The terminal will bind the second identity information to the scanned image of the unbound identity information, associate the corresponding transport sequence identifier to obtain the image to be inspected for security. When the matching first identity information corresponds to the scanned image of the bound identity information, it means that the second identity information at this time is not available and the identity verification fails. The terminal will obtain new first identity information and second identity information again, and perform identity verification on the new second identity information according to the new first identity information.

[0085] It should be noted that when the matching first identity information corresponds to the scanned image of the unbound identity information, it means that the second identity information corresponds to the scanned image of the unbound identity information, and in the scanned image queue, the first identity information corresponding to the scanned image that entered the queue before the scanned image corresponding to the matching first identity information is redundant data and needs to be removed. The terminal will bind the second identity information and the scanned image of the unbound identity information to obtain the image to be inspected for security, and remove the first identity information before the matching first identity information, as well as the scanned image corresponding to the first identity information before the first identity information is matched, and update the scanned image queue.

[0086] In this embodiment, by comparing the first identity information with the second identity information, when the second identity information corresponds to the first identity information, the second identity information and the scanned image are bound, and the corresponding transport sequence identifier is associated to obtain the image to be inspected, thereby achieving the acquisition of the image to be inspected.

[0087] In one embodiment, the image to be inspected is input into a trained target object detection model, and the trained target object detection model is used to perform object detection on the image to be inspected, and before obtaining the object detection result, the method further includes:

[0088] Obtain a training image set with object image annotations and an initial object detection model;

[0089] The initial object detection model is trained according to the training image set to obtain a trained target object detection model.

[0090] The training image set includes multiple training images, and the object image annotation is used to annotate the object images on the training images, including the positions of the object images on the training images and the object categories corresponding to the object images. The initial object detection model refers to a detection model that has not yet been parameter adjusted.

[0091] Specifically, before performing object detection, the terminal will obtain a training image set carrying object image annotations and an initial object detection model, input the training images in the training image set into the initial object detection model in sequence, obtain the detection results corresponding to the training images, calculate the model loss function by comparing the detection results and the object image annotations carried by the training images, adjust the parameters of the initial object detection model according to the model loss function, return to the step of inputting the training images into the initial object detection model in sequence, and obtain the detection results corresponding to the training images, until the model loss function meets the preset loss function requirements, and obtain the trained target object detection model. Among them, the preset loss function requirements can be set as needed, for example, the preset loss function requirements can be that the loss function value is less than the preset loss function threshold or the loss function converges, etc. This embodiment is not specifically limited here.

[0092] It should be noted that in this embodiment, when the terminal obtains the training image set with the image annotations of the carried items, because the main detection in this application is the scanned image, such as the X-ray image, considering the occlusion of the X-ray image and the color difference with the traditional image, when obtaining the training image set, it is necessary to pre-process the training image by means of image enhancement, etc., so as to improve the accuracy of the trained target object detection model. Furthermore, after obtaining the trained target object detection model, the terminal can also optimize the trained target object detection model through various methods such as Tensor RT acceleration and lightweight network design to further improve the security inspection efficiency.

[0093] In this embodiment, by training the initial object detection model according to the training image set, a trained target object detection model can be obtained.

[0094] In one embodiment, the image to be inspected is input into a trained target object detection model, and the trained target object detection model is used to detect the object category of the image to be inspected. The security inspection result includes:

[0095] The image to be inspected is input into the trained target object detection model, and the trained target object detection model is used to detect the object category of the image to be inspected, thereby obtaining the object category detection result;

[0096] When the item category in the item category detection result includes prohibited items, obtaining the category confidence corresponding to the prohibited items;

[0097] When the category confidence is greater than or equal to the preset confidence threshold, the security check result is obtained as the presence of the target prohibited item;

[0098] When the category confidence is less than the confidence threshold, a security inspection image judgment page is generated and displayed according to the image to be inspected and the prohibited items, an image judgment result corresponding to the security inspection image judgment page is received, and a security inspection result is obtained according to the image judgment result.

[0099] Among them, category confidence refers to the confidence output by the trained target object detection model that the image of the object in the image to be inspected belongs to a certain item category. The preset confidence threshold is used to judge whether the category confidence is credible. For example, the confidence threshold can be realized by pre-setting category management information, and the category management information includes pre-set land transport confidence threshold, air transport confidence threshold, type confidence threshold, etc. The security inspection image judgment page refers to the page for security inspectors to judge whether there are target prohibited items in the image to be inspected, and the image judgment result refers to the judgment result fed back by the security inspector. For example, the image judgment result can be one of qualified and prohibited.

[0100] Specifically, when the item category includes prohibited items, the terminal will obtain the category confidence corresponding to the prohibited items, and use the category confidence to confirm whether the target prohibited items actually exist in the image to be inspected for security. When the category confidence is greater than or equal to the preset confidence threshold, it means that the category confidence is credible, and the terminal can trust the item category detection result, and the security inspection result is that the target prohibited items exist. When the category confidence is less than the confidence threshold, it means that the category confidence is unreliable, and the terminal needs to further generate and display a security inspection image judgment page based on the image to be inspected for security and the prohibited items, so that the security inspector can confirm whether the target prohibited items actually exist in the image to be inspected for security based on the displayed security inspection image judgment page, and feedback the image judgment result corresponding to the security inspection image judgment page. After receiving the image judgment result, the terminal will obtain the security inspection result based on the image judgment result. Among them, when the image judgment result is qualified, it means that there are no target prohibited items in the image to be inspected for security, and the terminal can obtain the security inspection result that there are no target prohibited items. When the image judgment result is prohibited, it means that there are target prohibited items in the image to be inspected for security, and the terminal can obtain the security inspection result that there are target prohibited items.

[0101] Furthermore, the confidence threshold in this embodiment may correspond to the mode of transportation and the type of article, that is, different confidence thresholds may be set for different modes of transportation (such as land transportation, air transportation, water transportation, etc.) and different categories of articles. Therefore, after obtaining the category confidence corresponding to the prohibited items, it is necessary to determine the corresponding preset confidence threshold according to the mode of transportation of the items to be inspected and the category of the prohibited items, so as to improve the efficiency of security inspection. In addition, when the category confidence is less than the confidence threshold, when generating and displaying the security inspection image judgment page according to the image to be inspected and the prohibited items, the terminal will also refer to the pre-set flow direction screening information. The flow direction screening information may include a focus flow direction and a designated flow direction, etc. Among them, the focus flow direction refers to the flow direction that needs to be focused on. For the items to be inspected in this flow direction, a focus prompt will be generated in the security inspection image judgment page at the same time. The designated flow direction refers to the pre-set flow direction that needs to be processed, that is, only the images to be inspected corresponding to the items to be inspected in this flow direction need further judgment by the security inspector, and the security inspection result can be obtained directly through the confidence threshold judgment for the images to be inspected corresponding to the items to be inspected outside this flow direction. For example, when flow direction A is the designated flow direction, and the item category of the images to be inspected corresponding to the items to be inspected corresponding to flow direction B includes prohibited items, and the category confidence corresponding to the prohibited items is less than the confidence threshold, the security inspection result can be directly obtained as the absence of the target prohibited items.

[0102] In this embodiment, by judging whether there are target prohibited items on the image to be inspected based on the category confidence, the confidence threshold and the image judgment result, an accurate judgment can be made, thereby obtaining an accurate security inspection result.

[0103] In one embodiment, determining the target prompt terminal according to the transport sequence identifier includes:

[0104] Get the transport sequence information of each transport line;

[0105] According to the transport sequence information and the transport sequence identifier, the target transport line and the target prompt terminal on the target transport line corresponding to the sorting operation are determined.

[0106] The transport sequence information refers to the transport sequence identification information of the items to be inspected for security that are transported on the transport line, and the target transport line refers to the transport line that is transporting the items to be inspected for security that include the target contraband.

[0107] Specifically, the terminal will obtain the transport sequence information of each transport line, compare the transport sequence identifier and the transport sequence information, determine the target transport line that is transporting the items to be inspected corresponding to the transport sequence identifier, and determine the target prompt terminal corresponding to the sorting operation on the target transport line based on the preset relationship between the target transport line and the terminal. It should be noted that a collection device is provided on the transport line, and the collection device can be used to collect the transport sequence identifier. For example, the collection device can be a photoelectric sensor.

[0108] In this embodiment, by acquiring the transport sequence information of each transport line, the target transport line and the target prompt terminal corresponding to the sorting operation on the target transport line can be determined according to the transport sequence information and the transport sequence identifier.

[0109] In one embodiment, after the interception prompt is pushed to the target prompt terminal, the method further includes:

[0110] Receive an interception feedback result corresponding to the interception prompt;

[0111] When the interception feedback result is an interception error, the trained target object detection model is trained according to the object detection result in the interception feedback result and the image to be inspected for security.

[0112] Specifically, after completing the interception, the terminal will receive an interception feedback result corresponding to the interception prompt. When the interception feedback result is an interception error, it means that the accuracy of the trained target object detection model is not high. The terminal will mark the image to be inspected for security according to the object detection result in the interception feedback result, and train the trained target object detection model according to the marked image to be inspected for security, so as to improve the accuracy of the trained target object detection model. Preferably, the terminal will train the trained target object detection model only after receiving a preset number of interception feedback results of interception errors, wherein the preset number can be set as needed.

[0113] In this embodiment, after receiving the interception feedback result corresponding to the interception prompt, when the interception feedback result is an interception error, the trained target object detection model is trained according to the object detection result in the interception feedback result and the image to be inspected for security, so that an accurate trained target object detection model can be obtained.

[0114] In one embodiment, Figure 2 As shown, an application scenario diagram is provided to illustrate the article security inspection method of the present application, wherein the terminal includes a security inspection machine, a security inspection machine information interaction module, a core module, an image recognition module, an AI (Artificial Intelligence) recognition service and an interception module. The application of the article security inspection method in this application scenario is as follows:

[0115] The security inspection machine obtains the image code data (i.e. the image to be inspected for security), and uploads the image code data to the information interaction module of the security inspection machine according to the protocol. After receiving the image code data, the information interaction module of the security inspection machine outputs the image code data and creates an image judgment task to the core module. After receiving the image judgment task, the core module takes out a certain number of image judgment tasks from the image recognition module for image judgment. The image judgment task uses the image code data as a parameter to call the view interface in the AI ​​recognition service, and enables the AI ​​recognition service to return the image recognition result (i.e. the image to be inspected for security is input into the trained target object detection model, and the trained target object detection model is used to perform object category detection on the image to be inspected for security, to obtain the security inspection result). After receiving the image recognition result, the image recognition module feeds back the image recognition result to the core module, and the core module saves the image recognition result to the image judgment queue PendingTask, and performs the next step of processing according to the interception model.

[0116] When the interception mode is direct interception and the security check result shows that there are target prohibited items, an interception prompt is generated based on the target prohibited items and the bound identity information, and the target prompt terminal is determined based on the transport sequence identifier, and the interception prompt is pushed to the target prompt terminal.

[0117] When the interception mode is a hybrid interception mode (i.e., a hybrid interception mode of AI + manual), the image code data is fed back to the security inspector for judgment, the image judgment result of the security inspector is received, the image judgment result is saved to FinishTask, the status task in the data storage module is updated, and the security inspection result is obtained according to the image judgment result. When the security inspection result is that there are target prohibited items, an interception prompt is generated according to the target prohibited items and the bound identity information, and the target prompt terminal is determined according to the transport sequence identifier, and the interception prompt is pushed to the target prompt terminal. It should be noted that when the interception mode is a hybrid interception mode, the security inspector can pre-select the security inspection machine that needs to be judged. The core module will filter the image judgment queue PendingTask according to the security inspector's selection, and only push the image code data on the security inspection machine that needs to be judged selected by the security inspector to the display interface.

[0118] In one embodiment, Figure 3 As shown, an application scenario diagram of processing prohibited packages in a logistics transfer station is provided to illustrate the article security inspection method of the present application. The terminal includes a security inspection machine, a prohibited article identification system, and an AWSM system. The application of the article security inspection method in this application scenario is as follows:

[0119] The parcel is sorted online at the security inspection machine. The security inspection machine collects the order number and takes an X-ray photo of the package to obtain the security inspection data corresponding to the parcel (i.e., the image to be inspected), and outputs the security inspection data to the contraband identification system. The contraband identification system identifies contraband based on the security inspection data, determines contraband and suspected contraband based on the confidence threshold, and further pushes manual judgment for suspected contraband to determine the contraband among the suspected contraband (i.e., the image to be inspected is input into the trained target object detection model, and the trained target object detection model is used to detect the item category of the image to be inspected to obtain the security inspection result), generates contraband data based on the contraband, and pushes the wanted data ( The front-end APP obtains the abnormal express, scans the express and outputs the express information to the back-end. The back-end generates a prohibited goods verification task according to the express information to instruct the staff using the front-end APP to conduct a prohibited goods verification, and feedback the verification results, generates a verification data record according to the verification results, and feedbacks the verification data record to the prohibited goods identification system.

[0120] It should be understood that, although the steps in each flow chart involved in the above-described embodiment are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a portion of the steps in each flow chart involved in the above-described embodiment may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of the steps or stages in other steps.

[0121] In one embodiment, Figure 4 As shown, a flowchart is used to illustrate the article safety inspection method of the present application, and the article safety inspection method specifically includes the following steps:

[0122] Step 402, obtaining a transport sequence identifier corresponding to the item to be inspected and first identity information of the item before security inspection scanning;

[0123] Step 404, performing a security inspection scan on the item to be inspected, and obtaining a scan image corresponding to the security inspection scan;

[0124] Step 406, obtaining the second identity information of the item after the security check scan according to the corresponding transport sequence identifier;

[0125] Step 408, comparing the first identity information and the second identity information;

[0126] Step 410, when the second identity information corresponds to the first identity information, the second identity information and the scanned image are bound and associated with the corresponding transport sequence identifier to obtain the image to be inspected;

[0127] Step 412, obtaining a training image set with object image annotations and an initial object detection model;

[0128] Step 414, training the initial object detection model according to the training image set to obtain a trained target object detection model;

[0129] Step 416, input the image to be inspected into the trained target object detection model, perform object category detection on the image to be inspected using the trained target object detection model, and obtain an object category detection result;

[0130] Step 418, when the item category in the item category detection result includes prohibited items, obtain the category confidence corresponding to the prohibited items;

[0131] Step 420, when the category confidence is greater than or equal to the preset confidence threshold, the security check result is obtained as the presence of the target prohibited item;

[0132] Step 422, when the category confidence is less than the confidence threshold, a security inspection image judgment page is generated and displayed according to the image to be inspected and the prohibited object, an image judgment result corresponding to the security inspection image judgment page is received, and a security inspection result is obtained according to the image judgment result;

[0133] Step 424, when the security check result is that the target prohibited object exists, an interception prompt is generated according to the target prohibited object and the bound identity information;

[0134] Step 426, obtaining the transport sequence information of each transport line;

[0135] Step 428, determining the target transportation line and the target prompt terminal corresponding to the sorting operation on the target transportation line according to the transportation sequence information and the transportation sequence identifier;

[0136] Step 430, push the interception prompt to the target prompt terminal;

[0137] Step 432, receiving an interception feedback result corresponding to the interception prompt;

[0138] Step 434, when the interception feedback result is an interception error, the trained target object detection model is trained according to the object detection result in the interception feedback result and the image to be inspected for security.

[0139] In one embodiment, Figure 5 As shown, an article safety inspection device is provided, including: an acquisition module 502, a detection module 504, a processing module 506 and a push module 508, wherein:

[0140] An acquisition module 502 is used to acquire an image to be inspected corresponding to the item to be inspected, the image to be inspected carrying binding identity information and a transport sequence identifier;

[0141] The detection module 504 is used to input the image to be inspected into the trained target object detection model, and perform object category detection on the image to be inspected through the trained target object detection model to obtain the security inspection result;

[0142] The processing module 506 is used to generate an interception prompt according to the target prohibited object and the bound identity information when the security check result shows that the target prohibited object exists, and determine the target prompt terminal according to the transport sequence identifier;

[0143] The push module 508 is used to push the interception prompt to the target prompt terminal.

[0144] The above-mentioned article security inspection device, after obtaining the image to be inspected corresponding to the article to be inspected, first inputs the image to be inspected into the trained target article detection model, and performs article category detection on the image to be inspected through the trained target article detection model, so as to realize accurate article category recognition of the image to be inspected, improve the security inspection effect, and the security inspection result. When the security inspection result is that there are target prohibited items, an interception prompt is generated according to the target prohibited items and the bound identity information, and the target prompt terminal is determined according to the transport sequence identifier, and the interception prompt is pushed to the target prompt terminal to complete the automated security inspection, which can improve the security inspection efficiency.

[0145] In one embodiment, the acquisition module is also used to obtain the transport sequence identifier corresponding to the item to be inspected and the first identity information of the item before the security inspection scan, perform a security inspection scan on the item to be inspected, obtain a scanned image corresponding to the security inspection scan, obtain the second identity information of the item after the security inspection scan based on the corresponding transport sequence identifier, perform identity verification on the second identity information based on the first identity information, bind the scanned image to a code based on the identity verification result, and associate it with the corresponding transport sequence identifier to obtain the image to be inspected.

[0146] In one embodiment, the acquisition module is further used to compare the first identity information and the second identity information. When the second identity information corresponds to the first identity information, the second identity information and the scanned image are bound and associated with the corresponding transport sequence identifier to obtain the image to be inspected.

[0147] In one embodiment, the detection module is further used to obtain a training image set carrying object image annotations and an initial object detection model, and train the initial object detection model according to the training image set to obtain a trained target object detection model.

[0148] In one embodiment, the detection module is also used to input the image to be inspected for security into a trained target item detection model, perform item category detection on the image to be inspected for security by using the trained target item detection model, and obtain an item category detection result; when the item category in the item category detection result includes prohibited items, obtain the category confidence corresponding to the prohibited items; when the category confidence is greater than or equal to a preset confidence threshold, obtain the security inspection result that the target prohibited items exist; when the category confidence is less than the confidence threshold, generate and display a security inspection image judgment page based on the image to be inspected for security and the prohibited items, receive an image judgment result corresponding to the security inspection image judgment page, and obtain the security inspection result based on the image judgment result.

[0149] In one embodiment, the processing module is further used to obtain the transport sequence information of each transport line, and determine the target transport line and the target prompt terminal corresponding to the sorting operation on the target transport line according to the transport sequence information and the transport sequence identifier.

[0150] In one embodiment, the detection module is also used to receive an interception feedback result corresponding to the interception prompt. When the interception feedback result is an interception error, the trained target object detection model is trained according to the object detection result in the interception feedback result and the image to be inspected for security.

[0151] For the specific embodiments of the article security inspection device, please refer to the embodiments of the article security inspection method above, which will not be repeated here. Each module in the above-mentioned article security inspection device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0152] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 6 As shown. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a method for inspecting the safety of an item is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a key, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0153] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0154] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0155] Acquire an image to be inspected corresponding to the item to be inspected, the image to be inspected carrying binding identity information and a transport sequence identifier;

[0156] The image to be inspected is input into the trained target object detection model, and the trained target object detection model is used to detect the object category of the image to be inspected to obtain the security inspection result;

[0157] When the security check result shows that there are target prohibited items, an interception prompt is generated according to the target prohibited items and the bound identity information, and the target prompt terminal is determined according to the transport sequence identifier;

[0158] Push the interception prompt to the target prompt terminal.

[0159] In one embodiment, when the processor executes the computer program, the following steps are also implemented: obtaining the transport sequence identifier corresponding to the item to be security inspected and the first identity information of the item before the security inspection scan, performing a security inspection scan on the item to be security inspected to obtain a scanned image corresponding to the security inspection scan, obtaining the second identity information of the item after the security inspection scan according to the corresponding transport sequence identifier, performing identity verification on the second identity information according to the first identity information, binding the scanned image with a picture code according to the identity verification result, and associating the corresponding transport sequence identifier to obtain the image to be security inspected.

[0160] In one embodiment, when the processor executes the computer program, the following steps are also implemented: comparing the first identity information and the second identity information, when the second identity information corresponds to the first identity information, binding the second identity information and the scanned image, and associating the corresponding transport sequence identifier to obtain the image to be inspected.

[0161] In one embodiment, when the processor executes the computer program, the following steps are also implemented: obtaining a training image set carrying object image annotations and an initial object detection model, training the initial object detection model according to the training image set, and obtaining a trained target object detection model.

[0162] In one embodiment, when the processor executes the computer program, the following steps are also implemented: inputting the image to be inspected for security into a trained target item detection model, performing item category detection on the image to be inspected for security by using the trained target item detection model to obtain an item category detection result; when the item category in the item category detection result includes prohibited items, obtaining the category confidence corresponding to the prohibited items; when the category confidence is greater than or equal to a preset confidence threshold, obtaining a security inspection result indicating the presence of target prohibited items; when the category confidence is less than the confidence threshold, generating and displaying a security inspection image judgment page based on the image to be inspected for security and the prohibited items, receiving an image judgment result corresponding to the security inspection image judgment page, and obtaining a security inspection result based on the image judgment result.

[0163] In one embodiment, when the processor executes the computer program, the following steps are also implemented: obtaining the transport sequence information of each transport line, and determining the target transport line and the target prompt terminal corresponding to the sorting operation on the target transport line according to the transport sequence information and the transport sequence identifier.

[0164] In one embodiment, when the processor executes the computer program, the following steps are also implemented: receiving an interception feedback result corresponding to the interception prompt, and when the interception feedback result is an interception error, training a trained target object detection model according to the object detection result in the interception feedback result and the image to be inspected for security.

[0165] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0166] Acquire an image to be inspected corresponding to the item to be inspected, the image to be inspected carrying binding identity information and a transport sequence identifier;

[0167] The image to be inspected is input into the trained target object detection model, and the trained target object detection model is used to detect the object category of the image to be inspected to obtain the security inspection result;

[0168] When the security check result shows that there are target prohibited items, an interception prompt is generated according to the target prohibited items and the bound identity information, and the target prompt terminal is determined according to the transport sequence identifier;

[0169] Push the interception prompt to the target prompt terminal.

[0170] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining the transport sequence identifier corresponding to the item to be inspected and the first identity information of the item before the security inspection scan, performing a security inspection scan on the item to be inspected to obtain a scanned image corresponding to the security inspection scan, obtaining the second identity information of the item after the security inspection scan based on the corresponding transport sequence identifier, performing identity verification on the second identity information based on the first identity information, binding the scanned image with a picture code based on the identity verification result, and associating the corresponding transport sequence identifier to obtain the image to be inspected.

[0171] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: comparing the first identity information and the second identity information, when the second identity information corresponds to the first identity information, binding the second identity information and the scanned image, and associating the corresponding transport sequence identifier to obtain the image to be inspected.

[0172] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining a training image set carrying object image annotations and an initial object detection model, training the initial object detection model according to the training image set, and obtaining a trained target object detection model.

[0173] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: inputting the image to be inspected for security into a trained target item detection model, performing item category detection on the image to be inspected for security by using the trained target item detection model to obtain an item category detection result; when the item category in the item category detection result includes prohibited items, obtaining the category confidence corresponding to the prohibited items; when the category confidence is greater than or equal to a preset confidence threshold, obtaining a security inspection result indicating the presence of target prohibited items; when the category confidence is less than the confidence threshold, generating and displaying a security inspection image judgment page based on the image to be inspected for security and the prohibited items, receiving an image judgment result corresponding to the security inspection image judgment page, and obtaining a security inspection result based on the image judgment result.

[0174] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining the transport sequence information of each transport line, and determining the target transport line and the target prompt terminal corresponding to the sorting operation on the target transport line based on the transport sequence information and the transport sequence identifier.

[0175] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: receiving an interception feedback result corresponding to the interception prompt, and when the interception feedback result is an interception error, training a trained target object detection model according to the object detection result in the interception feedback result and the image to be inspected for security.

[0176] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0177] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0178] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A method for inspecting the safety of an article, characterized in that: The method comprises: Obtain the transport sequence identification corresponding to the item to be inspected and the first identity information of the item before the security inspection scan; Performing a security inspection scan on the item to be inspected, and obtaining a scanned image corresponding to the security inspection scan; According to the corresponding transport sequence identifier, obtain the second identity information of the item after the security inspection scan; Performing identity verification on the second identity information according to the first identity information, binding the scanned image with a code according to the identity verification result, and associating the corresponding transport sequence identifier to obtain an image to be inspected, wherein the image to be inspected carries the bound identity information and the transport sequence identifier; Inputting the image to be inspected into a trained target object detection model, performing object category detection on the image to be inspected using the trained target object detection model, and obtaining a security inspection result; When the security check result is that there is a target prohibited object, an interception prompt is generated according to the target prohibited object and the bound identity information, and a target prompt terminal is determined according to the transport sequence identifier; Push the interception prompt to the target prompt terminal.

2. The method according to claim 1, characterized in that The performing identity verification on the second identity information according to the first identity information, binding the scanned image with a code according to the identity verification result, and associating the corresponding transport sequence identifier to obtain the image to be inspected includes: comparing the first identity information and the second identity information; When the second identity information corresponds to the first identity information, the second identity information and the scanned image are bound and associated with the corresponding transport sequence identifier to obtain the image to be inspected.

3. The method according to claim 1, characterized in that Before inputting the image to be inspected into a trained target object detection model, and performing object category detection on the image to be inspected by using the trained target object detection model to obtain a security inspection result, the method further includes: Obtain a training image set with object image annotations and an initial object detection model; The initial object detection model is trained according to the training image set to obtain a trained target object detection model.

4. The method according to claim 1, characterized in that: The step of inputting the image to be inspected into a trained target object detection model, and performing object category detection on the image to be inspected by using the trained target object detection model to obtain a security inspection result comprises: Inputting the image to be inspected into a trained target object detection model, performing object category detection on the image to be inspected using the trained target object detection model, and obtaining an object category detection result; When the item category in the item category detection result includes prohibited items, obtaining the category confidence corresponding to the prohibited items; When the category confidence is greater than or equal to a preset confidence threshold, the security check result is obtained as the presence of the target prohibited item; When the category confidence is less than the confidence threshold, a security inspection image judgment page is generated and displayed according to the image to be inspected and the prohibited item, an image judgment result corresponding to the security inspection image judgment page is received, and a security inspection result is obtained according to the image judgment result.

5. The method according to claim 1, characterized in that Determining the target prompt terminal according to the transport sequence identifier includes: Get the transport sequence information of each transport line; According to the transport sequence information and the transport sequence identifier, a target transport line and a target prompt terminal on the target transport line corresponding to the sorting operation are determined.

6. The method according to claim 1, characterized in that After pushing the interception prompt to the target prompt terminal, the method further includes: Receiving an interception feedback result corresponding to the interception prompt; When the interception feedback result is an interception error, the trained target object detection model is trained according to the object detection result in the interception feedback result and the image to be inspected for security.

7. A safety inspection device, characterized in that: The device comprises: an acquisition module, used to acquire a transport sequence identifier corresponding to an item to be inspected and first identity information of the item before security inspection scanning, perform a security inspection scan on the item to be inspected, obtain a scanned image corresponding to the security inspection scan, acquire second identity information of the item after security inspection scanning according to the corresponding transport sequence identifier, perform identity verification on the second identity information according to the first identity information, bind the scanned image with a picture code according to the identity verification result, and associate the corresponding transport sequence identifier to obtain an image to be inspected, wherein the image to be inspected carries the bound identity information and the transport sequence identifier; A detection module, used to input the image to be inspected for security into a trained target object detection model, perform object category detection on the image to be inspected for security by using the trained target object detection model, and obtain a security inspection result; A processing module, configured to generate an interception prompt according to the target prohibited object and the binding identity information when the security check result indicates that a target prohibited object exists, and to determine a target prompt terminal according to the transport sequence identifier; A push module is used to push the interception prompt to the target prompt terminal.

8. The device according to claim 7, characterized in that The acquisition module is further used to compare the first identity information and the second identity information, and when the second identity information corresponds to the first identity information, bind the second identity information and the scanned image, and associate the corresponding transport sequence identifier to obtain the image to be inspected.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images