Inspection Task Processing Method, Device and Electronic Equipment

Abstract

The present disclosure provides a method, an apparatus, and an electronic device for processing inspection tasks. The method for processing inspection tasks includes: obtaining task information corresponding to a target inspection task, where the task information includes an item code and a scanned image of an item to be inspected corresponding to the target inspection task; obtaining a risk characteristic value of the item to be inspected according to the task information, where the risk characteristic value includes an underreporting characteristic value and / or a false alarm characteristic value; and determining whether to generate a manual drawing inspection task based on the target inspection task according to the risk characteristic value corresponding to each item to be inspected. Embodiments of the present disclosure can improve the processing efficiency and accuracy of inspection tasks during machine inspection at the same time.

Description

Background Art

[0002] Machine inspection refers to a technical means of scanning an item to be inspected through transmission means such as X-rays and analyzing the scanned image to determine whether the item to be inspected complies with the regulations. Machine inspection is usually applied in security inspections at various places, customs entry and exit inspections, and other occasions. In the related art, due to different image analysis algorithms, different experience levels of image analysis personnel, and the influence of the image analysis personnel themselves and the external environment during analysis, misjudgment or missed reports may occur in both automatic recognition of scanned images and manual review of images.

[0003] Therefore, how to improve the recognition accuracy of machine inspection scanned images has become a major problem in this field.

[0004] It should be noted that the information disclosed in the above background art section is only used to enhance the understanding of the background of the present disclosure, and thus may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0005] The purpose of the present disclosure is to provide a method, device, and electronic device for processing inspection tasks, which are used to improve the recognition efficiency and recognition accuracy of machine inspection scanned images at the same time.

[0006] According to the first aspect of the embodiments of the present disclosure, a method for processing inspection tasks is provided, including: obtaining task information corresponding to a target inspection task, where the task information includes the item code and scanned image of the item to be inspected corresponding to the target inspection task; obtaining a risk feature value of the item to be inspected according to the task information, where the risk feature value includes a missed report feature value and / or a false report feature value; and determining whether to generate a manual review task based on the target inspection task according to the risk feature value corresponding to each item to be inspected.

[0007] In an exemplary embodiment of the present disclosure, the determining whether to generate a manual review task based on the target inspection task according to the risk feature value corresponding to each item to be inspected includes: when the risk feature value of any one of the items to be inspected meets a preset condition, generating a manual review task according to the target inspection task; when the risk feature value of each item to be inspected does not meet the preset condition, recognizing the scanned image and generating a processing result for the target inspection task according to the recognition result of the scanned image.

[0008] In an exemplary embodiment of the present disclosure, the risk characteristic value includes a missed detection characteristic value, the preset condition includes that the missed detection characteristic value exceeds a first preset value, and generating an artificial drawing review task according to the target inspection task includes: when the missed detection characteristic value of a to-be-inspected item exceeds the first preset value, generating an artificial drawing review task and displaying the scanned image, and displaying the missed detection risk reminder information of the to-be-inspected item in the interface of the artificial drawing review task.

[0009] In an exemplary embodiment of the present disclosure, generating an artificial drawing review task according to the target inspection task further includes: when the missed detection characteristic value of a to-be-inspected item exceeds the first preset value, obtaining the historical risk image of the to-be-inspected item and providing the historical risk image, where the historical risk image is a historical scanned image related to the increase of the risk characteristic value of the to-be-inspected item.

[0010] In an exemplary embodiment of the present disclosure, the risk characteristic value includes a false alarm characteristic value, the preset condition includes that the false alarm characteristic value exceeds a second preset value, and generating an artificial drawing review task according to the target inspection task includes: when the false alarm characteristic value of a to-be-inspected item exceeds the second preset value, generating an artificial drawing review task and displaying the scanned image, and displaying the false alarm risk reminder information of the to-be-inspected item in the interface of the artificial drawing review task.

[0011] In an exemplary embodiment of the present disclosure, it further includes: obtaining the image analysis result of the target inspection task and the re-inspection result corresponding to the target inspection task, where the image analysis result includes an automatic image analysis result and an artificial drawing review recognition result; when the image analysis result shows no abnormality and the re-inspection result includes an abnormality, determining the to-be-inspected item corresponding to the re-inspection result and the abnormal area of the to-be-inspected item, increasing the missed detection characteristic value of the to-be-inspected item, and recording the abnormal area of the to-be-inspected item as the risk area of the to-be-inspected item; when the image analysis result is abnormal and the re-inspection result shows no abnormality, determining the to-be-inspected item corresponding to the recognition result and increasing the false alarm characteristic value of the to-be-inspected item.

[0012] In an exemplary embodiment of the present disclosure, the re-inspection result includes at least one of a re-image analysis result of the target inspection task, an inspection result of the target inspection task, and an external inspection result related to the target inspection task.

[0013] According to a second aspect of the embodiments of the present disclosure, there is provided an inspection task processing device, including: a task information determination module configured to obtain task information corresponding to a target inspection task, where the task information includes an item code and a scanned image of an item to be inspected corresponding to the target inspection task; a risk information determination module configured to obtain a risk characteristic value of the item to be inspected according to the task information, where the risk characteristic value includes a missed report characteristic value and / or a false report characteristic value; and an artificial drawing review task generation module configured to determine whether to generate an artificial drawing review task based on the target inspection task according to the risk characteristic value corresponding to each item to be inspected.

[0014] According to a third aspect of the present disclosure, there is provided an electronic device, including: a memory; and a processor coupled to the memory, where the processor is configured to execute the method according to any one of the above based on instructions stored in the memory.

[0015] According to a fourth aspect of the present disclosure, there is provided a computer-readable storage medium having a program stored thereon, where the program, when executed by a processor, implements the inspection task processing method according to any one of the above.

[0016] By determining whether to generate an artificial drawing review task for a target inspection task according to the risk characteristic value of each item to be inspected, the embodiments of the present disclosure can reduce the time occupied by inspection tasks with high drawing review difficulty during automatic drawing review and cannot obtain an accurate drawing review conclusion; at the same time, judging the target inspection task according to the risk characteristic value of the item to be inspected can avoid the drawbacks of false reports and omissions when inspecting only based on the scanned image, improve the inspection task processing efficiency, and improve the drawing review accuracy.

[0017] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings herein are incorporated into the specification and form a part of the specification, showing embodiments consistent with the present disclosure, and are used together with the specification to explain the principles of the present disclosure. Obviously, the accompanying drawings in the following description are only some embodiments of the present disclosure, and those of ordinary skill in the art can obtain other drawings without creative efforts based on these drawings.

[0019] Figure 1 is a flowchart of the inspection task processing method in an exemplary embodiment of the present disclosure.

[0020] Figure 2 is a schematic diagram of a machine inspection system in an embodiment of the present disclosure.

[0021] Figure 3It is a sub - flowchart of step S3 in an embodiment of the present disclosure.

[0022] Figures 4A to 4C They are three exemplary image information of manual drawing review tasks.

[0023] Figure 5 It is another flowchart of method 100 in an embodiment of the present disclosure.

[0024] Figure 6 It is an information update flowchart of the knowledge base system in an embodiment of the present disclosure.

[0025] Figure 7 It is an information update diagram of the knowledge base system in another embodiment of the present disclosure.

[0026] Figure 8 It is a task - processing flowchart of the machine inspection system in an embodiment of the present disclosure.

[0027] Figure 9 It is a block diagram of an inspection task - processing device in an exemplary embodiment of the present disclosure.

[0028] Figure 10 It is a block diagram of an electronic device in an exemplary embodiment of the present disclosure. Detailed implementation manners

[0029] Now, the exemplary embodiments will be described more comprehensively with reference to the accompanying drawings. However, the exemplary embodiments can be implemented in various forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be more thorough and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art. The described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of the embodiments of the present disclosure. However, those skilled in the art will realize that one or more of the specific details can be omitted in practicing the technical solutions of the present disclosure, or other methods, components, devices, steps, etc. can be adopted. In other cases, well - known technical solutions are not shown or described in detail to avoid obscuring the various aspects of the present disclosure.

[0030] In addition, the accompanying drawings are only schematic illustrations of the present disclosure, and the same reference numerals in the drawings denote the same or similar parts, so repeated descriptions thereof will be omitted. Some of the block diagrams shown in the drawings are functional entities, which do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0031] The following will describe the exemplary embodiments of the present disclosure in detail with reference to the accompanying drawings.

[0032] Figure 1 It is a flowchart of the inspection task processing method in the exemplary embodiment of the present disclosure.

[0033] Referring to Figure 1 , the inspection task processing method 100 may include:

[0034] Step S1, obtaining task information corresponding to the target inspection task, where the task information includes the item code and scanned image of the item to be inspected corresponding to the target inspection task;

[0035] Step S2, obtaining the risk characteristic value of the item to be inspected according to the task information, where the risk characteristic value includes an underreporting characteristic value and / or a false alarm characteristic value;

[0036] Step S3, determining whether to generate a manual drawing review task based on the target inspection task according to the risk characteristic value corresponding to each item to be inspected.

[0037] In the embodiment of the present disclosure, by determining whether to generate a manual drawing review task for the target inspection task according to the risk characteristic value of each item to be inspected, it is possible to reduce the situation that inspection tasks with high drawing review difficulty occupy a long time and cannot obtain accurate drawing review conclusions during the automatic drawing review process; at the same time, judging the target inspection task according to the risk characteristic value of the item to be inspected can avoid the drawbacks of false alarms and omissions when inspecting only based on the scanned image, improve the inspection task processing efficiency, and improve the drawing review accuracy.

[0038] The method provided in the embodiment of the present disclosure can be applied to each time node of the execution of the machine inspection task, such as after receiving the machine inspection task, after the execution of the pre - task classification (such as classifying the task into a key - attention task or an ordinary task), after the execution of the automatic drawing review, etc. In addition, the method provided in the embodiment of the present disclosure can be executed independently or in cooperation with multiple other functions, such as being executed simultaneously with the automatic drawing review task, being executed simultaneously with the task classification, etc. The finally generated manual drawing review task can be generated not only based on the risk characteristic value, but also based on the task results of the upstream time nodes and the task results of other functions executed simultaneously in cooperation. The method provided in the embodiment of the present disclosure is only an optional function in the machine inspection process and does not limit the overall process and overall functions involved in the machine inspection.

[0039] Next, each step of the inspection task processing method 100 will be described in detail.

[0040] In step S1, task information corresponding to the target inspection task is obtained, and the task information includes the item code and the scanned image of the item to be inspected corresponding to the target inspection task.

[0041] The target inspection tasks referred to in this disclosure include, but are not limited to, customs inspection tasks, and the scanned images include, but are not limited to, customs inspection scanned images. Each target inspection task may correspond to one or more scanned images, and the formation methods of the scanned images include, but are not limited to, methods such as X-ray, infrared detector, and thermal imaging.

[0042] The task information is mainly used to determine the type of the item to be inspected. In some embodiments, the task information can also be used to obtain the precautions corresponding to the item to be inspected. The task information includes the item code of the item to be inspected corresponding to the target inspection task. For example, in container inspection, if the shipper enters the items to be inspected as cotton and jute in the bill of lading or manifest corresponding to a container, then the items to be inspected corresponding to the target inspection task are cotton and jute, and the task information records the HSCODE codes (Harmonized System CODE, also known as commodity code) of cotton and jute.

[0043] Since the item code includes multiple digits and different numbers of digits represent different classification levels, when using the task information to determine the item type, the number of digits of the item code to be used can be set according to the actual situation. The more digits of the item code used, the more detailed the item type classification, and the fewer digits of the item code used, the broader the item type classification.

[0044] In addition, the task information can also include other information of the item to be inspected entered in various ways, such as various documents, various remarks, etc. The specific content of the task information is not limited in the implementation of this disclosure, and those skilled in the art can set the content of the task information to be used according to their needs.

[0045] When the embodiment of this disclosure is applied in the customs inspection process, the scanning image processing method 100 can be executed by the machine inspection system.

[0046] Figure 2 It is a schematic diagram of the machine inspection system in an embodiment of this disclosure.

[0047] Reference Figure 2 , the machine inspection system 200 may include a scanning system 21, a task processing system 20, and an artificial drawing review system 22. The task processing system 20 is connected to the scanning system 21 and the artificial drawing review system 22. The scanning system 21 is used to scan one or more items to be inspected and generate one or more scanned images corresponding to the items to be inspected; the artificial drawing review system 22 is used to provide a drawing review task interface for the drawing review personnel and receive the drawing review conclusion information input by the drawing review personnel.

[0048] The task processing system 20 can be implemented by one or more hardware devices, such as computers, servers, etc. In some embodiments, the task processing system 20 can also be implemented by a cloud computing system. The scanning system 21 can include multiple scanning terminals, and different scanning terminals can be located at different geographical locations (such as checkpoints). The manual drawing review system 22 can include multiple manual drawing review terminals, and each manual drawing review terminal at least includes an image display device and an information input device. Different manual drawing review terminals can be located at different geographical locations, such as different office locations, and different office locations can be in the same city or in different cities.

[0049] The task processing system 20 can at least include an information receiving module 201, an image processing module 202, an information storage module 203, and a comprehensive processing module 204. The information receiving module 201 is communicatively connected to the scanning system 21 for receiving scanned images. In addition, the information receiving module 201 can also receive externally transmitted information through a human-computer interaction unit, and this information includes but is not limited to a scanned image recognition program, a basis for judging recognition results, task information, a list of items to be inspected input externally, and a risk characteristic value corresponding to each item to be inspected; the image processing module 202 is used for recognizing the scanned image and outputting a processing result of the scanned image according to the recognition result; the information storage module 203 is, for example, a memory, and is used for storing information received by the information receiving module 201, such as scanned images, recognition results and processing results corresponding to each scanned image, and task information; the comprehensive processing module 204 is used for controlling the transmission of various types of information and tasks between the scanning system 21, the manual drawing review system 22, and the inspection system 20, and between the information receiving module 201, the image processing module 202, and the information storage module 203. At the same time, control instructions are sent to the information receiving module 201, the image processing module 202, and the information storage module 203.

[0050] Figure 2 In the illustrated embodiment, the task processing system 20 is used to execute the scanned image processing method 100 of the present disclosure embodiment. The functions of the various modules in the task processing system 20 vary according to different embodiments of the method 100, and the present disclosure does not strictly limit this.

[0051] In step S1, the task processing system 20 receives a target inspection task and task information corresponding to the target inspection task through the information receiving module 21. The task information includes one or more items to be inspected and scanned images of the one or more items to be inspected. In some embodiments, the item codes of the items to be inspected can be recorded in the task information through product codes.

[0052] In step S2, risk characteristic values of the item to be inspected are obtained according to the task information, where the risk characteristic values include false negative characteristic values and / or false positive characteristic values.

[0053] The risk characteristic value is a quantified value of the image analysis risk corresponding to each item to be inspected, which is pre-stored. The image analysis risk includes, but is not limited to, false negatives and false positives. Correspondingly, the risk characteristic values include, but are not limited to, false negative characteristic values and false positive characteristic values. Among them, a false negative means that an abnormality (such as concealment) is found in the item to be inspected during subsequent inspection, but the image analysis result shows no abnormality; a false positive means that the image analysis result shows an abnormality, but no abnormality is found during the actual inspection process.

[0054] In other embodiments, the risk characteristic value may only include false negative characteristic values or only include false positive characteristic values, or the risk characteristic value may also include or only include characteristic values corresponding to other influencing factors that can cause image analysis errors. In the embodiments of the present disclosure, the risk characteristic value is described by taking false negative characteristic values and false positive characteristic values as examples, and other types of characteristic values can be processed similarly according to the principles provided in the embodiments of the present disclosure.

[0055] The risk characteristic value can be calculated regularly based on the image analysis results in historical inspection data and updated in real time based on the most recent one or several inspection data. When the risk characteristic value includes false negative characteristic values and false positive characteristic values, the information storage module 203 in the task processing system 20 can pre-store false negative characteristic values and false positive characteristic values of various items to be inspected.

[0056] For example, the association relationship between the product code of the item to be inspected and the false negative characteristic value and the false positive characteristic value can be established through a data table. The false negative characteristic value and the false positive characteristic value can have preset initial values and be updated regularly based on the image analysis results in historical inspection data. In addition, the task processing system 20 can update the false negative characteristic value and the false positive characteristic value according to the feedback of each drawing review result. The process of the false negative characteristic value and the false positive characteristic value is described in detail in the subsequent embodiments.

[0057] One or more data tables can form a machine inspection knowledge base. The machine inspection knowledge base can be open and shared with one or more task processing systems to provide more accurate reference results through big data.

[0058] When multiple items to be inspected are determined in step S1, the image processing module 202 can directly retrieve the false negative characteristic values and false positive characteristic values corresponding to each item to be inspected from the information storage module according to the product code of the item to be inspected.

[0059] In step S3, it is determined whether to generate a manual drawing review task based on the target inspection task according to the risk characteristic value corresponding to each item to be inspected.

[0060] In one embodiment, before generating the manual drawing review task, the target inspection task can be roughly processed first to mark the target inspection tasks with obvious anomalies, reducing the burden of automatic drawing review and manual drawing review tasks.

[0061] The image processing module 202 can identify one or more items to be inspected in the target inspection task according to conventional image analysis algorithms, identify the types of each item to be inspected, and finally summarize the information to obtain the types of items to be inspected and the quantity of each type of item to be inspected, and compare them with the task information corresponding to the target inspection task. When the types and quantities of the items to be inspected match those recorded in the task information, risk feature values such as the omission feature value and false alarm feature value corresponding to each item to be inspected are obtained, and whether to generate a manual drawing review task is determined according to the risk feature values. When a target inspection task corresponds to multiple scanned images, it can be processed according to the recognition results of each scanned image.

[0062] Of course, in some embodiments, it is first determined whether to generate a manual drawing review task, and then the scanned images are identified and analyzed.

[0063] Figure 3 It is a sub - flowchart of step S3 in an embodiment of the present disclosure.

[0064] Reference Figure 3 In one embodiment, step S3 may include:

[0065] Step S31, when the risk feature value of any one of the items to be inspected meets a preset condition, generate a manual drawing review task according to the target inspection task;

[0066] Step S32, when the risk feature values of each of the items to be inspected do not meet the preset condition, identify the scanned image and generate a processing result for the target inspection task according to the recognition result of the scanned image.

[0067] In the embodiments of the present disclosure, when the risk feature values include omission feature values and / or false alarm feature values, the preset condition may be that the omission feature value exceeds a first preset value and / or the false alarm feature value exceeds a second preset value. The first preset value and the second preset value can be set by themselves according to the value criteria of the feature values. "Exceeds" can either mean greater than the first preset value or greater than the second preset value, or less than the first preset value or the second preset value. The specific meaning of "exceeds" can be determined according to the formation method and evaluation criteria of the omission feature value and the false alarm feature value, as long as it means that the omission feature value and the false alarm feature value reach the threshold.

[0068] In the embodiments of the present disclosure, the first preset value and the second preset value can be set according to the value ranges of the missed detection eigenvalue and the false detection eigenvalue and the statistical analysis results. The value ranges of the missed detection eigenvalue and the false detection eigenvalue can be the same. For example, both are 0 - 100. Correspondingly, the first preset value and the second preset value can also be the same. For example, both are 60. Alternatively, the first preset value and the second preset value can also be different according to the analysis results. In some embodiments, the value ranges of the missed detection eigenvalue and the false detection eigenvalue are also different according to the different frequencies of value increase. For example, the increase frequency of the missed detection eigenvalue is higher, and the types of corresponding increase scenarios (situations that cause the increase of the missed detection eigenvalue) are more, with a value range of 0 - 100, and each increase raises a value ranging from 0 to 10; the increase frequency of the false detection eigenvalue is lower, and the types of corresponding increase scenarios (situations that cause the increase of the false detection eigenvalue) are only the false detection type, then the value range can be set to 0 - 10, and each increase raises 1 value. The above setting methods are only examples, and those skilled in the art can set them according to the actual application scenarios.

[0069] When the missed detection eigenvalue exceeds the first preset value, it indicates that the item to be inspected has a relatively high missed detection of concealed items according to historical data records. When the false detection eigenvalue exceeds the second preset value, it indicates that the item to be inspected has a relatively high false detection and missed detection according to historical data records. Both of these situations indicate that the recognition result obtained when the machine inspection system automatically processes and recognizes the image may not be accurate and requires manual inspection. If any of the above two situations occurs, a manual drawing review task needs to be generated. And if neither of the above two situations occurs, it means that the current task is within the range that can be accurately judged by the task processing system. The task processing system can, according to conventional judgment criteria, such as when the task information and the types and quantities (or weights, volumes) of the items to be inspected are completely matched, determine that the target inspection task passes the image inspection (the processing result is passed).

[0070] Next, the formation process of the manual drawing review task is introduced.

[0071] The manual drawing review task can be generated by the task processing system 20 and sent to the manual drawing review system 22, so that the manual drawing review system 22 can assign the manual drawing review task to the drawing review personnel.

[0072] When generating the manual drawing review task according to the target inspection task, information such as the missed detection eigenvalue and the false detection eigenvalue of each item to be inspected corresponding to the target inspection task, and the task information (at least including the item code and the scanned image) can be sent to the manual drawing review system 22 to generate a manual drawing review task including the scanned image. At the same time, the efficiency of the drawing review personnel to complete the manual drawing review task can be directly improved through the display of the image information.

[0073] Figures 4A to 4C are three exemplary image information of the manual drawing review task.

[0074] Reference Figure 4A , in one embodiment, when the false alarm characteristic value of an item 41 to be inspected exceeds a first preset value, according to the false alarm information of the item to be inspected, the false alarm information may include, for example, the false alarm location and the false alarm item. When displaying the scanned image in the manual drawing review task, at the same time, false alarm risk prompt information of the item to be inspected is displayed in the interface, and the false alarm risk prompt information may include the false alarm location and the false alarm item. In some embodiments, the false alarm risk prompt information is implemented by text. In other embodiments, the area corresponding to the false alarm prompt information may also be marked in the scanned image, and the marking method may be, for example, highlighting or image flashing, and is represented by a dotted line box in Figure 4A . In addition, the false alarm risk prompt information can also be implemented in various media forms such as audio, video, pictures, animations, etc., and the present disclosure does not make special restrictions on this.

[0075] In some embodiments, the historical seizure records corresponding to the item 41 to be inspected can be obtained, and the corresponding relationship between the seized item and the hiding area (risk area) in each historical seizure record can be obtained, so as to summarize the most likely types of items hidden in each risk area. When displaying the false alarm risk prompt information of the item 41 to be inspected, one or more risk areas 42 and the most likely types of items hidden in each risk area 42 are displayed at the same time.

[0076] In some cases, there may be situations where the risk area 42 in the scanned image is blocked, the item 41 to be inspected itself is blocked, etc. At this time, the risk area 42 can be displayed through text information, or the risk area 42 and the most likely types of items hidden in each risk area 42 can be displayed at the same time.

[0077] Reference Figure 4B , when the false alarm characteristic value of an item 41 to be inspected exceeds a first preset value, the historical risk image of the item to be inspected can also be obtained, and the historical risk image 43 is provided. Among them, the historical risk image 43 is a historical scanned image related to the increase in the false alarm characteristic value of the item 41 to be inspected.

[0078] Among them, providing the historical risk image 43 is, for example, to display the historical risk image 43 or provide a viewing entry for the historical risk image 43. When displaying the historical risk image 43, it can be displayed on a single screen or together with the scanned image. As shown in Figure 4B , when displaying the historical risk image 43, the image 43 of the item 41 to be inspected in the historical risk image 43 and the relevant text information can be highlighted. The text information may include information such as the packaging method, placement posture, placement position, size and weight of the seized content of the item to be inspected in the inspection task corresponding to the historical risk image. In addition, a viewing entry for the historical seizure record of the item to be inspected can also be displayed.

[0079] Figure 4A and Figure 4B The illustrated embodiments may exist independently or simultaneously.

[0080] That is, in the machine inspection and verification system, when the inspector manually examines the drawings, on the basis of providing the historical risk images of the inspection items of the task, the false alarm information and seizure information of this type of item are provided, and corresponding missed alarm prompts are made by means of highlighting and flashing display lights. For example, for the inspection task of commodity A (commodity code: 12345678), after querying that the historical seizures of drugs for this type of inspection item are 20 times, the drawing examiner can be prompted to pay key attention to this task. If necessary, an entry for viewing the seizure history is provided for the drawing examiner to view the seizure details and the corresponding historical risk images. By analyzing the packaging method, placement posture, placement position, size and weight of the seized content of the item to be inspected in the historical risk images of the seizures, and comparing and analyzing with the current task image, more rigorous image analysis and judgment can be carried out to improve the inspection accuracy.

[0081] Reference Figure 4C , when the false alarm characteristic value of an item 44 to be inspected exceeds a second preset value, the scanned image corresponding to the target inspection task can be displayed in the manual drawing inspection task, and at the same time, the false alarm risk prompt information of the item to be inspected is displayed on the interface of the manual drawing inspection task. The false alarm risk prompt information is, for example, expressed in words, and the content includes information such as the item to be inspected, its false alarm times, false alarm content, false alarm position, etc. In some embodiments, a graphical false alarm risk reminder information 45 can also be marked in the area corresponding to the item 44 to be inspected. The graphical false alarm risk reminder information 45 can also be realized by means of color marking, flashing marking, etc. In Figure 4C , the false alarm risk reminder information 45 is represented by a dotted line box. In addition, the false alarm risk prompt information can also be realized by various media forms such as audio, video, pictures, animations, etc., and the present disclosure does not make special restrictions on this.

[0082] The function of generating a manual drawing inspection task according to the missed alarm characteristic value or the function of generating a manual drawing inspection task according to the false alarm characteristic value can be used alone or turned off alone, or can be used simultaneously, and those skilled in the art can set it according to the actual situation.

[0083] That is, in the embodiments of the present disclosure, the task processing system 20 can analyze the items to be inspected in the target inspection task according to the scanned image, inspection task information (such as inspection task information, inspection instruction information, etc.), and the omission information and false alarm information of each item to be inspected (such as the number of false alarms of the inspection system in the recent month, whether the false alarm version number is the same as the current one, etc.). It can determine whether to directly generate a manual drawing inspection task for the target inspection task with a relatively high drawing inspection difficulty by setting thresholds (whether the false alarm feature value is greater than the second preset value), decision trees (all the judgment processes included in the above embodiments), etc., and provide image analysis recommendation information (risk area / false alarm area marking), thereby effectively reducing the low task processing efficiency caused by inaccurate automatic drawing inspection, improving the processing efficiency and processing accuracy of the manual drawing inspection task, and further improving the recognition accuracy of the machine inspection system.

[0084] After the task processing system 20 generates a manual drawing inspection task and sends it to the manual drawing inspection system 22, it can receive the feedback result of the manual drawing inspection system 22 and other subsequent inspection results corresponding to the target inspection task (such as information on unpacking inspection, police investigation notice, etc.) to form a feedback on the target inspection task, update the omission feature value and false alarm feature value of the target inspection task, and improve the subsequent inspection accuracy.

[0085] Figure 5 It is another flowchart of the method 100 in an embodiment of the present disclosure.

[0086] Reference Figure 5 , in one embodiment, the method 100 further includes:

[0087] Step S51, obtaining the image analysis result of the target inspection task and the re-inspection result corresponding to the target inspection task, where the image analysis result includes an automatic image analysis result and a manual drawing inspection recognition result;

[0088] Step S52, when the image analysis result shows no abnormality and the re-inspection result includes an abnormality, determining the item to be inspected corresponding to the re-inspection result and the abnormal area of the item to be inspected, increasing the omission feature value of the item to be inspected, and recording the abnormal area of the item to be inspected as the risk area of the item to be inspected;

[0089] Step S53, when the image analysis result is abnormal and the re-inspection result shows no abnormality, determining the item to be inspected corresponding to the recognition result and increasing the false alarm feature value of the item to be inspected.

[0090] In Figure 5In the illustrated embodiment, the product code, false negative eigenvalue, and false positive eigenvalue of each item to be inspected can be recorded in a data table, and then, for each processing of the scan data to be processed, the false negative eigenvalue and false positive eigenvalue of one or more items to be inspected are updated. "Abnormal" can include situations such as concealment and restricted / prohibited items.

[0091] Among them, the image analysis result of the target inspection task refers to the automatic image analysis result directly output by the task processing system 20 and / or the manual drawing review recognition result, including recognition results such as "the target inspection task is abnormal" output when the type or quantity of the items to be inspected does not fully match, and recognition results generated according to the matching degree between the task information and the items to be inspected when the false negative eigenvalue of each item to be inspected does not exceed the first preset value and the false positive eigenvalue does not exceed the second preset value. The image analysis result can include two categories: normal and abnormal.

[0092] The image analysis result can reflect the processing accuracy of the machine inspection system for the inspection task. If the image analysis result is inaccurate, it indicates that the current system and the drawing reviewer need to improve their processing capabilities for this inspection task, and relevant information of this matter needs to be recorded to form a database for reference in subsequent task processing. Therefore, in the embodiments of the present disclosure, according to the relationship between the image analysis result and the re-inspection result of each inspection task, it is determined whether to update the false negative eigenvalue and false positive eigenvalue, and when it is determined to update, how much needs to be updated.

[0093] In the embodiments of the present disclosure, the re-inspection result includes at least one of the re-image analysis result of the target inspection task, the inspection result of the target inspection task, and the external inspection result related to the target inspection task. Among them, the re-image analysis result of the target inspection task can be the review image analysis result of the manual drawing review task, etc., the inspection result of the target inspection task can include inspection results generated by other direct contact means such as unpacking and detection by working dogs, and the external inspection result related to the target inspection task can include inspection results executed by the police or other law enforcement departments on the current inspection object (such as a container or a suitcase) leaving the checkpoint.

[0094] When the final inspection conclusion (including manual drawing review) of a target inspection task or the subsequent disposal conclusion of an external system is found to be abnormal, that is, the inspection task is considered to be seized, and the recognition result of the task processing system 20 is not found to be abnormal, the false negative eigenvalue of the item to be inspected with an abnormality corresponding to the target inspection task can be updated.

[0095] Table 1 is a record form for updating the false negative eigenvalue.

[0096] Table 1:

[0097] When a seizure event occurs, the machine inspection and verification system 200 or other relevant subsequent disposal systems make detailed records of the seizure situation (including the machine inspection task number with a seizure, the items to be inspected, and the seizure location) to generate seizure information. Therefore, the seizure information can be identified through technologies such as semantic recognition to extract the names of the items to be inspected corresponding to the seized content and the seizure location, so as to form a database that records the historical seizure data of each item to be inspected. For example, a container is loaded with commodity A and other items, and some drugs are seized in commodity A. According to the seizure information, the following information can be extracted:

[0098] Items to be inspected: Commodity A (Commodity code: 12345678);

[0099] Seized content: Drugs.

[0100] Next, the above information can be updated and added to the database that records the historical seizure data of each item to be inspected, forming the following data record:

[0101] Commodity code: 12345678;

[0102] Commodity name: Commodity A;

[0103] Number of seizures in the past year: 20;

[0104] Seized content in the past year: Drugs;

[0105] Seizure location in the past year: Under the table board, inside the packaging foam;

[0106] False alarm characteristic value: 89;

[0107] ……

[0108] The false alarm characteristic value of the above commodity A (commodity code: 12345678) is the content updated according to the current seizure result. The value of each update of the false alarm characteristic value can increase according to the update frequency. For example, when seized for the first time, the value of the false alarm characteristic value can be increased by 1, and when seized for the twentieth time, the value of the false alarm characteristic value can be increased by 10, and so on.

[0109] To improve the information processing efficiency, seizure information can be obtained regularly from multiple preset channels (such as this checkpoint, subsequent checkpoints, other law enforcement agencies) (for example, every 24 hours or every 12 hours), and the false alarm characteristic values of each item to be inspected can be calculated and updated centrally, and these false alarm characteristic values can be updated to each task processing system. Thus, when each inspection task is carried out, the false alarm characteristic value of the item to be inspected can be directly and quickly obtained to assist the task processing system 20 in making judgments.

[0110] When the final inspection conclusion of the on-machine inspection task or the subsequent disposal conclusion of the external system shows no abnormality, but the automatic image analysis conclusion or the manual drawing review conclusion shows an abnormality, it is considered that the inspection task conclusion corresponding to the target inspection task is a false alarm.

[0111] Structured data information can also be formed for the inspection tasks with false alarms, and the data relationships in Table 2 can be established.

[0112] Table 2:

[0113]

[0114] When there are multiple items to be inspected in the target inspection task, the system may not be able to accurately identify the product names of the items to be inspected corresponding to the false alarm content. At this time, the items to be inspected can be corrected through manual maintenance, input, or modification functions to accurately establish the corresponding relationship between the items to be inspected and the false alarm items.

[0115] Finally, according to the content shown in Table 2, the false alarm information table for each item to be inspected is updated, as exemplified below:

[0116] Product code: 10000000

[0117] Product name: Product B

[0118] False alarms by the manual drawing review system in the recent month: 10 times

[0119] False alarms by the image analysis system in the recent month: 2 times

[0120] Number of images in the recent month: 500

[0121] Version number: V2.0.0

[0122] False alarm feature value: 77

[0123] ……

[0124] Among them, the value of the false alarm feature value increases with the increase in the number of false alarms. The increased value can be the same each time a false alarm occurs, or it can increase with the increase in the number of false alarms and the false alarm frequency. In some embodiments, the false alarm feature value can also be generated based on the data in the recent preset period (such as the recent month). Thus, the false alarm feature value can be updated daily or weekly. The value of the false alarm feature value will not only increase with the increase in the number of false alarms but also decrease with the decrease in the recent number of false alarms to reflect the improvements brought about by factors such as the optimization of the image analysis model and the improvement of the skills of the drawing review personnel.

[0125] The above-mentioned recording of seizure information and false alarm information can also be achieved by building and maintaining a knowledge base system.

[0126] Figure 6It is a flowchart of information update of the knowledge base system in an embodiment of the present disclosure.

[0127] Refer to Figure 6 , in an embodiment, the process of updating the information of the knowledge base system according to the historical inspection tasks may include:

[0128] Step S61, run the machine inspection knowledge base data extraction tool;

[0129] Step S62, access the historical data of the machine inspection system;

[0130] Step S63, select a historical inspection task;

[0131] Step S64, determine whether the automatic image analysis conclusion or the manual drawing review conclusion of this task is a false alarm. If so, go to step S53; otherwise, go to step S66;

[0132] Step S65, parse the task information, obtain and save the product code of the item to be inspected, the automatic image analysis false alarm conclusion or the manual drawing review false alarm conclusion;

[0133] Step S66, determine whether the final inspection conclusion of this task finds any abnormality or there is a seizure in the subsequent disposal. If so, go to step S67; otherwise, go to step S68;

[0134] Step S67, parse the task information, obtain and save the product code of the item to be inspected, the inspection abnormality conclusion or the seized item and the seizure location;

[0135] Step S68, determine whether it is the last historical inspection task. If so, end the information extraction process; otherwise, return to step S63 to continue extracting historical information.

[0136] Figure 6 The data extraction process of the illustrated embodiment can be executed regularly to establish and maintain the knowledge base system to record seizure information and false alarm information.

[0137] Figure 7 It is an information update diagram of the knowledge base system in another embodiment of the present disclosure.

[0138] Refer to Figure 7 Refer to Figure 7 , in another embodiment, the process of updating the information of the knowledge base system according to the real-time inspection tasks may include:

[0139] Step S71, the machine inspection task process ends;

[0140] Step S72, determine whether the automatic image analysis conclusion or the manual drawing review conclusion of this task is a false alarm. If so, go to step S73; otherwise, go to step S74;

[0141] Step S73: Analyze the task information, and obtain and save the product code of the item to be inspected, the false alarm conclusion of automatic image analysis, or the false alarm conclusion of manual drawing inspection.

[0142] Step S74: Determine whether any abnormality is found in the final inspection conclusion of this task. If yes, proceed to Step S75; otherwise, proceed to Step S76.

[0143] Step S75: Analyze the task information, and obtain and save the product code of the item to be inspected, the abnormal conclusion found during inspection, or the seized items and the seizure location.

[0144] Step S76: Regularly obtain the feedback task seizure information on subsequent handling by other systems.

[0145] Step S77: Analyze the task seizure information, and add or update the corresponding false alarm and seizure information in the knowledge base.

[0146] Figure 7 The data extraction process of the illustrated embodiment can be executed after the completion of each machine inspection task process to update the seizure information and false alarm information in the knowledge base system. In addition, regularly obtaining the feedback task seizure information on subsequent handling by other systems can last for a preset duration (such as 3 days or 7 days) to cover the possible inspection actions on the inspection object (container / suitcase) of this machine inspection task.

[0147] Figure 6 and Figure 7 The inspection conclusion in the illustrated embodiment is the final inspection result of this task, including but not limited to the conclusion of unpacking and searching, the conclusion of working dog search, the subsequent law enforcement seizure conclusion of other law enforcement agencies, etc.

[0148] Figure 8 is the task processing flow chart of the machine inspection system in an embodiment of the present disclosure.

[0149] Reference Figure 8 , the task processing flow of the machine inspection system may include:

[0150] Step S81: The scanning system 21 scans the inspection object of the target inspection task to generate a scanned image.

[0151] Step S82: The information receiving module 201 in the task processing system 20 obtains the task information corresponding to the target inspection task. The task information includes the scanned image and the item information, and the item information includes the product code.

[0152] Step S83: The image processing module 202 in the task processing system 20 identifies the scanned image and determines the product code of the item to be inspected according to the information in the information storage module 203.

[0153] Step S84: The comprehensive processing module 204 in the task processing system 20 obtains the corresponding false alarm and seizure information in the knowledge base according to the product code of the item to be inspected. When the knowledge base is an external system, the false alarm and seizure information is obtained through the information receiving module 201. When the knowledge base is stored in the information storage module 203, the information storage module 203 is directly read.

[0154] Step S85: The comprehensive processing module 204 in the task processing system 20 obtains an automatic image analysis conclusion based on the scanned image, its corresponding item code, and historical false alarm and seizure information. The automatic image analysis conclusion includes generating a manual image review task, no abnormality found, and abnormality found.

[0155] Step S86: The comprehensive processing module 204 determines whether to generate a manual image review task. If yes, it proceeds to step S87; otherwise, it transfers to step S89.

[0156] Step S87: When the image review conclusion is to generate a manual image review task, the comprehensive processing module 204 generates a manual image review task and sends the manual image review task to the manual image review system 22.

[0157] Step S88: The manual image review system 22 assigns and displays the manual image review task according to the false alarm and seizure information provided by the knowledge base, and receives the manual image review conclusion of the manual image review task.

[0158] Step S89: The image review task ends, and the information on the execution of this task is extracted in real time to update the knowledge base.

[0159] Although Figure 8 the illustrated process is described by taking the behaviors of each module and system as examples, in other embodiments of the present disclosure, there can be various physical and logical divisions of the modules and physical and logical divisions of the systems. The present disclosure is not limited to the above division of the execution modules / systems.

[0160] In summary, the embodiments of the present disclosure record the corresponding relationships between the seized items and the items to be inspected, and between the false alarm items and the items to be inspected. After the inspection task process ends, the inspection task results can be extracted and analyzed, or the historical inspection data can be extracted and analyzed through a tool software, so as to optimize the processing flow of the inspection task according to the historical false alarm and seizure information in subsequent inspection tasks, directly generate a manual image review task for the inspection tasks with higher risks, improve the processing efficiency of the inspection tasks, effectively reduce the false alarm rate of image analysis, and improve the accuracy of image analysis.

[0161] When providing false alarm information and seizure information of such inspection items to inspection personnel during manual drawing inspection, and giving corresponding missed inspection prompts in the way of highlighting and flashing display lights, allowing inspection personnel to view the historical false alarms and seizure information of such inspection items during manual drawing inspection, and providing inspection record sheets and historical images for comparison and analysis, the efficiency and accuracy of manual drawing inspection can be improved.

[0162] That is, the embodiments of the present disclosure can not only assist in the analysis of manual drawing inspection for machine inspection, but also provide a set of corrective data for the machine inspection task processing system, improving the accuracy of machine inspection and reducing the false alarm rate of image analysis.

[0163] In some embodiments of the present disclosure, false alarm images can be collected based on a large number of false alarm and seizure case information, and a training set can be formed subsequently to train the automatic image recognition process of the task processing system 20 using missed inspection and false alarm cases, optimizing the image analysis ability of the task processing system 20.

[0164] In addition, a large number of false alarm and missed inspection case information can be used to provide targeted and guiding training for machine inspection image analysis personnel, improving the overall experience ability of machine inspection personnel, optimizing the recognition accuracy and the recognition accuracy of manual drawing inspection, while improving the drawing inspection efficiency and drawing inspection accuracy.

[0165] Corresponding to the above method embodiments, the present disclosure also provides an inspection task processing device, which can be used to execute the above method embodiments.

[0166] Figure 9 It is a block diagram of an inspection task processing device in an exemplary embodiment of the present disclosure.

[0167] Refer to Figure 9 , the inspection task processing device 900 may include:

[0168] A task information determination module 91, configured to obtain task information corresponding to a target inspection task, where the task information includes an item code and a scanned image of an item to be inspected corresponding to the target inspection task;

[0169] A risk information determination module 92, configured to obtain a risk characteristic value of the item to be inspected according to the task information, where the risk characteristic value includes a missed inspection characteristic value and / or a false alarm characteristic value;

[0170] A manual drawing inspection task generation module 93, configured to determine whether to generate a manual drawing inspection task based on the target inspection task according to the risk characteristic value corresponding to each item to be inspected.

[0171] Since the functions of the device 900 have been described in detail in their corresponding method embodiments, the present disclosure will not repeat them here.

[0172] It should be noted that although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of the two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0173] In an exemplary embodiment of the present disclosure, an electronic device capable of implementing the above method is also provided.

[0174] Those skilled in the art can understand that various aspects of the present invention can be implemented as a system, a method, or a program product. Therefore, various aspects of the present invention can be specifically implemented in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or an implementation combining hardware and software aspects, which can be collectively referred to as "circuit", "module", or "system" here.

[0175] The following refers to Figure 10 to describe the electronic device 1000 according to this embodiment of the present invention. Figure 10 The shown electronic device 1000 is only an example and should not impose any limitation on the functions and usage scope of the embodiments of the present invention.

[0176] As Figure 10 shown, the electronic device 1000 is presented in the form of a general-purpose computing device. The components of the electronic device 1000 may include but are not limited to: the above at least one processing unit 1010, the above at least one storage unit 1020, and a bus 1030 connecting different system components (including the storage unit 1020 and the processing unit 1010).

[0177] Among them, the storage unit stores program codes, and the program codes can be executed by the processing unit 1010, so that the processing unit 1010 executes the steps according to various exemplary embodiments of the present invention described in the above "Exemplary Method" section of this specification. For example, the processing unit 1010 can execute the method as shown in the embodiments of the present disclosure.

[0178] The storage unit 1020 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 10201 and / or a cache storage unit 10202, and may further include a read-only storage unit (ROM) 10203.

[0179] The storage unit 1020 may also include a program / utilities 10204 having a set (at least one) of program modules 10205. Such program modules 10205 include, but are not limited to: an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment.

[0180] The bus 1030 may represent one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus structures.

[0181] The electronic device 1000 may also communicate with one or more external devices 1100 (such as a keyboard, a pointing device, a Bluetooth device, etc.), and may also communicate with one or more devices that enable a user to interact with the electronic device 1000, and / or communicate with any device that enables the electronic device 1000 to communicate with one or more other computing devices (such as a router, a modem, etc.). Such communication may be carried out through the input / output (I / O) interface 1050. Moreover, the electronic device 1000 may also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through the network adapter 1060. As shown in the figure, the network adapter 1060 communicates with other modules of the electronic device 1000 through the bus 1030. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 1000, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0182] Through the description of the above embodiments, those skilled in the art can easily understand that the example embodiments described herein can be implemented by software, or can be implemented by the way of software combined with necessary hardware. Therefore, the technical solutions according to the embodiments of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which may be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which may be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the embodiments of the present disclosure.

[0183] In an exemplary embodiment of the present disclosure, there is also provided a computer-readable storage medium having a program product stored thereon that can implement the above-described method of this specification. In some possible implementation manners, various aspects of the present invention can also be implemented in the form of a program product, which includes program code. When the program product runs on a terminal device, the program code is used to cause the terminal device to execute the steps according to various exemplary embodiments of the present invention described in the above "Exemplary Method" section of this specification.

[0184] The program product for implementing the above method according to an embodiment of the present invention may be a portable compact disc read-only memory (CD-ROM) and includes program code, and can run on a terminal device, such as a personal computer. However, the program product of the present invention is not limited thereto. In this document, the readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0185] The program product may adopt any combination of one or more readable media. The readable media may be a readable signal medium or a readable storage medium. The readable storage medium may, for example, but not be limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0186] The computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries the readable program code. Such a propagated data signal may take various forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The readable signal medium may also be any readable medium other than the readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0187] The program code contained on the readable medium can be transmitted by any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination of the above.

[0188] Program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages such as Java, C++, etc., and also including conventional procedural programming languages such as the "C" language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user's computing device through any kind of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0189] In addition, the above drawings are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present invention, and are not for limiting purposes. It is easy to understand that the processes shown in the above drawings do not indicate or limit the chronological order of these processes. Additionally, it is also easy to understand that these processes may be executed, for example, synchronously or asynchronously in multiple modules.

[0190] Other embodiments of the present disclosure will be readily apparent to those skilled in the art after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include known common knowledge or conventional technical means in the technical field not disclosed herein. The specification and embodiments are to be considered exemplary only, and the true scope and spirit of the present disclosure are pointed out by the claims.

Claims

1. A method for processing inspection tasks, characterized in that, Including: Obtain task information corresponding to a target inspection task, where the task information includes the item code and scanned image of the item to be inspected corresponding to the target inspection task; Obtain the risk characteristic value of the item to be inspected according to the task information, where the risk characteristic value includes a missed declaration characteristic value and / or a false alarm characteristic value; Determine whether to generate an artificial drawing review task based on the target inspection task according to the risk characteristic value corresponding to each item to be inspected; Among them, the determining whether to generate an artificial drawing review task based on the target inspection task according to the risk characteristic value corresponding to each item to be inspected includes: When the risk characteristic value of any one of the items to be inspected meets a preset condition, generate an artificial drawing review task according to the target inspection task; The risk characteristic value includes a missed declaration characteristic value, the preset condition includes that the missed declaration characteristic value exceeds a first preset value, and the generating an artificial drawing review task according to the target inspection task includes: When the missed declaration characteristic value of an item to be inspected exceeds the first preset value, generate an artificial drawing review task and display the scanned image, and display the missed declaration risk reminder information of the item to be inspected in the interface of the artificial drawing review task.

2. The inspection task processing method according to claim 1, wherein The determining whether to generate an artificial drawing review task based on the target inspection task according to the risk characteristic value corresponding to each item to be inspected includes: When the risk characteristic value of each item to be inspected does not meet the preset condition, identify the scanned image and generate a processing result for the target inspection task according to the recognition result of the scanned image.

3. The inspection task processing method according to claim 1, characterized in that The generating an artificial drawing review task according to the target inspection task further includes: When the missed declaration characteristic value of an item to be inspected exceeds the first preset value, obtain the historical risk image of the item to be inspected and provide the historical risk image, where the historical risk image is a historical scanned image related to the increase in the risk characteristic value of the item to be inspected.

4. The inspection task processing method according to claim 1, wherein, The risk characteristic value includes a false alarm characteristic value, the preset condition includes that the false alarm characteristic value exceeds a second preset value, and the generating an artificial drawing review task according to the target inspection task includes: When the false alarm characteristic value of an item to be inspected exceeds the second preset value, generate an artificial drawing review task and display the scanned image, and display the false alarm risk reminder information of the item to be inspected in the interface of the artificial drawing review task.

5. The inspection task processing method according to any one of claims 1 to 4, characterized in that, Also including: Obtain the image analysis result of the target inspection task and the re-inspection result corresponding to the target inspection task, where the image analysis result includes an automatic image analysis result and an artificial drawing review recognition result; When the image analysis result shows no abnormality and the re-inspection result includes an abnormality, determine the item to be inspected corresponding to the re-inspection result and the abnormal area of the item to be inspected, increase the missed declaration characteristic value of the item to be inspected, and record the abnormal area of the item to be inspected as the risk area of the item to be inspected; When the image analysis result is abnormal and the re-inspection result shows no abnormality, determine the item to be inspected corresponding to the recognition result and increase the false alarm characteristic value of the item to be inspected.

6. The inspection task processing method according to claim 5, wherein, The retest result includes at least one of a re-image analysis result of the target inspection task, an inspection result of the target inspection task, and an external inspection result related to the target inspection task.

7. A verification task processing device, characterized in that, including: a task information determination module configured to obtain task information corresponding to a target inspection task, where the task information includes an item code and a scanned image of an item to be inspected corresponding to the target inspection task; a risk information determination module configured to obtain a risk characteristic value of the item to be inspected according to the task information, where the risk characteristic value includes a missed report characteristic value and / or a false report characteristic value; an artificial drawing review task generation module configured to determine whether to generate an artificial drawing review task based on the target inspection task according to the risk characteristic value corresponding to each item to be inspected; wherein, the artificial drawing review task generation module is configured to generate an artificial drawing review task according to the target inspection task when the risk characteristic value of any one of the items to be inspected meets a preset condition; the risk characteristic value includes a missed report characteristic value, and the preset condition includes that the missed report characteristic value exceeds a first preset value, and generating an artificial drawing review task according to the target inspection task includes: when the missed report characteristic value of an item to be inspected exceeds the first preset value, generating an artificial drawing review task and displaying the scanned image, and displaying missed report risk reminder information of the item to be inspected in the interface of the artificial drawing review task.

8. An electronic device, characterized in that, including: a memory; and a processor coupled to the memory, the processor being configured to execute the inspection task processing method according to any one of claims 1-6 based on instructions stored in the memory.

9. A computer-readable storage medium, on which a program is stored, and when the program is executed by a processor, the inspection task processing method according to any one of claims 1-6 is implemented.

Images