Dynamic sorting optimization control method and system in warehouse logistics

Abstract

The invention is suitable for the technical field of sorting control, and particularly relates to a dynamic sorting optimization control method and system in warehouse logistics, and the method comprises the steps: reading bar code data of all packages in warehouse logistics, and collecting attribute information which at least comprises sender information, cargo types, package sources and receiving addresses; selecting a plurality of trust units based on a preset evaluation rule; and when the attribute information contains a trusted unit, defining the corresponding parcel as a trusted parcel. By comparing the X-ray images with the target features, automatic sampling inspection of credible parcels can be achieved, manual dependence is reduced, the sorting efficiency is greatly improved, by conducting sampling inspection of different proportions on different parcel sources, high-risk parcel sources can be effectively focused, sorting resource configuration is optimized, the detection rate of risk parcels is increased, and the sorting efficiency is improved. Potential safety hazards are reduced, and efficient and stable operation of the sorting assembly line is guaranteed.

Description

Technical Field

[0001] The present invention relates to the field of sorting control technology, and in particular to a dynamic sorting optimization control method and system in warehousing logistics. Background Art

[0002] Sorting control refers to the process of optimizing and adjusting the sorting process according to package information, equipment status and operational requirements during the logistics and warehousing process.

[0003] In existing logistics sorting systems, to ensure safety and compliance, packages must be identified before the sorting process begins to determine whether any abnormal items are present. This identification is primarily accomplished through X-ray image scanning and computer vision recognition technology. However, comprehensive identification of all incoming packages, limited by current computing resources and image processing capabilities, would significantly increase the data processing load on the sorting system, leading to reduced sorting efficiency and making it difficult to meet the needs of large-scale, high-throughput parcel sorting in logistics scenarios.

[0004] Therefore, “how to conduct random inspections on packages with reliable sources” is the technical problem that the present invention needs to solve. Summary of the Invention

[0005] The purpose of the present invention is to provide a dynamic sorting optimization control method and system in warehousing logistics to solve the problem of "how to perform random inspections on packages with reliable sources" raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions: A dynamic sorting optimization control method in warehousing logistics, the method comprising: Read the barcode data of all packages in the warehouse logistics and collect attribute information, where the attribute information includes at least: sender information, cargo type, package source and delivery address. Based on the preset evaluation rules, select several trust units; When the attribute information contains a trust unit, the corresponding package is defined as a trusted package, the initial frequency of each package source is set, and the trusted packages are randomly inspected according to the initial frequency; Construct a package feature set, and select target features from the package feature set based on the cargo type. Use the X-ray security inspection machine pre-integrated in the sorting line to collect X-ray images of all trusted packages, construct an image comparison model, and train it using a pre-constructed historical dataset. Input the X-ray images and target features into the image comparison model to determine whether they are identical. If yes, continue to sort the trusted packages; if not, define the corresponding package source as a risk source, increase the initial frequency of the risk source, obtain the revised frequency, and perform random inspections on the trusted packages according to the revised frequency.

[0007] Furthermore, the step of reading the barcode data of the package in the warehousing logistics and identifying the attribute information includes: Receive image data taken by the pickup personnel and define the corresponding trusted packages as inspection-exempt units; An inspection-free mechanism is created based on the inspection-free unit.

[0008] Furthermore, when the attribute information includes a trust unit, the step of defining the corresponding package as a trusted package includes: Configuring factors influencing the trustworthy package, wherein the factors influencing the trustworthy package include at least: sender information and pickup personnel evaluation; Based on the feedback information uploaded by the pickup personnel, the trusted packages are added or removed.

[0009] Furthermore, the step of constructing a package feature set and selecting target features from the package feature set according to the cargo type includes: Based on the cargo type, the trusted packages are clustered into several categories, a standard image is selected from each category, and image features are extracted; Integrate all image features to generate a package feature set.

[0010] Furthermore, the step of collecting X-ray images of all trusted packages using an X-ray security inspection machine pre-integrated in the sorting line includes: Collect sensor data in the sorting line, wherein the sensor data includes at least: package weight, volume and infrared data; Compare the attribute information and the sensor data to determine whether there are any conflicting items. If so, write the conflicting items into a preset template, generate a disassembly and inspection task, and send the disassembly and inspection task to a preset terminal.

[0011] Furthermore, the step of increasing the initial frequency of the risk source to obtain a revised frequency, and performing random inspections on trusted packages according to the revised frequency includes: Obtain sampling inspection results and dynamically adjust the correction frequency; Integrate the inspection results and adjustments, generate an event report, and send it to the preset terminal.

[0012] Furthermore, the method further comprises: Construct a hidden danger feature set, compare the X-ray image with the hidden danger feature set, and determine whether all packages contain hidden danger features; If so, edit the emergency rules corresponding to the hidden danger characteristics one by one, and activate the corresponding emergency rules.

[0013] Furthermore, the system includes: A selection module is used to read the barcode data of all packages in the warehouse logistics, collect attribute information, where the attribute information includes at least: sender information, cargo type, package source and receiving address, and select a number of trust units based on preset evaluation rules; a random inspection module, configured to define a corresponding package as a trusted package when the attribute information includes a trust unit, set an initial frequency for each package source, and perform random inspections on the trusted packages according to the initial frequency; A comparison module is used to construct a package feature set and select target features from the package feature set based on the cargo type. X-ray images of all trusted packages are collected using an X-ray security inspection machine pre-integrated in the sorting line. An image comparison model is constructed and trained using a pre-constructed historical dataset. The X-ray images and target features are then input into the image comparison model. The judgment module is used to determine whether the two are the same. If so, the trusted packages will continue to be sorted. If not, the corresponding package source will be defined as a risk source, the initial frequency of the risk source will be increased, the corrected frequency will be obtained, and the trusted packages will be randomly inspected according to the corrected frequency.

[0014] Furthermore, the selection module includes: A definition unit is used to receive image data taken by the pickup personnel and define the corresponding trusted package as an inspection-exempt unit; A creation unit is used to create an inspection-free mechanism based on the inspection-free unit.

[0015] Furthermore, the sampling inspection module includes: a configuration unit, configured to configure influencing factors of the credible package, wherein the influencing factors include at least: sender information and pickup personnel evaluation; The adding and subtracting unit is used to add and subtract the trusted packages according to the feedback information uploaded by the pickup personnel.

[0016] Compared with the prior art, the present invention has the following beneficial effects: By determining the trust units, a data basis can be provided for adjusting the sorting frequency and strategy, which is conducive to the differentiated sorting of packages. By conducting random inspections on trusted packages, sorting delays caused by comprehensive inspections can be avoided while ensuring the safety and compliance of the packages, effectively improving the overall sorting efficiency. By comparing X-ray images and target features, automated inspections of trusted packages can be achieved, reducing manual dependence and greatly improving sorting efficiency. By conducting different proportions of inspections on different package sources, it is possible to effectively focus on high-risk package sources, optimize sorting resource allocation, improve the detection rate of risky packages, reduce safety hazards, and ensure the efficient and stable operation of the sorting line. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flowchart of a dynamic sorting optimization control method in warehousing logistics provided by an embodiment of the present invention; Figure 2 A block diagram of the first sub-process of the dynamic sorting optimization control method in warehousing logistics provided by an embodiment of the present invention; Figure 3 A second sub-flow chart of the dynamic sorting optimization control method in warehousing logistics provided by an embodiment of the present invention; Figure 4 A block diagram of the third sub-process of the dynamic sorting optimization control method in warehousing logistics provided by an embodiment of the present invention; Figure 5 A fourth sub-flow chart of the dynamic sorting optimization control method in warehousing logistics provided by an embodiment of the present invention; Figure 6 A block diagram of the dynamic sorting optimization control system in warehousing logistics provided by an embodiment of the present invention; Figure 7 A block diagram of the components of the selection module in the dynamic sorting optimization control system in warehousing logistics provided by an embodiment of the present invention; Figure 8 A block diagram of the composition of the sampling inspection module in the dynamic sorting optimization control system in warehousing logistics provided by an embodiment of the present invention; Figure 9 A block diagram of the composition of a comparison module in a dynamic sorting optimization control system in warehousing logistics provided by an embodiment of the present invention; Figure 10 This is a block diagram of the composition of the judgment module in the dynamic sorting optimization control system in warehousing logistics provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0019] In Example 1, Figure 1 The implementation process of the dynamic sorting optimization control method in warehousing logistics provided by the embodiment of the present invention is shown and described in detail below: S100: Read the barcode data of all packages in the warehouse logistics and collect attribute information, where the attribute information at least includes: sender information, cargo type, package source and receiving address, and select several trust units based on preset evaluation rules.

[0020] In the sorting line, the package barcode is generally scanned before sorting to use the logistics information therein for sorting; specifically, it includes: using a fixed laser scanner or CCD camera to read the barcode of the package. The barcode here refers to the identification code composed of parallel black lines of varying thicknesses on the express delivery label; the barcode serves as the unique identifier of the package and carries rich logistics and business information, such as attribute information; attribute information includes not only: sender information, cargo type, package source and receiving address, but also: packaging type, shipping time and sender, recipient contact information, etc.; it should be noted that the sender information refers to the specific sender, and the package source refers to the unit to which the package belongs; for example, among multiple packages sent by an e-commerce store, the sender information is the warehouse staff in the store, and the package source is also the e-commerce store.

[0021] Based on the evaluation rules, several trust units are selected from the package sources. The evaluation rules are pre-established by the sorting process manager. The evaluation rules can be: if the package source is an official store, authorized unit, administrative department, etc., it is defined as a trust unit. The trust unit can also be determined based on the credibility level or historical records of the package source.

[0022] S200: When the attribute information includes a trust unit, the corresponding package is defined as a trusted package, an initial frequency of each package source is set, and the trusted packages are randomly inspected according to the initial frequency.

[0023] After the trust unit is determined, all packages sent or received by the trust unit are defined as trusted packages. Trusted packages are packages with reliable and stable sources and low risk. An initial frequency is set for each type of trusted package. For example, if the package comes from the official store corresponding to daily chemical products, an initial frequency of 5 / time can be set, that is, 1 package is randomly selected from 5 packages. If the package comes from an administrative department, an initial frequency of 15 / time can be set, that is, 1 package is randomly selected from every 15 packages. The initial frequency should be determined by the sorting process manager based on actual experience.

[0024] In this application, random inspections of trusted packages are mainly conducted to identify abnormal packages, which mainly have abnormal physical characteristics (damaged packaging, liquid leakage) and abnormal contents (inconsistent with the declared item type, suspected of carrying contraband, dangerous goods, or concealed high-value items, etc.).

[0025] S300: Construct a package feature set, and select target features from the package feature set based on the cargo type. Use the X-ray security inspection machine pre-integrated in the sorting line to collect X-ray images of all trusted packages, construct an image comparison model, and use the pre-constructed historical data set for training. Input the X-ray images and target features into the image comparison model.

[0026] During daily sorting, X-ray images of all packages are collected, and image features are extracted and divided into several categories. According to the categories, the image features are integrated to generate a package feature set, where the package feature set refers to a collection of image feature vectors obtained through image processing and feature extraction algorithms based on X-ray images of packages of all categories. When sorting packages, the image features of the same package are extracted from the package feature set according to the cargo type in the package attribute information, and defined as target features. X-ray images of all trusted packages are collected using an X-ray security inspection machine, and the X-ray images and target features are input into the image comparison model to output the comparison results. The image comparison model should be trained in advance using historical data sets.

[0027] S400: Determine whether the two are the same. If yes, continue sorting the trusted packages. If not, define the corresponding package source as a risk source, increase the initial frequency of the risk source, obtain the revised frequency, and perform random inspections on the trusted packages according to the revised frequency.

[0028] If the comparison result shows that the X-ray image and the target features are the same, the sorting line will continue to be used to conduct random inspections on trusted packages. If the X-ray image and the target features are different, that is, the actual goods in the package are different from the type of goods recorded on the waybill, the corresponding package source will be defined as a risk source, and the initial frequency corresponding to the risk source will be increased to obtain a corrected frequency. In the subsequent random inspections of packages, the packages will be inspected according to this corrected frequency.

[0029] For example, the initial frequency of an official store for daily chemical products is 5 / times. During cargo sorting, the cargo type recorded on the package label is toothpaste, but a random inspection reveals that the cargo in the corresponding package is liquid (the toothpaste was mistakenly sent as shampoo). The package is opened and inspected, and in the subsequent random inspections, the initial frequency of the official store is increased to 3 / times. The reason for the discrepancy between the content recorded on the package label and the actual content may be a delivery error. Although shampoo is also a regular item, it also indicates that there may be loopholes in the official store's delivery management. To minimize the risk, the random inspection frequency should be increased.

[0030] In Example 2, Figure 2 The implementation process of the dynamic sorting optimization control method in warehousing logistics provided by an embodiment of the present invention is shown. The steps of reading the barcode data of the package in the warehousing logistics and identifying the attribute information are described in detail as follows: S101: Receive image data taken by the pickup personnel and define the corresponding package as an inspection-free unit.

[0031] In actual logistics, some goods are packed by pickup personnel. After the pickup personnel take image data of the goods in the package, the image data is uploaded to the logistics management platform, and the corresponding package is defined as an inspection-free unit.

[0032] S102: creating an inspection-free mechanism based on the inspection-free unit.

[0033] In actual sorting, the inspection-free mechanism is activated, where the inspection-free mechanism is: when the packages that have not been inspected and the inspection-free units in the trusted units pass through the sorting pipeline, they are sorted normally, but no X-ray images are collected and no data processing is performed.

[0034] In Example 3, Figure 3 The implementation process of the dynamic sorting optimization control method in warehousing logistics provided by an embodiment of the present invention is shown. The following details the steps of defining the corresponding package as a trusted package when the attribute information includes a trust unit, as follows: S201: Configuring factors influencing the trustworthy package, wherein the factors influencing the trustworthy package include at least sender information and pickup personnel evaluation.

[0035] As for which packages may be considered trusted packages, in addition to packages sent from official stores, authorized units or administrative departments, some packages can also be defined as trusted packages based on sender information and the evaluation of the pickup personnel. For example, if a pickup personnel evaluates that the goods were not packed at the time of pickup and were packed by the logistics station staff, the corresponding package can be defined as a trusted package because the package has been manually inspected by the logistics station staff.

[0036] S202: Based on the feedback information uploaded by the pickup personnel, the trusted packages are added or removed.

[0037] Trusted packages are added or removed based on the feedback information uploaded by the pickup personnel. The target terminal for uploading feedback information is the logistics management platform. "Add" means defining a package as a trusted package, and "remove" means deleting a package from the trusted package list.

[0038] In Example 4, Figure 4 The implementation process of the dynamic sorting optimization control method in warehousing logistics provided by an embodiment of the present invention is shown. The steps of constructing a package feature set and selecting target features from the package feature set based on the cargo type are described in detail below: S301: Based on the cargo type, the trusted packages are clustered into several categories, a standard image is selected from each category, and image features are extracted.

[0039] Trusted packages are clustered into several categories, such as electronics, apparel, books, and daily necessities, based on their physical properties, packaging form, and size. Representative samples are selected from each category, and image data collected when they pass through the X-ray security inspection machine, namely standard images, is obtained. Image features are then extracted, including edge structure, texture pattern, density distribution, and perspective characteristics.

[0040] S302: Integrate all image features to generate a package feature set.

[0041] Image features of all categories are uniformly integrated to generate a complete package feature set.

[0042] In Example 5, Figure 4 The implementation process of the dynamic sorting optimization control method in warehousing logistics provided by an embodiment of the present invention is shown. The following details the steps of using the X-ray security inspection machine pre-integrated in the sorting line to collect X-ray images of all trusted packages, as follows: S303: Collect sensor data in the sorting line, wherein the sensor data at least includes: package weight, volume and infrared data.

[0043] During the parcel sorting process, sensor data is collected using sensor equipment pre-integrated in the sorting line (such as dynamic weighing sensors and laser volume measuring instruments).

[0044] S304: Compare the attribute information and the sensor data to determine whether there are any conflicting items. If so, write the conflicting items into a preset template, generate an inspection task, and send the inspection task to a preset terminal.

[0045] The attribute information of each package (including sender information, cargo type, declared weight, volume, delivery address, etc.) and the corresponding sensor data (such as dynamic weighing, volume measurement, infrared sensing, etc.) are compared and analyzed. If obvious contradictions are found between the attribute information and the sensor data during the comparison process, such as the package label indicates a light item but the measured weight is far higher than the normal range, the declared volume does not match the measured volume, the cargo type is inconsistent with the image or infrared feature, etc., the information related to the contradiction is written into the preset inspection task template, an inspection task is generated, and the inspection task is sent to the preset terminal, where the preset terminal can be the inspection process management personnel terminal.

[0046] In Example 6, Figure 5 The implementation process of the dynamic sorting optimization control method in warehousing logistics provided by an embodiment of the present invention is shown. The following details the steps of increasing the initial frequency of the risk source, obtaining the revised frequency, and performing random inspections on trusted packages according to the revised frequency. S401: Obtain the sampling inspection results and dynamically adjust the correction frequency.

[0047] After each random inspection of trusted packages, the correction frequency is dynamically adjusted based on the inspection results.

[0048] S402: Integrate the inspection results and adjustments, generate an event report, and send it to a preset terminal.

[0049] Generate an event report based on the inspection results and the inspection frequencies before and after adjustment, and send the event report to the sorting process manager terminal.

[0050] In Example 7, different from Example 1, in this embodiment of the present invention, the method further includes: Construct a hidden danger feature set, compare the X-ray image with the hidden danger feature set, and determine whether all packages contain hidden danger features; If yes, edit the emergency rules corresponding to the hidden danger characteristics and activate the corresponding emergency rules A hidden danger feature set is constructed based on the X-ray image data of packages such as illegal items, abnormal structures and dangerous materials. In other words, the hidden danger feature set includes typical image features of various known high-risk items under X-ray images, such as density anomalies, structural anomalies, occlusion features or the perspective performance of specific materials. The X-ray image of each package is compared with the hidden danger feature set one by one, and the image recognition and feature matching algorithm is used to determine whether the X-ray images of all packages contain image features similar to the hidden danger feature set. If so, the corresponding image features are defined as hidden danger features, and the corresponding emergency rules are activated at the same time. Each hidden danger feature corresponds to an emergency rule, which is pre-established by professionals. The emergency rules are also the methods for handling the corresponding hidden danger features, such as reporting the hidden danger features, contacting relevant departments for handling, etc.

[0051] Figure 6 The following is a structural block diagram of a dynamic sorting optimization control system for warehousing and logistics provided by an embodiment of the present invention. The dynamic sorting optimization control system 1 for warehousing and logistics includes: The selection module 11 is used to read the barcode data of all packages in the warehouse logistics, collect attribute information, where the attribute information includes at least: sender information, cargo type, package source and receiving address, and select a number of trust units based on preset evaluation rules; A random inspection module 12 is configured to define a corresponding package as a trusted package when the attribute information includes a trust unit, set an initial frequency for each package source, and perform random inspections on the trusted packages according to the initial frequency; The comparison module 13 is configured to construct a package feature set and select target features from the package feature set based on the cargo type. The module collects X-ray images of all authentic packages using an X-ray security inspection machine pre-integrated in the sorting line, constructs an image comparison model, and trains the model using a pre-constructed historical dataset. The X-ray images and target features are then input into the image comparison model. The judgment module 14 is used to judge whether the two are the same. If so, the trusted packages are continued to be sorted. If not, the corresponding package source is defined as a risk source, the initial frequency of the risk source is increased, the revised frequency is obtained, and the trusted packages are randomly inspected according to the revised frequency.

[0052] Figure 7 The following is a structural block diagram of a dynamic sorting optimization control system in warehousing logistics provided by an embodiment of the present invention. The selection module 11 includes: A definition unit 111 is used to receive image data taken by a pickup person and define the corresponding package as an inspection-free unit; The creation unit 112 is configured to create an inspection-free mechanism based on the inspection-free unit.

[0053] Figure 8 The following is a structural block diagram of a dynamic sorting optimization control system in warehousing logistics provided by an embodiment of the present invention. The sampling inspection module 12 includes: A configuration unit 121 is configured to configure factors affecting the trustworthy package, wherein the factors at least include: sender information and pickup personnel evaluation; The adding and subtracting unit 122 is configured to add and subtract the trusted packages according to the feedback information uploaded by the pickup personnel.

[0054] Figure 9 The following is a structural block diagram of a dynamic sorting optimization control system in warehousing logistics according to an embodiment of the present invention. The comparison module 13 includes: An extraction unit 131 is configured to cluster the trusted packages into a plurality of categories according to the cargo type, select a standard image from each category, and extract image features; A generating unit 132 is used to integrate all image features and generate a package feature set; The collection unit 133 is used to collect sensor data in the sorting line, wherein the sensor data includes at least: package weight, volume and infrared data; The writing unit 134 is used to compare the attribute information and the sensor data to determine whether there are any conflicting items. If so, the conflicting items are written into a preset template, a disassembly and inspection task is generated, and the disassembly and inspection task is sent to a preset terminal.

[0055] Figure 10 The following is a structural block diagram of a dynamic sorting optimization control system in warehousing logistics provided by an embodiment of the present invention. The judgment module 14 includes: An adjustment unit 141 is used to obtain the sampling inspection results and dynamically adjust the correction frequency; The sending unit 142 is used to integrate the sampling inspection results and adjustments, generate an event report, and send it to a preset terminal.

[0056] The selection module 11 is mainly used to complete step S100, the sampling module 12 is mainly used to complete step S200, the comparison module 13 is mainly used to complete step S300, and the judgment module 14 is mainly used to complete step S400; The definition unit 111 is mainly used to complete step S101, and the creation unit 112 is mainly used to complete step S102; The configuration unit 121 is mainly used to complete step S201, and the increase / decrease unit 122 is mainly used to complete step S202; The extraction unit 131 is mainly used to complete step S301, the generation unit 132 is mainly used to complete step S302, the collection unit 133 is mainly used to complete step S303, and the writing unit 134 is mainly used to complete step S304; The adjusting unit 141 is mainly used to complete step S401, and the issuing unit 142 is mainly used to complete step S402.

[0057] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0058] The above-described embodiments merely illustrate several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

[0059] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A dynamic sorting optimization control method in warehousing logistics, characterized in that: The method comprises: Read the barcode data of all packages in the warehouse logistics and collect attribute information, where the attribute information includes at least: sender information, cargo type, package source and delivery address. Based on the preset evaluation rules, select several trust units; When the attribute information contains a trust unit, the corresponding package is defined as a trusted package, the initial frequency of each package source is set, and the trusted packages are randomly inspected according to the initial frequency; Construct a package feature set, and select target features from the package feature set based on the cargo type. Use the X-ray security inspection machine pre-integrated in the sorting line to collect X-ray images of all trusted packages, construct an image comparison model, and train it using a pre-constructed historical dataset. Input the X-ray images and target features into the image comparison model to determine whether they are identical. If yes, continue to sort the trusted packages; if not, define the corresponding package source as a risk source, increase the initial frequency of the risk source, obtain the revised frequency, and perform random inspections on the trusted packages according to the revised frequency.

2. The dynamic sorting optimization control method in warehousing logistics according to claim 1 is characterized in that: The step of reading the barcode data of the package in the warehousing logistics and identifying the attribute information includes: Receive image data taken by the pickup personnel and define the corresponding package as an inspection-free unit; An inspection-free mechanism is created based on the inspection-free unit.

3. The dynamic sorting optimization control method in warehousing logistics according to claim 2 is characterized in that: When the attribute information includes a trust unit, the step of defining the corresponding package as a trusted package includes: Configuring factors influencing the trustworthy package, wherein the factors influencing the trustworthy package include at least: sender information and pickup personnel evaluation; Based on the feedback information uploaded by the pickup personnel, the trusted packages are added or removed.

4. The dynamic sorting optimization control method in warehousing logistics according to claim 1 is characterized in that: The step of constructing a package feature set and selecting target features from the package feature set according to the cargo type includes: Based on the cargo type, the trusted packages are clustered into several categories, a standard image is selected from each category, and image features are extracted; Integrate all image features to generate a package feature set.

5. The dynamic sorting optimization control method in warehousing logistics according to claim 1 is characterized in that: The steps of collecting X-ray images of all trusted packages using an X-ray security inspection machine pre-integrated in the sorting line include: Collect sensor data in the sorting line, wherein the sensor data includes at least: package weight, volume and infrared data; Compare the attribute information and the sensor data to determine whether there are any conflicting items. If so, write the conflicting items into a preset template, generate a disassembly and inspection task, and send the disassembly and inspection task to a preset terminal.

6. The dynamic sorting optimization control method in warehousing logistics according to claim 5 is characterized in that: The steps of increasing the initial frequency of the risk source to obtain a revised frequency, and performing random inspections on trusted packages according to the revised frequency include: Obtain sampling inspection results and dynamically adjust the correction frequency; Integrate the inspection results and adjustments, generate an event report, and send it to the preset terminal.

7. The dynamic sorting optimization control method in warehousing logistics according to claim 1 is characterized in that: The method further comprises: Construct a hidden danger feature set, compare the X-ray image with the hidden danger feature set, and determine whether all packages contain hidden danger features; If so, edit the emergency rules corresponding to the hidden danger characteristics one by one, and activate the corresponding emergency rules.

8. A dynamic sorting optimization control system in warehousing logistics, characterized in that: The system comprises: A selection module is used to read the barcode data of all packages in the warehouse logistics, collect attribute information, where the attribute information includes at least: sender information, cargo type, package source and receiving address, and select a number of trust units based on preset evaluation rules; a random inspection module, configured to define a corresponding package as a trusted package when the attribute information includes a trust unit, set an initial frequency for each package source, and perform random inspections on the trusted packages according to the initial frequency; A comparison module is used to construct a package feature set and select target features from the package feature set based on the cargo type. X-ray images of all trusted packages are collected using an X-ray security inspection machine pre-integrated in the sorting line. An image comparison model is constructed and trained using a pre-constructed historical dataset. The X-ray images and target features are then input into the image comparison model. The judgment module is used to determine whether the two are the same. If so, the trusted packages will continue to be sorted. If not, the corresponding package source will be defined as a risk source, the initial frequency of the risk source will be increased, the corrected frequency will be obtained, and the trusted packages will be randomly inspected according to the corrected frequency.

9. The dynamic sorting optimization control system in warehousing logistics according to claim 8, characterized in that: The selection module includes: A definition unit, used to receive image data taken by the pickup personnel and define the corresponding package as an inspection-exempt unit; A creation unit is used to create an inspection-free mechanism based on the inspection-free unit.

10. The dynamic sorting optimization control system in warehousing logistics according to claim 9, characterized in that: The sampling inspection module includes: a configuration unit, configured to configure influencing factors of the credible package, wherein the influencing factors include at least: sender information and pickup personnel evaluation; The adding and subtracting unit is used to add and subtract the trusted packages according to the feedback information uploaded by the pickup personnel.

Images