Customs cargo detection system and method

Abstract

The invention provides a customs cargo detection system and method, and the method comprises the steps: an information checking module can check an electronic file of a to-be-detected customs cargo, and transmits the electronic file and an obtained checking result to an information tracking module, so that the information tracking module can obtain a first parameter corresponding to the to-be-detected customs cargo; according to the mathematical models corresponding to the plurality of risk assessment factors and the first parameter, the risk level corresponding to the to-be-detected customs cargo can be assessed, the detection module can detect the to-be-detected customs cargo according to the target detection path matched with the risk level, and the customs clearance efficiency of the to-be-detected customs cargo can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of customs cargo security inspection, and in particular to a customs cargo inspection system and method. Background Art

[0002] As an innovative regulatory model, smart customs has become a key way to cope with the growing customs business volume. However, among related technologies, the development potential of smart customs has not yet been fully tapped, resulting in low customs clearance efficiency for customs goods. Summary of the invention

[0003] The object of the present invention is to provide a customs cargo inspection system and method to improve the customs cargo clearance efficiency.

[0004] The present invention provides a customs cargo inspection system, which includes: an information inspection module, an information tracking module and a detection module; the information inspection module is used to receive declaration list information corresponding to the customs cargo to be inspected, generate an electronic file and an electronic identity tag corresponding to the customs cargo to be inspected according to the declaration list information, and inspect the archive information in the electronic file based on a pre-connected database to obtain the inspection result, send the electronic file and the inspection result to the information tracking module, and send the electronic identity tag to the detection module; the information tracking module is used to obtain mathematical models corresponding to a plurality of preset risk assessment factors, obtain a first parameter corresponding to the customs cargo to be inspected based on the preset historical archive, the electronic archive and the inspection result, determine the risk level corresponding to the customs cargo to be inspected based on each mathematical model and the first parameter, and determine the target detection path corresponding to the customs cargo to be inspected based on the risk level; wherein the plurality of risk assessment factors include: customs clearance frequency, trajectory behavior, historical information and specified characteristics; the first parameter includes: the number of customs clearances of the declaring party corresponding to the customs goods to be inspected in the first historical time period, the number of customs clearances of the associated party corresponding to the declaring party in the first historical time period, the customs clearance frequency factor corresponding to the goods belonging to the same category as the customs goods to be inspected in the first historical time period, the historical customs clearance qualification rate of the declaring party and the associated party in the first historical time period, the historical punctuality rate of the declaring party and the associated party in the first historical time period, the probability of carrying risky substances in the transportation path, the punctuality rate of arrival at the customs inferred by the preset logistics path model this time, the punctuality rate of arrival at the customs inferred by the logistics path model of the freight ship carrying the customs goods to be inspected last time, the operating and financial status of the declaring party, the legal risks of the declaring party, and the material integrity of the declaration list information; the detection module is used to verify the identity of the customs goods to be inspected according to the electronic identity tag. If the verification is successful, the customs goods to be inspected are inspected according to the target detection path to obtain the target detection result.

[0005] The present invention provides a method for detecting customs goods, which includes: an information checking module receives declaration list information corresponding to customs goods to be detected, generates an electronic file and an electronic identity tag corresponding to the customs goods to be detected according to the declaration list information, checks the file information in the electronic file based on a pre-connected database to obtain the inspection result, sends the electronic file and the inspection result to the information tracking module, and sends the electronic identity tag to the inspection module; the information tracking module obtains mathematical models corresponding to a plurality of preset risk assessment factors, obtains a first parameter corresponding to the customs goods to be detected based on a preset historical file, the electronic file and the inspection result, determines the risk level corresponding to the customs goods to be detected based on each mathematical model and the first parameter, and determines the target detection path corresponding to the customs goods to be detected based on the risk level; wherein the plurality of risk assessment factors include: customs clearance frequency, trajectory behavior, historical information and specified characteristics; the first parameter Including: the number of customs clearances of the declaring party corresponding to the customs goods to be inspected in the first historical time period, the number of customs clearances of the associated parties corresponding to the declaring party in the first historical time period, the customs clearance frequency factor corresponding to the goods belonging to the same category as the customs goods to be inspected in the first historical time period, the historical customs clearance qualification rate of the declaring party and the associated parties in the first historical time period, the historical punctuality rate of the declaring party and the associated parties in the first historical time period, the probability of carrying risky substances in the transportation route, the punctuality rate of arrival at the customs inferred by the preset logistics path model this time, the punctuality rate of arrival at the customs inferred by the logistics path model of the freight ship carrying the customs goods to be inspected last time, the operating and financial status of the declaring party, the legal risks of the declaring party, and the material integrity of the declaration list information; the detection module verifies the identity of the customs goods to be inspected according to the electronic identity tag. If the verification is successful, the customs goods to be inspected are inspected according to the target detection path to obtain the target detection result.

[0006] The customs goods inspection system and method provided by the present invention can be used to inspect the electronic files of the customs goods to be inspected by the information inspection module, and send the electronic files and the inspection results to the information tracking module, so that the information tracking module can obtain the first parameter corresponding to the customs goods to be inspected. According to the mathematical model corresponding to multiple risk assessment factors and the first parameter, the risk level corresponding to the customs goods to be inspected can be evaluated and obtained. The inspection module can inspect the customs goods to be inspected according to the target inspection path matching the risk level, which helps to improve the customs clearance efficiency of the customs goods to be inspected. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0008] Figure 1 A schematic diagram of a customs cargo inspection system provided by an embodiment of the present invention; Figure 2 A schematic diagram of the layout of an unmanned inspection park provided by an embodiment of the present invention; Figure 3 A schematic diagram of an action route of a weighing robot provided in an embodiment of the present invention; Figure 4 An overall schematic diagram of a separate detection zone provided by an embodiment of the present invention; Figure 5 A schematic diagram of a single detection room provided by an embodiment of the present invention; Figure 6 A schematic diagram of a weighing robot provided by an embodiment of the present invention; Figure 7 A schematic diagram of a detection robot provided by an embodiment of the present invention; Figure 8 An overall schematic diagram of a joint inspection area provided by an embodiment of the present invention; Fig. 9 A schematic diagram of a joint detection room provided by an embodiment of the present invention; Fig.10 A schematic diagram of a customs cargo inspection system provided by an embodiment of the present invention; Fig.11 A schematic diagram of another customs cargo inspection system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0009] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0010] As an innovative supervision model, smart customs has become a key way to cope with the growing volume of customs business, and has far-reaching significance for improving logistics efficiency, reducing costs, strengthening supervision and security. As an important part of digital transformation, smart customs has significantly improved work efficiency and ensured smooth and safe trade through early information layout and deep digital application, the construction of digital infrastructure and the application of intelligent supervision technology development. However, in actual applications, the synergistic development potential between various intelligent technologies (such as machine learning, artificial intelligence, big data processing, 5G communication, Internet of Things, etc.) has not been fully explored, and there is a problem of information asymmetry between traditional operation management and lack of integrated operation and supervision capabilities. At present, there is still a lack of comprehensive smart customs management systems on the market that can complete real-time data collection, intelligent integration, monitoring feedback, and visual operation, resulting in the problem of low customs clearance efficiency of related technologies.

[0011] To facilitate understanding of this embodiment, a customs cargo inspection system disclosed in an embodiment of the present invention is first introduced. Figure 1 As shown, the system includes: an information checking module 10, an information tracking module 11 and a detection module 12; the information checking module 10 is used to receive the declaration list information corresponding to the customs goods to be detected, generate the electronic file and electronic identity tag corresponding to the customs goods to be detected according to the declaration list information, check the file information in the electronic file based on the pre-connected database, obtain the inspection result, send the electronic file and the inspection result to the information tracking module 11, and send the electronic identity tag to the detection module 12; The above-mentioned customs goods to be tested refer to commodities that need to be approved and supervised by the customs. These commodities need to be inspected and supervised by the customs during the import and export process. The above-mentioned declaration list information usually includes: customs declaration form information, packing list information, invoice, contract and other information; the above-mentioned electronic archive can be understood as entering the declaration list information of the customs goods to be tested into a digital system to create an electronic record containing all necessary information. The electronic archive usually includes the following content: product source, type and quantity, processing and manufacturing method, transaction flow and other link data of the customs goods to be tested; the above-mentioned electronic identity tag can be represented by numbers, letters, etc., and can also be in the form of a two-dimensional code, etc., which can be set according to actual needs, and is not limited here. Each customs goods to be tested has a unique corresponding electronic identity tag, and the identity of the customs goods to be tested can be identified by the electronic identity tag. There can be many types of the above-mentioned database, for example, it can include a database platform corresponding to the tax department, a database platform corresponding to the business department, a database platform of the ship operation management and logistics transportation department, etc., and can be connected to different database platforms for the information inspection module 10 according to actual needs, which is not limited here.

[0012] In actual implementation, customs clearance laws and regulations, policy documents, etc. can be input into the information verification module 10 in advance, and different database platforms can be connected to the information verification module 10 in advance. Then, the information verification module 10 can establish a corresponding electronic file for the customs goods to be tested based on the declaration list information of the customs goods to be tested submitted by the declarant, and can use QR code technology to generate an electronic identity tag for identity identification in the customs park; the information verification module 10 can verify the authenticity of the file information in the electronic file based on the information in the pre-connected database to verify its accuracy and legality, and obtain the inspection results, for example, based on the authenticity and completeness of the invoices and trade contracts of the customs goods to be tested in the database of the tax department, and based on the commercial The database of the service platform verifies the origin, source, batch number and other information of the customs goods to be inspected, and verifies the authenticity of the logistics information based on the database of the ship operation management system and the logistics transportation department; and inputs the inspection result and the electronic identity tag into the information tracking module 11. For example, the inspection result can be that the electronic file is authentic, or that the electronic file has an abnormality. If there is an abnormality, the corresponding display terminal can also display the abnormal information, such as whether there are contraband, counterfeit goods, etc.; In addition, the inspection strategy can be adjusted for specific types during special periods, where special periods refer to: sudden epidemics require the detection of specific infectious agents or easy-to-carry carriers. For example, if an animal epidemic occurs, special monitoring of animals and meat products is required. That is, the information inspection module 10 can access different information for inspection based on the differences in ports and changes in actual conditions, rather than remaining unchanged.

[0013] The information tracking module 11 is used to obtain mathematical models corresponding to a plurality of preset risk assessment factors, obtain a first parameter corresponding to the customs goods to be inspected based on preset historical archives, electronic archives and inspection results, determine the risk level corresponding to the customs goods to be inspected based on each mathematical model and the first parameter, and determine the target inspection path corresponding to the customs goods to be inspected based on the risk level; wherein the plurality of risk assessment factors include: customs clearance frequency, trajectory behavior, historical information and specified characteristics; the first parameter includes: the number of customs clearances of the declaring party corresponding to the customs goods to be inspected in the first historical time period, the number of customs clearances of the associated party corresponding to the declaring party in the first historical time period, the customs clearance frequency factor corresponding to the goods belonging to the same category as the customs goods to be inspected in the first historical time period, the historical customs clearance qualified rate of the declaring party and the associated party in the first historical time period, the historical punctuality rate of the declaring party and the associated party in the first historical time period, the probability of carrying risky substances in the transportation path, the punctuality rate of arrival at the customs inferred by the preset logistics path model this time, the punctuality rate of arrival at the customs inferred by the logistics path model of the freight ship carrying the customs goods to be inspected last time, the operating and financial status of the declaring party, the legal risk of the declaring party, and the material integrity of the declaration list information; The above-mentioned historical archives usually record the archival information of all goods that have been inspected by the customs within a specified historical time period. In this embodiment, the electronic archives of the customs goods to be inspected can be uploaded and saved in the historical archives; in this embodiment, multiple risk assessment factors may include: customs clearance frequency, trajectory behavior, historical information and specified characteristics; wherein the customs clearance frequency may include the number of customs clearances and customs clearance records of the declaring party corresponding to the customs goods to be inspected and its corresponding related parties within the first historical time period. According to the customs clearance frequency, it is helpful to identify patterns that may have abnormal activities. Usually, the customs clearance frequency should comply with trade policies, market demands, and enterprises should have a complete supply chain of such goods in the country. For example, daily necessities, electronic products, food and other products with stable demand that have been applied for by the declaring party for many times have experienced drastic changes in import volume within a certain time interval, which is considered abnormal; the import frequency of the declaring party's imported goods that does not comply with seasonal changes is considered abnormal; for example, the declaring party's import frequency from the war-torn origin is too high, which is considered abnormal, etc. The trajectory behavior may include: the associated parameters of the customs goods to be tested in the process of arriving at the final destination from the departure place; the associated parameters include at least one of the following: the country of origin of the customs goods to be tested, the port of departure, the parameters of the ship transporting the customs goods to be tested (such as age, ship parameters, loading capacity, etc.), the cargo information of the customs goods to be tested (such as quantity, weight, mention, special storage environment, etc.), the weather conditions of the sea areas along the way, and the logistics distribution conditions of the domestic section, etc. Constraints (such as transportation costs, path safety, etc.) can be introduced, and linear programming, dynamic programming and other modeling methods can be used to construct a logistics path model for the declarant. The logistics path model can be used to infer the time when the customs goods to be tested arrive at the customs; historical information may include: a retrospective review of all declaration list information submitted by the declarant in the first historical time period, focusing on the types and quantities of goods declared by the declarant. The characteristics of the customs goods to be tested may include quantity, quantity fluctuation trend, historical penalty records for illegal operations, etc. It may also include pathogens or alien invasive species that the customs goods to be tested themselves and their places of origin are prone to carry in specific seasons or periods, and the distribution of biological and chemical pollution sources that may exist in each link of the entire supply chain; the above-mentioned specified characteristics may include: the difference in the domestic and foreign sales ratio of the declarant corresponding to the customs goods to be tested, the financial status, daily operational compliance and legal disputes, the completeness of the declaration list information, etc. It is particularly worth noting that the completeness of the declaration list information means supporting the declarant to exchange data for convenience. Compared with only providing a small amount of materials with no abnormalities in the basic requirement documents, the system focuses on the real and effective amount of materials provided by the declarant for evaluation, and accumulates credit based on the data materials; the specific content that needs to be considered in each risk assessment factor can be selected according to actual needs and is not limited here.

[0014] The number of risk levels can be set according to actual needs, for example, three levels can be included, namely, high risk level, medium risk level and low risk level; the number of target detection paths is usually the same as the number of risk levels, that is, different risk levels usually correspond to different target detection paths; the first historical time period can be set according to actual needs, for example, it can be a time period in the last six months, etc.; the related parties can be understood as companies that have equity relations (such as holding 25% or more) with the declarant, senior managers or directors, units with delegation relations, companies with direct or indirect control relations in terms of funds, etc. queried in the industrial and commercial management system connected to the information verification module 10; the customs clearance frequency factor can be obtained in advance, and the customs clearance frequency factor usually refers to the relevant customs clearance information of other declarants and related parties at the same time. Usually, different customs clearance frequency factors corresponding to different customs goods to be tested are different, and different first historical time periods correspond to different customs clearance frequency factors; the historical customs clearance pass rate can be understood as the proportion of customs goods with qualified customs clearance of the declarant and related parties of the customs goods to be tested in the first historical time period to all customs goods of the declarant and related parties. The above historical punctuality rate can be understood as the punctuality rate of all customs goods declared for customs clearance by the declaring party and related parties of the customs goods to be inspected in the first historical time period. The probability of carrying risky substances in the above transportation route can be understood as the possibility that the customs goods to be inspected carry risky substances in the entire transportation route from the departure place to the destination. The above punctuality rate of arrival at the customs inferred by the preset logistics path model can be understood as: the cargo ship loaded with customs goods to be inspected usually needs to declare 24 hours before entering the port for customs clearance. According to the time of its declaration and the location information when the declaration information is uploaded, the arrival time of the cargo ship at the customs can be inferred in combination with the departure time. If the ship deviates during the voyage, this value will change greatly. The punctuality rate can be calculated based on the inferred arrival time and the actual arrival time. This value can reflect the probability of introducing unplanned or unexpected risks to a certain extent. The operating and financial status of the above declaring party can be expressed in numbers, for example, it can be expressed in numbers between 0 and 10, where 0 indicates a poor operating and financial status and 10 indicates a good operating and financial status. The above-mentioned legal risks can be understood as the possibility of negative legal liability or consequences for the declarant due to changes in the external legal environment of the declarant or the declarant's own actions or inactions based on legal provisions or contractual agreements; the above-mentioned material completeness can be used to characterize the completeness of the declaration list information, for example, it can be expressed as a number between 0-5, where 0 indicates very poor material integrity and 5 indicates good material integrity, etc.

[0015] In this embodiment, multiple risk assessment factors and mathematical models corresponding to each risk assessment factor can be pre-configured. The information tracking module 11 can obtain each mathematical model, and obtain the first parameter corresponding to the customs goods to be inspected according to the preset historical archives, the electronic archives corresponding to the customs goods to be inspected, and the inspection results, so as to evaluate the risk level corresponding to the customs goods to be inspected according to each mathematical model and the first parameter, and then determine the corresponding target detection path according to the risk level, that is, different detection methods can be used for customs goods to be inspected with different risk levels. For example, the first-level risk corresponds to a separate detection path, the second-level risk corresponds to a joint detection path, and the third-level risk can be directly cleared, etc., among which the first-level risk corresponds to high risk, the second-level risk corresponds to medium risk, and the third-level risk corresponds to low risk. This risk level grading method can achieve rapid customs clearance for low-risk customs goods and accurate detection of high-risk customs goods.

[0016] The detection module 12 is used to verify the identity of the customs goods to be detected according to the electronic identity tag. If the verification is successful, the customs goods to be detected are detected according to the target detection path to obtain the target detection result. The detection module 12 can verify the identity of the customs goods to be detected according to the electronic identity tag of the customs goods to be detected. After the verification is successful, the customs goods to be detected can be detected according to the target detection path. For example, the weight, quality, composition, safety performance, etc. of the customs goods to be detected can be detected to obtain the target detection result corresponding to the customs goods to be detected.

[0017] In the above-mentioned customs cargo inspection system, the information verification module can inspect the electronic files of the customs cargo to be inspected, and send the electronic files and the inspection results to the information tracking module, so that the information tracking module can obtain the first parameter corresponding to the customs cargo to be inspected. According to the mathematical model corresponding to multiple risk assessment factors and the first parameter, the risk level corresponding to the customs cargo to be inspected can be evaluated. The inspection module can inspect the customs cargo to be inspected according to the target inspection path matching the risk level, which helps to improve the customs clearance efficiency of the customs cargo to be inspected.

[0018] Furthermore, the information tracking module is further used to: calculate a first evaluation result corresponding to the customs clearance frequency based on a first mathematical model corresponding to the customs clearance frequency and a first specified parameter in the first parameter; The above-mentioned first mathematical model is: S1=T1×(1+k)+T2×(1+k)+T3×k-L2×k×f; wherein S1 represents the first evaluation result; T1 represents the number of customs clearances of the declaring party in the first historical time period; T2 represents the number of customs clearances of the related party in the first historical time period; T3 represents the customs clearance frequency factor corresponding to the goods belonging to the same category as the customs goods to be inspected in the first historical time period; L2 represents the historical punctuality rate of the declaring party and the related party in the first historical time period; these data T1-T3, L2 can be determined based on historical archives; k and f are preset coefficients, and the specific values ​​can be adjusted according to the historical records of the customs port using this technology; The calculation formula corresponding to T3 is: ; Among them, T1 represents the number of customs clearances of the declarant in the first historical time period; t i It indicates the time difference between the customs clearance time of the declarant at the i-th customs clearance and the customs clearance time of the previous customs clearance in the first historical time period. For example, the customs clearance time corresponding to the third customs clearance is 2 hours, which can also be understood as the total duration corresponding to the arrival at the customs and passing through the customs at the third customs clearance of the declarant is 2 hours; the corresponding customs clearance time of the previous customs clearance, i.e. the second customs clearance, is 1.5 hours, then t3 is: 2-1.5=0.5 hours; It represents the average of the time difference between the customs clearance times of two consecutive customs clearances of goods belonging to the same category as the customs goods to be inspected in the first historical time period; It represents the standard deviation of the time difference between two adjacent customs clearances of goods belonging to the same category as the customs goods to be inspected in the first historical time period; f1 is a preset coefficient, which can be adjusted according to the historical records of the customs, and the preset coefficients corresponding to different customs can be the same or different.

[0019] The parameters in the first mathematical model correspond to the first specified parameters. The first specified parameters obtained can be substituted into the first mathematical model to calculate the first evaluation result corresponding to the clearance frequency.

[0020] Based on a second mathematical model corresponding to the trajectory behavior and a second specified parameter in the first parameter, a second evaluation result corresponding to the trajectory behavior is calculated; The second mathematical model is: S2 = G1 + G2 × L2 × (1 + j); where S2 represents the second evaluation result; G1 represents the on-time rate of arrival at the customs inferred by the preset logistics path model this time; G2 represents the on-time rate of arrival at the customs inferred by the logistics path model last time on the freight ship loaded with customs goods to be inspected; L2 represents the historical on-time rate of the declarant and the related party in the first historical time period; j is a preset coefficient, and the specific value can be adjusted according to the historical records of the customs port using this technology; Among them, the calculation formula corresponding to G1 is: ; Among them, t represents the estimated arrival time at the customs inferred by the preset logistics path model; Indicates the average of the historical arrival times corresponding to the transport path in the first historical time period. For example, if there are three transport records corresponding to the transport path in the first historical time period, the average of the three corresponding historical arrival times can be calculated. The historical arrival time can be understood as the duration of the transport corresponding to the transport path. For example, if the three historical arrival times are 20 hours, 21 hours and 22 hours respectively, the average can be calculated to be 21 hours. It represents the standard deviation of the historical arrival time corresponding to the transportation path in the first historical time period; is the polynomial coefficient, which can be adjusted according to the historical records of different customs. k1 is the polynomial coefficient, which represents the historical record number of the freight ship arriving at the customs in the first historical time period. The parameters in the second mathematical model correspond to the second specified parameters. The acquired second specified parameters can be substituted into the second mathematical model to calculate the second evaluation result corresponding to the trajectory behavior.

[0021] Based on a third mathematical model corresponding to the historical information and a third specified parameter in the first parameter, a third evaluation result corresponding to the historical information is calculated; The third mathematical model is: S3=L1×L2+L3×G1×j×u-T1×k×f; wherein S3 represents the third evaluation result; L1 represents the historical customs clearance qualification rate of the declarant and the related parties in the first historical time period; L2 represents the historical punctuality rate of the declarant and the related parties in the first historical time period; L3 represents the probability of carrying risky substances in the transportation path; these data L1-L3 can be determined based on historical archives; G1 represents the punctuality rate of arrival at the customs inferred by the preset logistics path model this time; T1 represents the number of customs clearances of the declarant in the first historical time period; j, u, k, f are preset coefficients, and the specific values ​​can be adjusted according to the historical records of the customs port using this technology; Among them, the calculation formula corresponding to L3 is: ; in, Historical risk data indicating the transport route, for example, may include the categories of risky substances carried by cargo ships on the transport route during a first historical time period; Represents data associated with environmental factors, such as For example, if the environment of the departure place or source country of the goods to be inspected is poor, the probability of carrying risky substances will be relatively high. A relatively large value can be taken. If the environment of the departure place or source country corresponding to the customs goods to be tested is good, the probability of carrying risky substances will be relatively low. Can take relatively small values; Represents data associated with time factors, such as For example, if the departure place or source country of the goods to be inspected is in the summer, the probability of risk substances appearing will be relatively high. A relatively large value can be taken. If the departure place or source country corresponding to the customs goods to be inspected is in winter, the probability of risk substances appearing will be relatively low. Can take relatively small values; , and are the weight coefficients of each factor respectively; The parameters in the third mathematical model correspond to the third specified parameters. The obtained third specified parameters can be substituted into the third mathematical model to calculate the third evaluation result corresponding to the trajectory behavior.

[0022] Based on a fourth mathematical model corresponding to the specified feature and a fourth specified parameter in the first parameter, a fourth evaluation result corresponding to the specified feature is calculated; The fourth mathematical model is: S4= Q1×Q2×(1+g)+Q3×(1+g)+L1×u; wherein S4 represents the fourth evaluation result; Q1 represents the operating and financial status of the declarant, Q2 represents the legal risk of the declarant, and Q3 represents the material completeness of the declaration list information; g and u are preset coefficients, and the specific values ​​can be adjusted according to factors such as high-risk cargo varieties and biological risks in the historical records of the customs port using this technology.

[0023] Among them, the calculation formula corresponding to Q1 is: ; Where q represents the current financial ratio of the applicant (for example, it may include debt-to-asset ratio and current ratio, etc.), It represents the average value of the financial ratio of the declarant during the first historical period; represents the standard deviation of the financial ratios of the declarant during the first historical period; , represents polynomial coefficients; represents the weight of the ith financial risk of the declarant in the first historical period; n represents the total number of financial risk assessments of the declarant in the first historical period; The parameters in the fourth mathematical model correspond to the fourth specified parameters. The fourth specified parameters obtained can be substituted into the fourth mathematical model to calculate the fourth evaluation result corresponding to the trajectory behavior.

[0024] According to the preset weight coefficient, the first evaluation result, the second evaluation result, the third evaluation result and the fourth evaluation result are weightedly calculated to obtain the comprehensive score result corresponding to the customs goods to be inspected; the above-mentioned weight coefficient can be set according to actual needs. Usually, different customs ports have different inspection focuses. For example, the transportation convenience of international routes will make the products of some countries and regions enter from a certain port. The inspection focus of different countries and regions will be different. The appropriate weight can be set for each risk assessment factor according to actual needs; in addition, for the same customs, different types of customs goods to be inspected that arrive at the customs can also use different weight coefficients, which can be flexibly set and are not limited here; after obtaining the above-mentioned first evaluation result, the second evaluation result, the third evaluation result and the fourth evaluation result, weighted calculation can be performed according to the preset weight coefficient to finally obtain the comprehensive score result corresponding to the customs goods to be inspected.

[0025] According to the comprehensive scoring results and the preset risk thresholds, determine the risk level corresponding to the customs goods to be inspected; the number of the above risk thresholds can be set according to actual needs. Different numbers of risk thresholds correspond to different numbers of risk levels. For example, if there are two risk thresholds, there are three risk levels that can be divided. In actual implementation, after calculating the final comprehensive scoring results of the customs goods to be inspected, the risk level corresponding to the customs goods to be inspected can be determined based on the threshold range of the comprehensive scoring results. For example, according to the comprehensive scoring results and the preset risk thresholds, the risk level division formula for the customs goods to be inspected is: ; Among them, S total Indicates the comprehensive scoring result; and Indicates the preset risk threshold.

[0026] Furthermore, the detection module is also used to: when the customs goods to be detected are placed on a preset weighing device, obtain the initial weight data of the customs goods to be detected through the weighing device; The above-mentioned weighing equipment can be a weighing device such as a scale. When the customs goods to be inspected are unloaded on the scale, the corresponding weight data can be displayed by the instrument of the scale. In this embodiment, when the customs goods to be inspected arrive at the customs port, a small recording device with a magnetic suction cup can be adsorbed on the side of the container of the customs goods to be inspected by a preset weighing robot. The screen of the small recording device can display the electronic identity tag of the customs goods to be inspected. The electronic tag is bound to the goods information and cannot be changed manually. The small recording device cooperates with the surveillance camera in the unmanned inspection park of the port to track and locate the goods throughout the process. The mechanical devices and detection modules in the park will confirm the identity and automatically operate by identifying the electronic tag on the screen of the small recording device. The customs goods to be inspected can be placed on the weighing device through the lifting machinery of the mechanical system. The electronic identity tag displayed by the small recording device can be scanned by the preset weighing robot to confirm the identity of the customs goods to be inspected, and the initial weight data of the customs goods to be inspected can be weighed by the weighing device.

[0027] Place the preset weighing robot at the first preset position on the weighing device, and obtain the first total weight data of the customs goods to be inspected and the weighing robot through the weighing device; wherein the weighing robot is configured with a counterweight that matches the initial weight data; control the weighing robot to reach the second preset position on the weighing device, and obtain the second total weight data of the customs goods to be inspected and the weighing robot through the weighing device; control the weighing robot to reach the third preset position on the weighing device, and obtain the third total weight data of the customs goods to be inspected and the weighing robot through the weighing device; control the weighing robot to reach the fourth preset position on the weighing device, and obtain the fourth total weight data of the customs goods to be inspected and the weighing robot through the weighing device; wherein the first preset position, the second preset position, the third preset position and the fourth preset position are the positions of the four corners on the weighing device; control the weighing robot to return to the first preset position on the weighing device, and obtain the fifth total weight data of the customs goods to be inspected and the weighing robot through the weighing device; The weighing robot is mainly composed of a camera scanning system, a recording device placement system, a counterweight system, and a motion system. The camera scanning system is used to capture the initial image of the container of the customs goods to be inspected, scan the lead seal of the customs goods to be inspected to identify the identity, and is responsible for positioning through the environmental image. The recording device placement system is located below the camera scanning system and is used to store a small recording device. When the container lead seal is confirmed to be intact and the number is correct, the weighing robot can adsorb the small recording device on the side of the container; when the customs goods to be inspected are inspected and leave the park, the small recording device is recovered by the exit. After accumulating a certain number, the mechanical system transmits it to the weighing robot in the weight detection area, so that the weighing robot can be adsorbed on other containers of customs goods to be inspected. The surveillance camera at the entrance of the port unmanned inspection park and the camera of the weighing robot jointly capture the first image of the container of the customs goods to be inspected, and identify the appearance, lead seal, and container number. The camera at each inspection checkpoint cooperates with the electronic tag on the small recording device for identity authentication to identify the container attribute characteristics of the customs goods to be inspected, such as the shape, appearance, color, etc. of the container. Cameras at various locations within the checkpoint record video of each area. At the same time, robot cameras within the park capture and supplement images from a lower angle to create image data of goods without blind spots.

[0028] The counterweight system is located below the recording device placement system. It is a heavy object with a fixed weight. The counterweight is related to the size of the container of the customs goods to be inspected. For example, the counterweight range of a 20-foot standard container is 0.02-0.06 tons, the counterweight range of a 40-foot standard container is 0.02-0.07 tons, and the counterweight range of special containers (such as refrigerated containers, open-top containers and frame containers) is 0.1%-0.2% of the total cargo weight. The motion system uses crawler wheels, which disperse pressure by increasing the contact area with the ground and have greater load capacity and stability. This method selects weighing robots with different counterweights according to the different weight specifications of the customs goods to be inspected, and uses the counterweight to calibrate the weight of the customs goods to be inspected, which can further improve the accuracy of the calibration, thereby improving the accuracy of weighing the customs goods to be inspected.

[0029] Taking into account the existing weighing system, due to the uneven weight distribution of the cargo body of the customs goods to be inspected in the container, the gravity effect on the elastic body of the weighing sensor in the weighing equipment is uneven, which will directly lead to the weighing equipment being unable to achieve the expected accuracy during the weighing process. To solve this problem, this embodiment can control the weighing robot to enter the weighing equipment after weighing the above-mentioned initial weight data; the above-mentioned first preset position, second preset position, third preset position and fourth preset position can be located at the four corners of the weighing equipment. Specifically, after the weighing robot enters the weighing equipment, it is first controlled to be located at the initial fixed point position at the lower left edge. After that, the weighing equipment weighs the first total weight data of the customs goods to be inspected and the weighing robot, and then controls the weighing robot to move counterclockwise to the fixed position at the lower right edge of the weighing equipment, and then the weighing equipment weighs the second total weight data of the customs goods to be inspected and the weighing robot, and continues to control the weighing robot to move counterclockwise to successively weigh the third total weight data when the weighing robot is located at the fixed position at the upper right edge of the weighing equipment, and the fourth total weight data when the weighing robot is located at the fixed position at the upper left edge of the weighing equipment, and finally controls the weighing robot to return to the initial fixed position, and again weighs the fifth total weight data of the customs goods to be inspected and the weighing robot.

[0030] The initial weight data, the first total weight data, the second total weight data, the third total weight data, the fourth total weight data, and the fifth total weight data are input into a pre-trained weight detection model to output the weighing data of the customs goods to be detected through the weight detection model; the weight detection model calculates the accurate weighing data of the customs goods to be detected according to the weight fluctuation. The weight detection model can be trained based on the preset historical data as a training set, and the weighing data obtained by the above calculation and the initial weight data, the first total weight data, the second total weight data, the third total weight data, the fourth total weight data, and the fifth total weight data can be supplemented into the training set to continuously enrich the data in the training set. Compared with the direct measurement by weighing equipment in the traditional technology, this method can correct the error caused by the uneven placement of the goods in the customs goods to be detected.

[0031] In this embodiment, a weighing robot with known weights is applied at four fixed positions outside the container of the customs goods to be inspected, and the weight fluctuations displayed on the weighing device are observed and recorded. The trained and optimized weight detection model can calculate the actual weighing data of the customs goods to be inspected according to the initial weight data, the first total weight data, the second total weight data, the third total weight data, the fourth total weight data, and the fifth total weight data, thereby effectively improving the accuracy of weighing the customs goods to be inspected.

[0032] This method uses the weight of the weighing robot as the correction amount, collects different data at different locations of the scale, and uses it to correct the weighing results of the scale to offset the weighing error caused by uneven weight distribution. Accurate data can be obtained without direct contact or opening the container. In addition, with the accumulation of more sample data and the continuous improvement of the algorithm itself, the weight detection model can further learn and evolve to improve accuracy.

[0033] Obtain the declaration data of the customs goods to be inspected; wherein the declaration data includes: the declared quantity, the declared type and the declared weight; send the declaration data and the weighing data to a pre-trained recognition model to confirm through the recognition model whether the weight difference between the weighing data and the declared weight satisfies a preset fluctuation range; if the weight difference satisfies the preset fluctuation range, determine the weighing data as the target weight data corresponding to the customs goods to be inspected; if the weight difference does not satisfy the preset fluctuation range, transfer the customs goods to be inspected to a preset first buffer zone, so that customs inspection personnel can inspect the customs goods to be inspected.

[0034] The above preset fluctuation range can be set according to actual needs and is not limited here; considering that in actual applications, there is often data fluctuation between the weighing data of the customs goods to be tested output by the weight detection model and the declared weight, for example, due to different batches and different packaging materials, the weight of the customs goods to be tested will change, or the product is counted by piece, and due to the process differences between each piece, the weight of the customs goods to be tested will also fluctuate, but the weight difference between the weighing data and the declared weight should meet the preset fluctuation range. After confirming that the weight difference meets the preset fluctuation range, the weighing data can be determined as the final target weight data corresponding to the customs goods to be tested, and the weighing data of the customs goods to be tested is determined to match the declared quantity and declared type. If the weight difference does not meet the preset fluctuation range, it is judged that the customs goods to be tested have the risk of false reporting, and the mechanical system can transfer the customs goods to be tested to the first buffer zone to wait for the customs inspection personnel to conduct manual re-inspection of the customs goods to be tested. This can avoid the problem of other goods being smuggled into the customs goods to be tested.

[0035] Weight inspection is the first mandatory inspection item for customs goods to be inspected entering the park. If the weight inspection of the customs goods to be inspected is qualified, the risk level of the customs goods to be inspected confirmed by the information tracking module can be maintained, and the customs goods to be inspected can be inspected or directly released according to the target inspection path of the corresponding risk level; if the weight inspection is abnormal, the risk level of the customs goods to be inspected can be raised to level one risk, and they can be transferred to the first buffer zone for manual re-inspection, thus completing the whole process of weighing the customs goods to be inspected.

[0036] Further, the detection module is also used for: if the risk level of the customs goods to be detected is a level one risk, determining that the target detection path corresponding to the customs goods to be detected is a separate detection path; according to the separate detection path, the customs goods to be detected are transferred to a separate detection room, and the container of the customs goods to be detected is opened by a hook device, so as to detect the goods body of the customs goods to be detected by an online detection device in the separate detection room, and obtain a first cargo detection result; wherein the first cargo detection result is used to indicate: whether there are illegal items in the cargo body; In actual implementation, when it is determined that the risk level of the customs goods to be inspected is a first-level risk, the corresponding target detection path can be determined as a separate detection path. In this case, the customs goods to be inspected can be transported to a separate detection area, which includes multiple separate detection rooms. A separate detection room is a detection unit. The customs goods to be inspected can be transferred to the door of one of the separate detection rooms through a mechanical system. After the door camera recognizes the electronic identity tag displayed by the small recording device adsorbed on the container of the customs goods to be inspected, the customs goods to be inspected are transferred to the separate detection room through a ground transmission device, the room door is closed, and the container is opened through a hook device on the top of the separate detection room. The cargo body inside the container is scanned and detected by the online detection device in the room. The detection time is usually 30-35 minutes. After the detection is completed, when leaving the separate detection room, the electronic identity tag is checked again to generate the corresponding first cargo detection result. According to the first cargo detection result, it can be confirmed whether there are illegal items in the cargo body. The illegal items are usually specific items that are carried without permission or authorization in violation of relevant laws and regulations.

[0037] In the process of detecting the cargo body of the customs cargo to be inspected by the online inspection device, the environment in the separate inspection room is inspected by the inspection robot to obtain a first environment inspection result; wherein the first environment inspection result is used to indicate whether there is abnormal gas and / or pathogenic microorganisms in the environment in the separate inspection room; In this embodiment, during the process of detecting the cargo body of the customs goods to be detected, the detection time window can be used to synchronously detect the environment in the separate detection room where the customs goods to be detected are located by the detection robot based on the auxiliary detection line. In the separate detection area, the detection robot enters the separate detection room one by one for detection. In practical applications, the detection robot may include a robot camera system, an electronic nose system, a rapid acquisition and detection system, and a motion system. In the robot camera system, the top camera can be used to rotate and shoot the environment and goods images, and the spatial coordinates can be established to supplement and verify the transit route of the customs goods to be detected in the customs. A fill light is set next to the bottom camera to clearly shoot the bottom image of the customs goods to be detected. Subsequently, the feature area of ​​the bottom image can be intelligently selected to check whether the customs goods to be detected are leaking or attached with abnormal substances. The electronic nose system can use a highly sensitive gas sensor array to capture and analyze the odor molecules in the environment in real time. When abnormal odors or suspected dangerous gases are found, an alarm mechanism will be triggered to achieve rapid and accurate identification of specific chemical substances. The electronic nose system can detect hazardous chemicals, dangerous gases produced by rotting plants and animals, illegal drugs and other illegal items.

[0038] The rapid collection and detection system can check for bioaerosol risks, including seasonal or regional epidemics and infectious pathogens. Specific detection items can be set by the regulatory center based on climate, regional epidemiology and relevant documents. The inlet of the rapid collection and detection system is connected to the wet cyclone collection chamber, and a rotating airflow is generated in the chamber. Sampling is performed at an airflow rate of 400-500 L / min (this airflow rate can cover the diameter range of common pathogenic or harmful particles, and reduce the damage to particles caused by physical collisions, thereby improving detection accuracy). The aerosol particles in the collection chamber collide inertially under the action of centrifugal force and surface tension, and are separated and collected into the medium liquid at the bottom of the chamber according to the mass of the particles, and are transported to the outlet at the same time. The outlet is connected to the microfluidic detection chip. Under the action of capillaries, the liquid enters the pre-embedded parallel test strips in the chip. Each test strip corresponds to a detection target substance. The camera at the end of the chip scans the test results on the test strip: when only the control line (T line) appears, the result is negative; when the detection line (C line) and T line appear, the result is positive, where negative means that the specified microorganism is not detected, and positive means that the specified microorganism is detected. In the motion system, in order to ensure that the detection robot can move flexibly and efficiently in each detection unit, four sets of high-performance universal wheels are set under the detection robot for movement. They are made of high-strength materials and have excellent load-bearing capacity and wear resistance, which improves the driving stability of the detection robot and ensures that the detection robot remains stable when running for a long time. Usually, the electronic nose system is mainly used for chemical gas detection, and the rapid collection and detection system is mainly used for pathogenic microorganism detection.

[0039] By comprehensively screening the microbial components contained in suspended particles in the air, especially those pathogens that are known to be easily attached to the surface of customs goods to be inspected and survive during transportation and are highly contagious, such as bacterial spores, viruses and other high-risk pathogenic microorganisms, it is possible to avoid the risk of epidemics caused by high-risk transmission factors through logistics transportation and achieve early warning functions. High-risk transmission factors that may be carried by containers and packaging are checked to prevent them from spreading across regions or even across borders through the logistics chain. The detection robot can also upload the test results obtained to the information tracking module to accumulate historical information on risk assessment and improve the accuracy of determining the risk level corresponding to the customs goods to be inspected.

[0040] If the first cargo inspection result indicates that there are no illegally carried items in the cargo body, and the first environmental inspection result indicates that there are no abnormal gases and pathogenic microorganisms in the environment of the separate inspection room, the customs goods to be inspected are allowed to pass customs; if the first cargo inspection result indicates that there are illegally carried items in the cargo body, and / or the first environmental inspection result indicates that there are abnormal gases and / or pathogenic microorganisms in the environment of the separate inspection room, the customs goods to be inspected will be transferred to a preset first buffer zone so that customs inspection personnel can conduct inspection and quarantine work on the customs goods to be inspected.

[0041] In actual implementation, when both the first cargo inspection result and the first environment inspection result indicate that there is no abnormality in the customs cargo to be inspected, it can be considered that the inspection of the customs cargo to be inspected has passed, and the customs cargo to be inspected can be allowed to pass customs; when the first cargo inspection result and / or the first environment inspection result indicate that there is an abnormality in the customs cargo to be inspected, an instruction can be issued through the supervision center to transfer the customs cargo to be inspected to the first buffer zone by the mechanical system, waiting for the customs inspection personnel to conduct manual re-inspection of the customs cargo to be inspected.

[0042] Furthermore, the detection module is also used for: if the risk level of the customs goods to be detected is a level 2 risk, determining that the target detection path corresponding to the customs goods to be detected is a joint detection path; according to the joint detection path, transferring the customs goods to be detected to one of the detection compartments in the joint detection room; wherein the joint detection room includes four detection compartments; when the customs goods to be detected are placed in the four detection compartments, opening the container of each customs goods to be detected by a hook device, so as to perform a mixed detection on the cargo bodies of the four customs goods to be detected by an online detection device in the joint detection room, and obtaining a second cargo detection result; wherein the second cargo detection result is used to indicate: whether there are any illegally carried items in the cargo body of at least one of the customs goods to be detected; In actual implementation, when it is determined that the risk level of the customs goods to be inspected is a secondary risk, the corresponding target inspection path can be determined as a joint inspection path. In this case, the customs goods to be inspected can be transported to the joint inspection area, which includes multiple joint inspection rooms, each of which includes four inspection compartments. A joint inspection room is a joint inspection unit, and four customs goods to be inspected can be transported to the door of the joint inspection room through a mechanical system. After the door camera recognizes the electronic identity tag displayed by the small recording device adsorbed on the container of each customs goods to be inspected, each customs goods to be inspected will be transferred to the joint inspection room separately through the ground conveyor. Each inspection compartment in the room, close the room door of the joint inspection room, open each container through the hook device on the top of the joint inspection room, and scan and inspect the cargo bodies inside the four containers through the online inspection device in the room. The inspection time is usually 30-35 minutes. After the inspection is completed, when leaving the joint inspection room, check the electronic identity tag again to generate a corresponding second cargo inspection result. According to the second cargo inspection result, it can be confirmed whether there are illegally carried items in the cargo body of at least one of the customs cargoes to be inspected. In this embodiment, four inspection compartments are set in each joint inspection room, that is, four customs cargoes to be inspected can be mixed and inspected at the same time.

[0043] In the process of performing mixed detection on the cargo bodies of the four customs cargoes to be detected by the online detection device, the detection robot performs mixed detection on the environment in the four detection compartments to obtain a second environment detection result; wherein the second environment detection result is used to indicate whether there is abnormal gas and / or pathogenic microorganisms in the environment of at least one detection compartment; In this embodiment, during the inspection of the cargo bodies of four customs goods to be inspected in the same joint inspection room, the inspection time window can be utilized to synchronize the inspection of the environment in the joint inspection room where the four customs goods to be inspected are located based on the auxiliary inspection route. In the joint inspection area, the inspection robot uses four inspection compartments as a group of inspection units, and enters each inspection compartment in turn for mixed collection and inspection. In this embodiment, four customs goods to be inspected can be inspected. The number comprehensively considers the inspection time of the inspection robot and the inspection effect of each customs goods to be inspected. If the number is more than four, it will usually affect the inspection effect of the customs goods to be inspected. If the number is less than four, it will affect the inspection efficiency of the customs goods to be inspected.

[0044] If the second cargo inspection result indicates that there are no illegally carried items in the cargo bodies of the four customs cargoes to be inspected, and the second environmental inspection result indicates that there are no abnormal gases and pathogenic microorganisms in the environments of the four inspection compartments, the four customs cargoes to be inspected are allowed to pass customs; if the second cargo inspection result indicates that there are illegally carried items in the cargo body of at least one customs cargo to be inspected, and / or the second environmental inspection result indicates that there are abnormal gases and / or pathogenic microorganisms in the environment of at least one inspection compartment, the risk level of each customs cargo to be inspected in the four inspection compartments is upgraded to level one risk, and the cargoes are transferred to the preset second buffer zone, so that the four customs cargoes to be inspected that have been upgraded to level one risk will be re-inspected separately according to separate inspection paths.

[0045] In actual implementation, when both the second cargo inspection result and the second environment inspection result are normal, it can be considered that the inspection of the four customs goods to be inspected has passed, and the four customs goods to be inspected can be allowed to pass customs, and subsequent domestic logistics transportation can be carried out; when the second cargo inspection result is abnormal, it can be considered that there are illegally carried items in the cargo body of at least one customs cargo to be inspected, and when the second environment inspection result is abnormal, it can be considered that there are abnormal gases and / or pathogenic microorganisms in the environment of at least one inspection compartment. In the case of abnormalities in the second cargo inspection result, the risk level of each customs cargo to be inspected in the four inspection compartments can be upgraded to level one risk, and an instruction can be issued through the supervision center to transfer the customs goods to be inspected to the second buffer zone by the mechanical system, so that the four customs goods to be inspected that have been upgraded to level one risk can be re-inspected separately according to a separate inspection path. Through a separate inspection process, the customs goods to be inspected with abnormalities can be locked, and the remaining customs goods to be inspected without abnormalities can be allowed to pass customs normally.

[0046] In order to effectively realize on-site, rapid and accurate identification of products, the above-mentioned online detection device combines an X-ray transmission instrument and a spectrum analyzer to perform non-destructive rapid detection on the customs goods to be inspected. First, based on X-ray three-dimensional imaging technology, identify the type and quantity of customs goods to be inspected, detect whether there are illegal items or abnormal conditions (for example, the shape and appearance are different from common items, hidden disguised items, etc.); then, use the artificial intelligence system to process the image obtained based on X-ray three-dimensional imaging technology, and mobilize the robotic arm to take samples for spectral analysis according to the internal structure and size of the goods; based on Raman spectroscopy and infrared spectroscopy technology, combine deep learning to establish an identification model to extract feature data, and realize the automated on-site identification and rapid detection of high-value goods and dangerous chemicals, including rapid identification of chemical fingerprint information of raw materials, identification of reagents or solids and detection of doping, rapid identification of toxic and harmful substances in animals, plants and their products, and trace detection of surface residues on packaging, etc. For example, the sample can be placed in the instrument measurement area so that the laser focus of the spectrometer is aimed at the sample surface, the measurement is started in the operation interface, the laser irradiates the sample, the collected data is displayed in the form of a spectrum, and the molecular vibration information of the sample is recorded. The spectral data can be analyzed, interpreted and compared using software. The specific detection method can refer to the relevant technology, which will not be repeated here.

[0047] The detection system in related technologies mainly focuses on the detection of the goods themselves, while ignoring the pathogenic substances that may be carried by the cargo boxes and packaging materials. This solution uses a detection robot, which can use the time window of the online detection device to detect the cargo body of the customs goods to be inspected, and simultaneously perform the task of collecting and quickly analyzing the aerosol samples in the space (i.e. the detection room), thereby improving the comprehensiveness of the detection.

[0048] It can be seen from the above embodiments that the detection module can perform quantity, quality, composition and safety performance tests on the customs goods to be inspected, and the detection process involves a weight detection area, a separate detection area, a joint inspection area, a detection robot, a first buffer zone and a second buffer zone. In this embodiment, three detection routes are divided based on the three levels of risk levels. The first route is used to inspect the customs goods to be inspected with a first-level risk. These customs goods to be inspected are customs goods that need to be inspected. They are transferred from a mechanical platform to a separate inspection area, and are inspected one-to-one by an online detection device and a detection robot equipped with an artificial intelligence recognition system and a rapid detection device. The second route is used to inspect the customs goods to be inspected with a second-level risk. They are transferred from a mechanical platform to a joint inspection area, and are jointly inspected by an online detection device and a detection robot. The third route is an auxiliary detection route, in which the detection robot cyclically detects key risk control factors of environmental aerosols on the paths of the first two routes, including chemical leaks, flammable and explosive monitoring, and pathogenic microorganisms.

[0049] Furthermore, the system also includes: a risk control module; the risk control module is used to: obtain the number of customs goods to be inspected in the joint inspection room; if within a first preset time period, the number of customs goods to be inspected in the joint inspection room is less than four, and according to a pre-trained logistics prediction model, it is determined that within a first future time period, there are no customs goods to be inspected with a risk level of level 2 arriving at the customs, then according to a preset first withdrawal rule, the first designated customs goods are withdrawn from the customs goods to be inspected with a risk level of level 3; the risk level of the first designated customs goods is increased to level 2 risk, and they are transported to the joint inspection room, so that the customs goods to be inspected in the joint inspection room are inspected according to the joint inspection path.

[0050] The above-mentioned first preset time period and the first future time period can be set according to actual needs. For example, the first preset time period can be 20 minutes, etc., and the first future time period can be 10 minutes, etc.; the above-mentioned first withdrawal rule can be to withdraw according to a preset withdrawal ratio, or to withdraw according to a preset withdrawal quantity, etc., which can be selected according to actual needs and is not limited here; the number of the above-mentioned first designated customs goods can be one or more, which can be selected according to actual needs; in actual implementation, when the second-level risk goods in the joint inspection room do not meet the quantity requirements of the mixed inspection of 4 customs goods to be inspected within 20 minutes, and according to the pre-trained logistics prediction model, it is determined that there are no second-level risk customs goods to be inspected arriving at the customs within 10 minutes, the risk module can withdraw part of the customs goods to be inspected from the third-level risk customs goods to be inspected, increase the risk level of these withdrawn customs goods to be inspected to second-level risk, and transport them to the joint inspection room for inspection, thereby reducing the waiting time in the joint inspection room.

[0051] Furthermore, multiple joint inspection rooms form a joint inspection area; multiple separate inspection rooms form a separate inspection area; the above-mentioned first buffer zone can be set between the separate inspection area and the manual inspection area outside the park, and is used to store customs goods to be inspected with a first-level risk that have not passed the inspection, as well as customs goods to be inspected that have failed the weight inspection; the supervision center plans the transit time and transit plan based on the algorithm according to the amount of goods to be inspected manually and the storage capacity of goods in the buffer zone, and the mechanical system transports them out of the area for manual re-inspection, wherein the planning method can refer to the relevant technology and will not be repeated here. The above-mentioned second buffer zone can be set between the joint inspection area and the separate inspection area, and is used to store customs goods to be inspected with a second-level risk that have not passed the joint inspection. The supervision center plans the optimal inspection time and the room for inspection based on the algorithm according to the real-time inspection situation in the park and the amount of goods to be inspected, and the mechanical system executes instructions to transport them to the designated separate inspection room in the separate inspection area for re-inspection.

[0052] The risk control module is used to: calculate the first filling degree of the joint inspection area according to the number of customs goods to be inspected in each joint inspection room; calculate the second filling degree of the separate inspection area according to the number of customs goods to be inspected in each separate inspection room; if the first filling degree reaches the preset filling degree threshold, and the second filling degree is lower than the preset filling degree threshold, and it is determined according to the logistics prediction model that there are customs goods to be inspected with a risk level of level 2 arriving at the customs in the second future time period, according to the preset second withdrawal rule, the second designated customs goods are withdrawn from the customs goods to be inspected with a risk level of level 2; the risk level of the second designated customs goods is increased to level 1 risk, and the goods are transported to the separate inspection room so as to be inspected according to the separate inspection path; The first filling degree can be used to indicate the occupancy rate of multiple joint inspection rooms in the joint inspection; the second filling degree can be used to indicate the occupancy rate of multiple separate inspection rooms in the separate inspection area; the filling degree threshold can be set according to actual needs, for example, it can be 85%; the second future time period can be set according to actual needs, for example, it can be 30 minutes; the second withdrawal rule can be to withdraw according to a preset withdrawal ratio, or to withdraw according to a preset withdrawal quantity, etc., which can be selected according to actual needs and is not limited here; the number of the second designated customs goods can be one or more, which can be selected according to actual needs; in actual implementation, when the joint inspection room in the joint inspection area has a filling degree of 85%, the separate inspection room in the separate inspection area has not reached a filling degree of 85%, and according to the logistics prediction model, it is determined that there will be customs goods to be inspected with a second-level risk arriving at the customs within 30 minutes, the risk module can withdraw part of the customs goods to be inspected from the customs goods to be inspected with a second-level risk, raise the risk level of these withdrawn customs goods to be inspected to a first-level risk, and transfer them to the separate inspection room for inspection, so as to reduce the inspection pressure of the joint inspection area.

[0053] If both the first filling degree and the second filling degree reach the preset filling degree thresholds, and it is determined according to the logistics prediction model that within the third future time period, there will be customs goods to be inspected with a risk level of target risk arriving at the customs, a deceleration instruction is generated so that the mechanical transport device reduces the transportation rate of the customs goods to be inspected that have arrived at the customs and have not been inspected according to the deceleration instruction; wherein the target risk is: primary risk and / or secondary risk. The above-mentioned third future time period can be set according to actual needs, for example, it can be 30 minutes, etc.; in actual implementation, when the joint inspection area and the separate inspection area are both 85% full, and according to the logistics prediction model, it is determined that there will be first-level risk and / or second-level risk customs goods to be inspected arriving at the customs within 30 minutes, the risk module can generate a deceleration instruction to enable the mechanical transport device to reduce the transportation rate of the customs goods to be inspected that have arrived at the customs and have not been inspected according to the deceleration instruction, so as to disperse the customs goods to be inspected in the transportation path, avoid congestion in the inspection area, and minimize potential risks while ensuring customs clearance efficiency, thereby improving the overall operation level of the park.

[0054] As the core of the unmanned inspection park, the above-mentioned supervision center undertakes decision-making support and global supervision and control tasks, deeply integrates artificial intelligence and big data analysis capabilities, supports seamless docking and collaboration with other systems, and ensures efficient and accurate management and scheduling of various functional areas in the entire park. The supervision center can include a main control module, a risk control module, and a data processing module. These modules are described below: The main control module can automatically summarize and analyze detailed data from all links in the park, including but not limited to key indicators such as cargo flow, customs clearance time, and abnormal event records. At the same time, it also summarizes and organizes risk information input from the outside, such as global trade dynamics, epidemic development, changes in the international security situation, etc., and forms risk assessment and checkpoint detection reports every day through multi-dimensional data fusion and intelligent analysis. Once an emergency or abnormal situation occurs, the emergency plan can be quickly activated, and the resources of all parties can be coordinated for rapid disposal to ensure the safe and stable operation of the park. The specific analysis method is as follows: obtain cargo flow data, customs clearance data, abnormal detection data, etc. in the park from the data module; risk information outside the park, including changes in global trade policies, international security situation database, epidemic data, and epidemic prevention and control policy adjustment information. Obtain the sampling plan and data in the park on the same day from the risk control module. Preprocess the acquired data, such as deleting duplicate data and estimating incomplete data. The processed data is deeply analyzed through preset data models and association rules. For example, the cargo flow data of the park is combined with trade data to analyze the impact of changes in trade policies on the import and export volume of specific goods; abnormal data detection is associated with changes in the epidemic situation to identify potential biochemical risk factors; customs clearance time data is analyzed with cargo type data to compare the customs clearance efficiency of the park's cargo, and the calculation method of the risk assessment system, various calculation models used in the park, and AI algorithm parameters are optimized in combination with the customs business characteristics and management needs of the park; the customs clearance data of the day is analyzed with the customs clearance data of the past to predict future changes in cargo flow or seasonal epidemic trends. The data analysis results are automatically generated and pushed to management personnel through pre-stored standardized report templates, and retained in the data module for historical tracing. The report cannot be modified after it is generated to avoid human tampering.

[0055] The risk control module can generate and dynamically adjust the sampling plan based on the customs' normal inspection capacity, historical sampling data and real-time customs clearance status, and check the risk level of goods of various risk levels on the day to ensure the targetedness and effectiveness of the sampling work.

[0056] The data processing module is responsible for comprehensively collecting and archiving all data related to cargo declaration and inspection in the park, and establishing an artificial intelligence model based on big data to conduct in-depth mining and intelligent analysis of real-time data, thereby providing data support for the supervision center. It mainly includes data collection and storage functions, data analysis functions, and data transmission functions. The following describes these three functions respectively.

[0057] Data collection and storage function: collect commodity information and declarant information of the declared customs goods to be inspected to establish a corresponding database and index, which is convenient for data search and extraction of the information verification module. The information verification module generates an exclusive electronic identity tag for the customs goods to be inspected based on the Electronic Product Code Information Service (EPCIS), which is used for the identification and positioning of the customs goods to be inspected in the park. The artificial intelligence platform determines whether the goods inspection can be passed directly or whether a higher risk level inspection is required based on the various inspection data in the inspection area, and finally records the relevant data in the system for data backtracking.

[0058] Data analysis function: Establish artificial intelligence models based on big data and conduct comprehensive analysis of real-time data. Customized construction is carried out according to the needs of each functional area, and in-depth analysis and efficiency verification are carried out using parameters such as real-time detection data, historical data, and risk factors. The analysis and verification methods can refer to relevant technologies and will not be repeated here. Continuously evaluate and optimize artificial intelligence models to make them closer to actual business needs. Through in-depth mining and correlation analysis of data, abnormal conditions of detected items or potential risks in the park are discovered, and customs clearance conditions and risk conditions are provided to the supervision center.

[0059] Data transmission function: In order to achieve the real-time and accuracy of data, data cleaning and pre-processing technology is used to remove duplication, correct errors, normalize and other processing of the original data to improve the quality and application value of the data. At the same time, in conjunction with the transmission device of the information system, it ensures that every link of the customs goods to be inspected in the customs clearance process can be recorded and updated in time to form a complete data chain.

[0060] Furthermore, the detection module is also used to: if the risk level of the customs goods to be detected is level three risk and the customs goods to be detected have not been selected, allow the customs goods to be detected to pass customs. In this embodiment, all customs goods to be detected can be divided into three risk levels based on the preset risk threshold. Level one high-risk customs goods need to be tested separately, level two medium-risk customs goods need to be tested in combination, and level three low-risk customs goods can be released directly without additional procedures, which simplifies the customs clearance process for low-risk customs goods and improves customs clearance efficiency.

[0061] Further, the system also includes: an information recording module; the information recording module is used to receive the target detection result, convert the target detection result according to the preset format, and obtain the converted detection result; save the converted detection result to the blockchain. The above preset format can be set according to actual needs; in actual implementation, in order to ensure the efficiency and accuracy of information access and sharing in the park, and to avoid data tampering, attack, leakage and other security risks, the target detection result obtained by the detection module can be uploaded to the blockchain for storage after format conversion, so as to ensure that the target detection result is fully recorded and cannot be modified artificially, and to ensure the integrity and authenticity of all data. In addition, the blockchain technology can also ensure the security of data transmission inside and outside the park, including the inspection platform in the information inspection module accessing other data platforms for information authenticity comparison, and it is necessary to avoid the risk of data leakage or tampering in the process; the risk level of the customs goods to be inspected generated by the information tracking module is safely stored and accurately transmitted to the inspection module to correctly execute the inspection steps; in addition, in view of the situation that information from different data sources is continuously uploaded to the information recording module during the declaration and inspection process of the customs goods to be inspected, a compatible heterogeneous regional chain information service system can be established to collaboratively manage information. Specifically, the entire process from application to the end of testing is a process of continuous accumulation of information. Initially, a goods database of the customs goods to be tested is established based on the information in the declaration list. In each link of subsequent information verification, risk assessment, and test results, new results are uploaded to the goods database, and finally a complete database of the entire process of the customs goods to be tested is formed. Although the regional chain can realize data security sharing, when a large amount of data in different links is continuously accumulated and uploaded, there may be an explosion of data on the chain. This solution coordinates and manages data by building a compatible service system, and converts heterogeneous data from different databases, different detection nodes, different formats of data, and different service systems into unified standards to avoid data explosion.

[0062] This plan targets the main application scenarios of customs clearance of goods for export in the mainland, and aims to respond to the full-chain supervision requirements of "source control + port supervision + subsequent verification + combating smuggling", and build an integrated smart customs comprehensive management system on the existing basis. By using advanced information technologies such as Electronic Product Code Information Service (EPCIS), virtual reality (AR), big data, artificial intelligence (AI), 5G communication, and combining blockchain, Internet of Things and other means, we will focus on exploring aspects such as monitoring path optimization, regional module design, technology integration, and information security integration, in order to optimize the flow between goods elements and digital elements, reduce manual intervention, and further improve the efficiency and quality of customs supervision. At the same time, we will enhance risk warning and tracking capabilities, achieve full traceability of the logistics chain, credit chain, and supervision chain, and help achieve a more intelligent customs clearance process. This not only provides a new reference for the construction of smart customs at the current stage, but also sets a new benchmark for promoting a new round of opening up and injects strong momentum.

[0063] In the current customs assessment and detection system, factors such as the risk of transmission of historical and seasonal infectious pathogens and the risk of carrying dangerous factors in the path have not been fully considered. These pathogenic substances or biochemical factors usually have a long incubation period and strong survival ability. In the context of globalization, convenient transportation conditions may aggravate the threat they pose to domestic public security, and once an epidemic spreads, it becomes particularly difficult to trace the source. This solution aims to solve the above problems by combining the historical and seasonal pathogenic microbial data records in the origin and logistics transportation process, and introducing the information of dangerous factors carried by cargo ships in the past to conduct a more comprehensive and in-depth risk assessment. The risk assessment system of this solution comprehensively considers four main dimensions: customs clearance frequency, trajectory behavior, historical information and designated characteristics, in order to achieve accurate judgment of the potential risks of imported goods. The risk assessment and hierarchical management mechanism proposed by the system helps to optimize the efficiency of resource allocation. At the same time, when an infectious disease transmission event occurs, the data based on the statistics of this mechanism can more efficiently track the potential initial infection source, thereby providing strong support for timely prevention and control measures. In addition, the system can also help relevant departments better understand the changing trend of disease transmission patterns between different regions, laying the foundation for formulating more scientific and reasonable public health strategies.

[0064] In the above system, the information verification module can build an inspection and identification platform to automatically identify the validity of the information of the customs goods to be inspected, and pre-judge the authenticity and credibility of the declaration data of the goods entering the customs; secondly, the information tracking module can evaluate the risk level of the customs goods to be inspected based on risk assessment factors such as customs clearance frequency, trajectory behavior, historical information and specified characteristics, so as to plan the corresponding customs clearance inspection path according to the risk level; finally, the information recording module can collect and archive data throughout the process based on the target detection results transmitted back by the detection module in real time, and support data access and traceability after customs clearance.

[0065] In addition, the system also includes an emergency processing module, which can monitor the environment in the park in real time and is responsible for safety management. Through real-time data from camera and detection technology, it can quickly analyze, respond to and effectively handle various emergency events in the port unmanned detection park based on intelligent means, and quickly report emergencies to the main control module of the supervision center and accurately implement the treatment plan. The emergency processing module usually includes a fire protection system and a public security system.

[0066] Firefighting system: Based on the detection results of high-definition cameras deployed in the park and detection robots equipped with electronic noses, real-time monitoring and data analysis of potential risk factors such as liquid leakage and biochemical abnormality levels are carried out, and early warning signals are immediately issued when abnormal conditions are detected. Once a major accident such as fire, explosion, attack, etc. occurs, the firefighting system will quickly activate the emergency plan and automatically dispatch the mechanical system to take necessary protective measures, such as shutting down the ventilation system to prevent the spread of harmful gases, starting the fire extinguishing device for preliminary firefighting, and the monitoring center will start the data protection program. At the same time, the system will immediately send detailed alarm information to the local fire department, including the specific location, nature and possible impact range of the accident, so that firefighters can arrive at the scene as soon as possible and carry out rescue work. For details, please refer to the relevant technology, which will not be repeated here.

[0067] Public Security System: Based on the photos of people captured by the camera at the entrance of the park, the personnel on the transport ship are monitored and recorded in real time. When the camera captures the photo of a suspicious person, the system will automatically compare and analyze it with the image in the database of cross-border illegal persons. After confirming the identity of the illegal person, the public security system will immediately send an alarm to the nearest public security agency and provide detailed information of the person (such as name, identity characteristics, etc.) to assist the police in taking prompt action. At the same time, it also assists the fire protection system in providing support for retrospective investigation of unexpected dangerous situations. For details, please refer to the relevant technology, which will not be repeated here.

[0068] Mechanical operation module: responsible for the loading, unloading and transportation of goods in the customs port and the park. It integrates intelligent technology to form a highly automated and intelligent logistics processing system, realizes accurate navigation and positioning of goods in the park, intelligent path planning, emergency safety braking measures, and mechanical operations such as lifting, conveying, loading and unloading of goods. Specifically, it includes intelligent control systems and mechanical devices.

[0069] Intelligent control system: The uneven inspection time for goods of different risk levels and the dynamic changes in the volume of goods of various risk levels every day, coupled with the uncertainty of the cargo inspection results, make it difficult to fully predict the subsequent transportation routes of the goods. The intelligent control system builds a dynamic model of the goods in the park based on three-dimensional modeling, and uses AI real-time data analysis technology to scientifically plan the movement path of the container and issue instructions to the mechanical device. When it is monitored that the amount of goods at a certain risk level exceeds the preset capacity, the system sends real-time feedback to the supervision center and promptly adjusts the handling strategy for risk-level goods to ensure smooth and efficient logistics processes. For example, when the number of second-level risk goods waiting in line for inspection reaches 70% of the buffer zone capacity, it is determined that the inspection load is large, and the random inspection of third-level risk goods is terminated. When the number of first-level risk goods waiting in line for inspection reaches 50% of the buffer zone capacity, it is determined that the inspection load is large, and the frequency of random inspection of second-level risk goods is reduced to one-third of the original frequency. This value is determined according to the nature and inspection time of different risk goods. Third-level risk goods have lower risks, so there is no big problem without random inspections, and the alarm threshold can be set at 70%. However, second-level risk goods must continue to be inspected, and the inspection time for first-level risk goods is longer, so the plan can be changed at a lower threshold.

[0070] Mechanical devices: Integrate intelligent control systems and high-precision sensors, and can autonomously complete efficient loading operations of container cargo under the control of the intelligent control system, execute mobile paths, and accelerate the flow of cargo within the port. Including lifting machinery and handling machinery, mainly used for rapid loading and unloading of containers and short-distance transshipment. Conveying machinery is used for continuous transshipment operations.

[0071] For easier understanding, see Figure 2 A schematic diagram of the layout of an unmanned inspection park is shown in FIG. Figure 3 A schematic diagram of the action route of a weighing robot is shown in FIG. Figure 4 The overall schematic diagram of a separate detection area shown in FIG. Figure 5 A schematic diagram of a single detection room is shown in FIG. Figure 6 A schematic diagram of a weighing robot is shown in FIG. Figure 7The schematic diagram of a detection robot shown in the figure; the information verification module verifies the authenticity of the declaration information and other information of the declarant, and determines that the customs goods to be detected are first-level risk goods based on the evaluation factor information tracking module, or are selected as first-level risk goods by the risk control module. When the batch of customs goods to be detected arrives at the unmanned detection park 100 of the port, it is unloaded by the lifting machinery and transferred to the scale 129, and the park camera and the camera 121 of the weighing robot scan and confirm the identity of the container. The corresponding counterweight system 123 is configured according to the weight of the container. After the weighing robot obtains the initial weight data at the initial fixed point position 125 entering the weight detection area 120 under the action of the motion system 124 of the weighing robot, it goes to the next fixed point position 2 126, fixed point position 3 127, fixed point position 4 128 in sequence, and finally returns to the initial fixed point position 125 to weigh the weight to complete the weighing test, and the small recording device of the electronic tag used to display the identity of the goods on the recording device placement system 122 of the weighing robot is adsorbed on the side wall of the container. When the weight data of the goods to be inspected by the customs is abnormal, the mechanical operation module will set a transportation route to transfer the goods to the first buffer zone 170, and will enter the manual inspection area outside the park for inspection. When the weight data of the goods meets the preset fluctuation range, the system confirms that the weight and quantity of the goods are consistent, and according to the transportation route set by the mechanical operation module, the transport machinery will transfer the goods to the separate inspection area 150. When the goods arrive at the door 152 of the separate inspection room, the camera recognizes the electronic tag to confirm the identity. The inspection robot determines the position by the camera 131 of the inspection robot, and the motion system 135 of the inspection robot provides power to follow into the room 151. After opening the container, the online inspection device 153 of the separate inspection room performs non-destructive inspection items in sequence and issues the inspection results (a total of 30-35 minutes). During this period, the electronic nose system 132 of the inspection robot is in working state. At the same time, the rapid collection system 133 collects environmental aerosols for 30 minutes in the room, and then the rapid detection system 134 performs rapid test on the test strips and issues the test results (a total of 35 minutes). After the inspection, the camera 131 of the inspection robot rechecks the electronic identity tag of the customs goods to be inspected, binds the two test results and uploads them to the information recording module. When the two test results are normal, the supervision center can issue instructions, and the mechanical system will transmit the customs goods to be inspected to the exit 180 and issue customs clearance documents. When at least one of the test results is abnormal, the supervision center can issue an instruction to transfer the customs goods to be tested to the first buffer zone 170 by the mechanical system for manual re-inspection. All test records will be archived in the information recording module.

[0072] like Figure 8 An overall schematic diagram of a joint inspection area is shown, and Fig. 9The schematic diagram of a joint inspection room shown in the figure, the information inspection module verifies the authenticity of the declaration list information and other information of the declarant, and the information tracking module determines that the customs goods to be inspected are secondary risk goods, or are selected as secondary risk goods by the risk control module based on the preset risk assessment factor. When the batch of customs goods to be inspected arrives at the entrance 110 of the unmanned inspection park 100 of the port, the lifting machinery completes the unloading and transfers to the weighbridge 129, and the park camera and the camera 121 of the weighing robot scan and confirm the identity of the container. The corresponding counterweight system 123 is configured according to the weight of the container. After the weighing robot obtains the initial weight data at the initial fixed point position 125 entering the weight detection area 120 under the action of the motion system 124 of the weighing robot, it goes to the next fixed point position 2 126, fixed point position 3 127, fixed point position 4 128 in sequence and finally returns to the initial fixed point position 125 to weigh the weight to complete the weighing test, and the small recording device of the electronic tag used to display the identity of the goods on the weighing robot recording device placement system 122 is adsorbed on the side wall of the container. When the cargo weight data is abnormal, the mechanical operation module will set a transportation route to transfer the cargo to the first buffer zone 170 for manual cargo inspection. When the cargo weight data meets the preset fluctuation range and the system confirms that the cargo weight is consistent with the quantity, the transport machinery will transfer the cargo to the joint inspection area 140 according to the transportation route set by the mechanical operation module. The cargo arrives at the door 142 of the joint inspection room 141, and the camera recognizes the electronic tag to confirm the identity, and enters the No. 1 inspection compartment 143. Subsequent cargo enters the No. 2 inspection compartment 144, the No. 3 inspection compartment 145, and the No. 4 inspection compartment 146 in turn. When the cargo in the four inspection compartments has entered, the inspection robot is positioned by the camera 131 of the inspection robot, and the motion system 135 of the inspection robot provides power to follow into the room 141. After opening the container, the online inspection device 147 of the joint inspection room will perform non-destructive inspection items in turn and issue inspection results (a total of 30-35 minutes). During this time, the electronic nose system 132 of the inspection robot is in working condition, and the inspection robot first enters the No. 1 inspection compartment. The test compartment 143 collects environmental aerosols for 5 minutes, then enters the second test compartment 144 and continues to collect for 5 minutes. After collecting in four test compartments in turn, the rapid collection system 133 collects environmental aerosols for 30 minutes, and then the rapid detection system 134 performs rapid detection of test strips and issues the test results (a total of 35 minutes). After the test is completed, the camera 131 of the simultaneous detection robot checks the electronic identity tags of the four customs goods to be tested again, binds the two mixed inspection results and uploads them to the information recording module. When the two mixed inspection results are normal, it can be determined that the four customs goods to be tested have passed the test. The supervision center will issue an instruction, and the mechanical system will transmit the four customs goods to be tested to the exit 180 and issue the corresponding customs clearance documents.When at least one of the test results is abnormal, it can be determined that the four customs goods to be tested have failed the test and are all upgraded to level 1 risk goods. The mechanical system transfers the goods to the second buffer 160 and re-inspects them according to the level 1 risk goods process. All test records will be archived in the information recording module.

[0073] Another embodiment is provided below to illustrate the processing of customs goods to be inspected with a risk level of level three. Specifically, the information verification module verifies the authenticity of the declaration list information and other information of the declarant. The information tracking module determines that the customs goods to be inspected are level three risk goods based on the preset risk assessment factors and have not been selected for inspection. When the batch of customs goods to be inspected reaches the entrance 110 of the unmanned inspection park 100 of the port, the unloading is completed by the lifting machinery and transferred to the weighbridge 129. The park camera and the camera 121 of the weighing robot scan and confirm the identity of the container. Then, referring to the method in the above embodiment, the customs goods to be inspected are weighed. Under the action of the motion system 124 of the weighing robot, the weighing robot adsorbs the small recording device for displaying the identity electronic tag of the goods on the recording device placement system 122 of the weighing robot to the side wall of the container, and the mechanical device transfers it to the exit 170 and issues a customs clearance document. The customs clearance record will be archived in the information recording module.

[0074] like Fig.10 A schematic diagram of a customs cargo inspection system is shown in FIG. Fig.11 A schematic diagram of another customs cargo inspection system is shown in FIG. Fig.10 The system includes: information module, supervision center, and detection module. The supervision center can check the status of goods in each area of ​​the park based on image recognition algorithms, and obtain the quantity of goods in each area and the occupancy of scene space, and assist the mechanical system to adjust and plan the subsequent transportation routes of goods and the detection lines of the detection robots. Fig.11From a general perspective, the information module includes an information verification module, an information tracking module and an information recording module; the supervision center includes: a main control module, a risk module (corresponding to the above-mentioned risk control module) and a data module; the information module can obtain the park situation, such as cargo flow, customs clearance time, abnormal events, emergencies, etc., and can also obtain declaration materials, such as product information, enterprise status, customs clearance records, ship information, shipping routes, etc.; it can also obtain risk control data, such as domestic and foreign epidemic prevention and control, global economic trade, laws and regulations, etc.; the information module can analyze the information obtained to obtain the risk level corresponding to the customs goods to be inspected, and can send the risk level to the main control module in the supervision center, and determine the target detection path corresponding to the customs goods to be inspected through the main control module. The mechanical operation module can transport the customs goods to be inspected to the corresponding target detection path, and the detection module can detect the customs module to be inspected according to the target detection path, and the target detection results obtained can be fed back to the information recording module for recording and storage.

[0075] Furthermore, Fig.11 In the information verification module, the information verification module can access the external database and the declaration information of the customs goods to be tested (declaration list information), generate the corresponding cargo file (corresponding to the above-mentioned electronic file) according to the declaration information, and determine the cargo risk corresponding to the customs goods to be tested (corresponding to the above-mentioned risk level) through the information tracking module. The information recording module can record the cargo file and cargo risk; the information tracking module can send the cargo arrival time to the risk module (corresponding to the above-mentioned risk control module), and the risk control module can determine the sampling plan according to the amount of cargo tested in each area; it can also send the dynamic information of risk sampling to the main control module; the data module obtains information such as new policies and new epidemics, as well as the cargo files and inspection information obtained from the information recording module. According to the information, the park inspection situation and the risk control situation outside the park can be determined and sent to the main control module. The main control module can generate a daily park report based on all the information received. In addition, when the inspection module inspects the customs goods to be inspected, it usually first inspects the weight of the customs goods to be inspected in the weight inspection area, and then conducts quality inspection and safety inspection in a separate inspection area or a joint inspection area according to the target inspection path corresponding to the customs goods to be inspected. During the inspection process, abnormal gases, pathogenic microorganisms, etc. are also inspected by the inspection robot, and the final target inspection results are sent to the information recording module through the transmission device and the recording device (corresponding to the above-mentioned small recording device) for recording and preservation. In addition, the recording device can display the electronic identity tag corresponding to the customs goods to be inspected, and the recording device and the camera system in the park jointly identify and verify the customs goods to be inspected.

[0076] The above customs cargo inspection system has established a model for evaluating the risk level of cross-border cargo. By reading the historical information of ships and cargoes and combining it with the customs database, it quantitatively evaluates the risks of cross-border cargoes in terms of customs clearance frequency, trajectory behavior, historical information and specified characteristics. It classifies cargoes by combining cargo weight, cargo images, historical information of ships and other contents through the model, and adopts different subsequent processing methods to achieve rapid customs clearance of low-risk cargoes and accurate inspection of high-risk cargoes, improve the customs clearance efficiency of unmanned smart ports, and reduce the workload of customs clearance inspection of cross-border cargoes as much as possible. The system optimizes the customs inspection methods for goods with different levels and types of risks, improving customs clearance efficiency. For example, joint inspection of goods with secondary risks can complete the inspection more efficiently than the traditional one-by-one sampling method, and when planning joint inspections, random inspection goods among the third-level risk goods can be inserted according to actual conditions to reduce the waiting time for joint inspections.

[0077] The system has established a safe and efficient information transmission mechanism to realize the encrypted transmission of cargo information, cargo inspection results and other data in the unmanned smart port. On the one hand, it avoids data errors caused by human tampering or operational errors, and on the other hand, it facilitates the reading, writing and confirmation of information, eliminating a large amount of data verification and form filling in traditional ports, realizing intelligent management and improving port efficiency.

[0078] The system has established a set of unmanned detection methods for chemical, biological and other dangerous factors and prohibited items. Compared with the existing automated detection technology, this solution provides the connection method of various automated detection technologies and the operation logic in the port. This logic helps to ensure the authenticity of the information of sampling, detection and other steps, rather than simply connecting the existing technology through a robot with wheels. The contents that cannot be simply deduced include: using a weighing robot to calibrate the weight of the scale, a method that can improve the accuracy of the scale, and can help the weight detection model correctly analyze the risks of smuggling, concealing and other behaviors; the path of multi-sample joint inspection, in which the method of simultaneous detection of four detection compartments takes into account the running time, sampling time and the shortest time required for effective sample collection. If the number is too large, effective sampling cannot be achieved, and if the number is too small, time is wasted. This path is the optimal solution for balancing efficiency and accuracy; container buffer strategy, two buffers are designed according to the risk level of the goods to reduce the risk of cross-infection. According to the risk of cross-infection between goods, the goods with abnormal results of individual detection are placed in the edge area of ​​the park to effectively prevent the spread of pathogenic aerosols in the park. At the same time, considering the impact of transportation distance and mechanical equipment layout on transshipment efficiency, the positional relationship proposed in this plan is an optimal solution that takes into account both risk control and efficiency, and can improve the customs clearance efficiency and inspection accuracy of unmanned smart ports.

[0079] The embodiment of the present invention provides a method for customs cargo inspection, the method comprising the following steps: Step 1: The information verification module receives the declaration list information corresponding to the customs goods to be tested, generates the electronic files and electronic identity tags corresponding to the customs goods to be tested according to the declaration list information, verifies the file information in the electronic files based on the pre-connected database, obtains the verification results, sends the electronic files and the verification results to the information tracking module, and sends the electronic identity tags to the detection module; Step 2: The information tracking module obtains mathematical models corresponding to the preset multiple risk assessment factors, obtains the first parameter corresponding to the customs goods to be tested based on the preset historical archives, electronic archives and inspection results, and determines the risk level corresponding to the customs goods to be tested based on each mathematical model and the first parameter; determines the target detection path corresponding to the customs goods to be tested based on the risk level; wherein the multiple risk assessment factors include: customs clearance frequency, trajectory behavior, historical information and specified characteristics; the first parameter includes: the number of customs clearances of the declaring party corresponding to the customs goods to be tested in the first historical time period, the number of customs clearances of the associated party corresponding to the declaring party in the first historical time period, the customs clearance frequency factor corresponding to the goods belonging to the same category as the customs goods to be tested in the first historical time period, the historical customs clearance qualification rate of the declaring party and the associated party in the first historical time period, the historical punctuality rate of the declaring party and the associated party in the first historical time period, the probability of carrying risky substances in the transportation path, the punctuality rate of arrival at the customs inferred by the preset logistics path model this time, the punctuality rate of arrival at the customs inferred by the logistics path model of the freight ship carrying the customs goods to be tested last time, the operating and financial status of the declaring party, the legal risk of the declaring party, and the material integrity of the declaration list information; Step three: the detection module verifies the identity of the customs goods to be detected according to the electronic identity tag. If the verification is successful, the customs goods to be detected are detected according to the target detection path to obtain the target detection result.

[0080] In the above-mentioned customs goods inspection method, the information inspection module can inspect the electronic files of the customs goods to be inspected, and send the electronic files and the inspection results to the information tracking module, so that the information tracking module can obtain the first parameter corresponding to the customs goods to be inspected. According to the mathematical model corresponding to multiple risk assessment factors and the first parameter, the risk level corresponding to the customs goods to be inspected can be evaluated and obtained. The inspection module can inspect the customs goods to be inspected according to the target inspection path matching the risk level, which helps to improve the customs clearance efficiency of the customs goods to be inspected.

[0081] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A customs cargo inspection system, characterized in that: The system includes: an information checking module, an information tracking module and a detection module; The information verification module is used to receive the declaration list information corresponding to the customs goods to be tested, generate the electronic file and electronic identity tag corresponding to the customs goods to be tested according to the declaration list information, and based on the pre-connected database, verify the file information in the electronic file to obtain the verification result, send the electronic file and the verification result to the information tracking module, and send the electronic identity tag to the detection module; The information tracking module is used to obtain mathematical models corresponding to the preset multiple risk assessment factors, obtain the first parameter corresponding to the customs goods to be inspected based on the preset historical archives, the electronic archives and the inspection results, determine the risk level corresponding to the customs goods to be inspected based on each of the mathematical models and the first parameter, and determine the target inspection path corresponding to the customs goods to be inspected based on the risk level; The detection module is used to verify the identity of the customs goods to be detected according to the electronic identity tag. If the verification is successful, the customs goods to be detected are detected according to the target detection path to obtain a target detection result.

2. The system according to claim 1, characterized in that Multiple risk assessment factors include: customs clearance frequency, trajectory behavior, historical information and specified characteristics; the first parameter includes: the number of customs clearances of the declaring party corresponding to the customs goods to be inspected within the first historical time period, the number of customs clearances of the associated party corresponding to the declaring party within the first historical time period, the customs clearance frequency factor corresponding to the goods belonging to the same category as the customs goods to be inspected in the first historical time period, the historical customs clearance qualification rate of the declaring party and the associated party within the first historical time period, the historical punctuality rate of the declaring party and the associated party within the first historical time period, the probability of carrying risky substances in the transportation route, the punctuality rate of arrival at the customs inferred by the preset logistics path model this time, the punctuality rate of arrival at the customs inferred by the logistics path model last time on the freight ship carrying the customs goods to be inspected, the operating and financial status of the declaring party, the legal risks of the declaring party, and the material completeness of the declaration list information.

3. The system according to claim 2, characterized in that The information tracking module is also used for: Based on the first mathematical model corresponding to the customs clearance frequency and the first designated parameter in the first parameter, a first evaluation result corresponding to the customs clearance frequency is calculated; Based on a second mathematical model corresponding to the trajectory behavior and a second specified parameter in the first parameters, a second evaluation result corresponding to the trajectory behavior is calculated; Based on a third mathematical model corresponding to the historical information and a third specified parameter in the first parameter, a third evaluation result corresponding to the historical information is calculated; Based on a fourth mathematical model corresponding to the specified feature and a fourth specified parameter in the first parameter, a fourth evaluation result corresponding to the specified feature is calculated; Performing weighted calculation on the first evaluation result, the second evaluation result, the third evaluation result, and the fourth evaluation result according to a preset weight coefficient to obtain a comprehensive scoring result corresponding to the customs goods to be inspected; The risk level corresponding to the customs goods to be inspected is determined based on the comprehensive scoring result and the preset risk threshold.

4. The system according to claim 3, characterized in that The first mathematical model is: S1=T1×(1+k)+T2×(1+k)+T3×k-L2×k×f; Wherein, S1 represents the first evaluation result; T1 represents the number of customs clearances of the declarant in the first historical time period; T2 represents the number of customs clearances of the associated party in the first historical time period; T3 represents the customs clearance frequency factor corresponding to the goods belonging to the same category as the customs goods to be inspected in the first historical time period; L2 represents the historical punctuality rate of the declarant and the associated party in the first historical time period; k and f are preset coefficients; The calculation formula corresponding to T3 is: ； Wherein, T1 represents the number of customs clearances of the declaring party in the first historical time period; t i represents the time difference between the customs clearance time of the declaring party at the i-th customs clearance and the customs clearance time of the previous customs clearance during the first historical time period; represents the average of the time differences between two adjacent customs clearances of goods belonging to the same category as the customs goods to be inspected during the first historical time period; represents the standard deviation of the time difference between two adjacent customs clearances of goods belonging to the same category as the customs goods to be inspected in the first historical time period; f1 is a preset coefficient; The second mathematical model is: S2= G1+G2×L2×(1+j); Wherein, S2 represents the second evaluation result; G1 represents the on-time rate of arrival at the customs inferred by the preset logistics path model this time; G2 represents the on-time rate of arrival at the customs inferred by the logistics path model last time on the freight ship carrying the customs goods to be inspected; L2 represents the historical on-time rate of the declarant and the related party in the first historical time period; j is the preset coefficient; Among them, the calculation formula corresponding to G1 is: ； Among them, t represents the estimated arrival time at the customs inferred by the preset logistics path model; represents the average of the historical arrival times corresponding to the transportation path in the first historical time period; represents the standard deviation of the historical arrival time corresponding to the transportation path in the first historical time period; is a polynomial coefficient; k1 is a polynomial coefficient, representing the historical record number of the cargo ship arriving at the customs during the first historical time period; The third mathematical model is: S3=L1×L2+L3×G1×j×u-T1×k×f; Among them, S3 represents the third evaluation result; L1 represents the historical customs clearance pass rate of the declarant and the related party in the first historical time period; L2 represents the historical punctuality rate of the declarant and the related party in the first historical time period; L3 represents the probability of carrying risky substances in the transportation path; G1 represents the punctuality rate of arrival at the customs inferred by the logistics path model this time; T1 represents the number of customs clearances of the declarant in the first historical time period; j, u, k, f are preset coefficients; Among them, the calculation formula corresponding to L3 is: ； in, Represents historical risk data for transportation routes; Represents data associated with environmental factors; Represents data associated with time factors; , and are the weight coefficients of each factor respectively; The fourth mathematical model is: S4 = Q1×Q2×(1+g)+Q3×(1+g)+L1×u; Among them, S4 represents the fourth assessment result; Q1 represents the operating and financial status of the declarant, Q2 represents the legal risk of the declarant, and Q3 represents the material completeness of the declaration list information; g and u are preset coefficients; Among them, the calculation formula corresponding to Q1 is: ； Where q represents the current financial ratio of the reporting party, represents the average value of the financial ratio of the declaring party during the first historical period; represents the standard deviation of the financial ratio of the declaring party during the first historical period; , represents polynomial coefficients; represents the weight of the ith financial risk of the declaring party during the first historical period; and n represents the total number of times the financial risk of the declaring party is assessed during the first historical period.

5. The system according to claim 1, characterized in that The detection module is also used for: After the customs goods to be inspected are placed on a preset weighing device, the initial weight data of the customs goods to be inspected are obtained through the weighing device; Placing a preset weighing robot at a first preset position on the weighing device, and obtaining first total weight data of the customs goods to be inspected and the weighing robot through the weighing device; wherein the weighing robot is configured with a counterweight that matches the initial weight data; Controlling the weighing robot to reach a second preset position on the weighing device, and obtaining second total weight data of the customs goods to be inspected and the weighing robot through the weighing device; Controlling the weighing robot to reach a third preset position on the weighing device, and obtaining third total weight data of the to-be-inspected customs goods and the weighing robot through the weighing device; Controlling the weighing robot to reach a fourth preset position on the weighing device, and obtaining fourth total weight data of the customs goods to be inspected and the weighing robot through the weighing device; wherein the first preset position, the second preset position, the third preset position and the fourth preset position are positions of four corners on the weighing device; Controlling the weighing robot to return to a first preset position on the weighing device, and obtaining fifth total weight data of the to-be-inspected customs goods and the weighing robot through the weighing device; Inputting the initial weight data, the first total weight data, the second total weight data, the third total weight data, the fourth total weight data, and the fifth total weight data into a pre-trained weight detection model, so as to output the weighing data of the customs goods to be detected through the weight detection model; Obtaining the declaration data of the customs goods to be inspected; wherein the declaration data includes: declared quantity, declared type and declared weight; Sending the declared data and the weighing data to a pre-trained recognition model to confirm through the recognition model whether the weight difference between the weighing data and the declared weight meets a preset fluctuation range; If the weight difference satisfies the preset fluctuation range, the weighing data is determined as the target weight data corresponding to the customs goods to be inspected; If the weight difference does not satisfy the preset fluctuation range, the customs goods to be inspected are transferred to a preset first buffer zone so that customs inspection personnel can inspect the customs goods to be inspected.

6. The system according to claim 1, characterized in that The detection module is also used for: If the risk level of the customs goods to be inspected is level one risk, determining that the target inspection path corresponding to the customs goods to be inspected is a separate inspection path; According to the separate inspection path, the customs goods to be inspected are transferred to a separate inspection room, and the container of the customs goods to be inspected is opened by a hook device, so that the goods body of the customs goods to be inspected is inspected by an online inspection device in the separate inspection room to obtain a first goods inspection result; wherein the first goods inspection result is used to indicate whether there are any illegally carried items in the goods body; In the process of detecting the cargo body of the customs cargo to be detected by the online detection device, the environment in the separate detection room is detected by the detection robot to obtain a first environment detection result; wherein the first environment detection result is used to indicate whether there is abnormal gas and / or pathogenic microorganisms in the environment of the separate detection room; If the first cargo inspection result indicates that there are no illegally carried items in the cargo body, and the first environment inspection result indicates that there are no abnormal gases and pathogenic microorganisms in the environment of the separate inspection room, the customs clearance of the cargo to be inspected is allowed; If the first cargo inspection result indicates that there are illegally carried items in the cargo body, and / or the first environmental inspection result indicates that there are abnormal gases and / or pathogenic microorganisms in the environment of the separate inspection room, the customs cargo to be inspected will be transferred to a preset first buffer zone so that customs inspection personnel can conduct inspection and quarantine on the customs cargo to be inspected.

7. The system according to claim 6, characterized in that The detection module is also used for: If the risk level of the customs goods to be inspected is level 2 risk, determining that the target inspection path corresponding to the customs goods to be inspected is a joint inspection path; According to the joint inspection route, the customs goods to be inspected are transferred to one of the inspection compartments in the joint inspection room; wherein the joint inspection room includes four inspection compartments; When the customs goods to be inspected are placed in the four inspection compartments, each container of the customs goods to be inspected is opened by a hook device, so that the cargo bodies of the four customs goods to be inspected are mixedly inspected by the online inspection device in the joint inspection room to obtain a second cargo inspection result; wherein the second cargo inspection result is used to indicate whether there are any illegally carried items in the cargo body of at least one of the customs goods to be inspected; In the process of performing mixed detection on the cargo bodies of the four customs cargoes to be detected by the online detection device, the detection robot performs mixed detection on the environments in the four detection compartments to obtain a second environment detection result; wherein the second environment detection result is used to indicate whether there is abnormal gas and / or pathogenic microorganisms in the environment of at least one detection compartment; If the second cargo inspection result indicates that there are no illegally carried items in the cargo bodies of the four customs cargoes to be inspected, and the second environment inspection result indicates that there are no abnormal gases and pathogenic microorganisms in the environments of the four inspection compartments, the four customs cargoes to be inspected are allowed to pass customs; If the second cargo inspection result indicates that there are illegally carried items in the cargo body of at least one of the customs goods to be inspected, and / or the second environment inspection result indicates that there are abnormal gases and / or pathogenic microorganisms in the environment of at least one inspection compartment, the risk level of each of the customs goods to be inspected in the four inspection compartments is upgraded to level one risk and transferred to the preset second buffer zone, so that the four customs goods to be inspected that have been upgraded to level one risk will be re-inspected separately according to the separate inspection path.

8. The system according to claim 7, characterized in that The system further comprises: a risk control module; the risk control module is used to: Obtaining the quantity of customs goods to be inspected in the joint inspection room; If the number of customs goods to be inspected in the joint inspection room is less than four within the first preset time period, and it is determined according to the pre-trained logistics prediction model that no customs goods to be inspected with a risk level of Level 2 will arrive at the customs within the first future time period, the first designated customs goods are selected from the customs goods to be inspected with a risk level of Level 3 according to the preset first selection rule; The risk level of the first designated customs goods is upgraded to level 2 risk, and the goods are transported to the joint inspection room, so that the customs goods to be inspected in the joint inspection room are inspected according to the joint inspection path.

9. The system according to claim 8, characterized in that A plurality of joint inspection rooms constitute a joint inspection area; a plurality of separate inspection rooms constitute a separate inspection area; the risk control module is used to: Calculating a first filling degree of the joint inspection area according to the quantity of customs goods to be inspected in each of the joint inspection rooms; Calculating a second filling degree of the separate inspection area according to the quantity of customs to be inspected in each of the separate inspection rooms; If the first filling degree reaches a preset filling degree threshold, and the second filling degree is lower than the preset filling degree threshold, and it is determined according to the logistics prediction model that in the second future time period, there are customs goods to be inspected with a risk level of level 2 arriving at the customs, then according to the preset second selection rule, the second designated customs goods are selected from the customs goods to be inspected with a risk level of level 2; Raising the risk level of the second designated customs goods to level one risk, and transferring the second designated customs goods to the separate inspection room, so as to inspect the second designated customs goods according to the separate inspection path; If both the first fullness and the second fullness reach the preset fullness thresholds, and it is determined according to the logistics prediction model that within a third future time period, there are customs goods to be inspected with a risk level of target risk arriving at the customs, a deceleration instruction is generated so that the mechanical transport device reduces the transport rate of the customs goods to be inspected that have arrived at the customs and have not been inspected according to the deceleration instruction; wherein the target risk is: primary risk and / or secondary risk.

10. The system according to claim 1, characterized in that The detection module is also used for: If the risk level of the customs goods to be inspected is level three risk and the customs goods to be inspected have not been selected, the customs goods to be inspected are allowed to pass customs.

11. The system according to claim 1, characterized in that The system further comprises: an information recording module; The information recording module is used to receive the target detection result, and convert the format of the target detection result according to a preset format to obtain a converted detection result; The converted detection result is saved in the blockchain.

12. A customs cargo inspection method, characterized in that: The method comprises: The information checking module receives the declaration list information corresponding to the customs goods to be tested, generates the electronic file and electronic identity tag corresponding to the customs goods to be tested according to the declaration list information, checks the file information in the electronic file based on the pre-connected database, obtains the checking result, sends the electronic file and the checking result to the information tracking module, and sends the electronic identity tag to the detection module; The information tracking module obtains mathematical models corresponding to the preset multiple risk assessment factors, obtains a first parameter corresponding to the customs goods to be inspected based on the preset historical archives, the electronic archives and the inspection results, determines the risk level corresponding to the customs goods to be inspected based on each of the mathematical models and the first parameter, and determines the target inspection path corresponding to the customs goods to be inspected based on the risk level; The detection module verifies the identity of the customs goods to be detected according to the electronic identity tag. If the verification is successful, the customs goods to be detected are detected according to the target detection path to obtain a target detection result.

Images