Headless package matching method, device and equipment

Abstract

The present invention discloses a method, device and equipment for matching headless packages. The method is applied to the field of logistics technology. The method includes: inputting sample logistics data of sample headless packages into a machine learning model to extract package features and transportation information, and obtaining sample package features and sample transportation information; obtaining headless package attribute information of each headless package in a database; based on the sample package features and sample transportation information, searching the database for headless package attribute information that matches the sample headless package, and outputting the sample headless package attribute information; based on the difference between the sample headless package attribute information and the sample headless package attribute label, obtaining a headless package information matching model; obtaining logistics data of the headless package to be matched, inputting the logistics data into the headless package information matching model, and obtaining the headless package attribute information corresponding to the headless package to be matched. The present invention can quickly identify and process headless packages, thereby improving the overall logistics efficiency.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method, device and equipment for matching headless packages. Background Art

[0002] In the express logistics industry, accurate identification and matching of parcels is a key link to ensure delivery efficiency and customer satisfaction. However, due to various reasons (such as label detachment, barcode damage, missing information, etc.), some parcels may have "headless" problems, that is, parcels that cannot be identified and matched by conventional methods. These problems not only affect the timeliness and accuracy of delivery, but also increase the operating costs and management difficulties of express companies. Summary of the invention

[0003] The present invention provides a method, device and equipment for matching packages with headerless items, which can reduce logistics losses caused by headerless items and improve the delivery rate of packages; and enhance customer service experience through real-time monitoring and rapid response.

[0004] In one aspect, the present invention provides a method for matching a package without a header, the method comprising: Obtaining sample logistics data of a sample headerless package in a logistics platform; the sample headerless package is marked with a sample headerless attribute label; Inputting the sample logistics data of the sample headerless package into the machine learning model to extract package features and transportation information, and obtain sample package features and sample transportation information; Get the headless item attribute information of each headless item in the database; Based on the sample package characteristics and the sample transportation information, searching the database for headless item attribute information matching the sample headless item package, and outputting the sample headless item attribute information; Based on the difference between the sample headless item attribute information and the sample headless item attribute label, the machine learning model is trained to obtain a headless item information matching model; The logistics data of the headless item to be matched is obtained, and the logistics data is input into the headless item information matching model to obtain the headless item attribute information corresponding to the headless item to be matched.

[0005] In an exemplary embodiment, obtaining sample logistics data of a sample headerless package in a logistics platform includes: Obtaining initial logistics data of the headerless logistics package in the logistics platform; Performing data cleaning on the initial logistics data to obtain sample logistics data of the headerless logistics package; By analyzing the logistics process, set the preset time corresponding to suspected headless items; A headless logistics package whose cumulative unscanned time is greater than or equal to the preset time and less than the alert time is marked as a suspected headless package; Mark the headless logistics package whose accumulated unscanned time is greater than or equal to the alert time as a headless package to be matched; Based on the suspected headless piece and the headless piece to be matched, the sample headless piece package is constructed, and sample logistics data of the sample headless piece package is obtained. In an exemplary embodiment, the method further comprises: Determine the first headless file whose cumulative unscanned time reaches a preset time as a suspected headless file; The headless item whose second cumulative unscanned time in the suspected headless item reaches the warning time is determined as the headless item to be matched; the second cumulative unscanned time is the same as the start time corresponding to the first cumulative unscanned time, and the end time of the second cumulative unscanned time is later than the end time of the first cumulative unscanned time.

[0006] In an exemplary embodiment, the training of the machine learning model based on the difference between the sample headless component attribute information and the sample headless component attribute label to obtain the headless component information matching model includes: Based on the difference between the attribute information of the sample headless component and the attribute label of the sample headless component, the machine learning model is trained to obtain an initial matching model; Using a cross-validation method to determine the performance index of the initial matching model; In the case where the performance index meets the preset conditions, determining the initial matching model as the headless information matching model; When the performance indicator does not meet the preset conditions, adjusting the model parameters of the initial matching model to obtain a current matching model; A cross-validation method is used to determine the current performance index of the current matching model, and based on the current performance index, the model parameters of the current matching model are readjusted until the current performance index of the current matching model meets the preset conditions, thereby obtaining the headless information matching model.

[0007] In an exemplary embodiment, the method further comprises: Obtaining the headless piece images of each headless piece, and performing image cropping, scaling, and grayscale processing on the headless piece images to obtain processed images; Creating a SIFT feature extractor and using a preset function to extract key points and descriptors from the processed image; Storing key points and descriptors corresponding to each headless component in the database; The acquiring of the logistics data of the headless item to be matched, inputting the logistics data into the headless item information matching model, and obtaining the headless item attribute information corresponding to the headless item to be matched includes: Obtaining a target headless piece image of a headless piece to be matched, and performing image cropping, scaling, and grayscale processing on the target headless piece image to obtain a target processed image; Creating a SIFT feature extractor and using a preset function to extract target key points and target descriptors in the processed image of the target; Input the target key points and target descriptors of the headless item to be matched and the key points and descriptors corresponding to each headless item in the database into the headless item information matching model for matching analysis to obtain a matching result; Using a ratio test method to verify the accuracy of the matching result; When the accuracy of the matching result meets the preset matching condition, the headless component attribute information corresponding to the headless component to be matched is determined according to the matching result.

[0008] In an exemplary embodiment, the using a ratio test method to verify the accuracy of the matching result includes: Obtain the nearest neighbor headless component and the next nearest neighbor headless component corresponding to the headless component to be matched in the matching results; According to the key points and the descriptors, a ratio test method is used to calculate a first distance between the headless piece to be matched and the nearest neighbor headless piece and a second distance between the headless piece to be matched and the next nearest neighbor headless piece; calculating a distance ratio between the first distance and the second distance; When the distance ratio is less than a preset threshold, the nearest neighbor headless piece is determined to be a headless piece that matches the headless piece to be matched, and it is determined that the accuracy of the matching result meets the preset matching condition.

[0009] In an exemplary embodiment, the method includes: Obtaining the sample package size, sample package weight, sample shipping place, sample destination, sample transportation route and sample transportation time of the sample headerless package; Setting the weights of the sample package size and the sample package weight as the first weight; The weights of the sample shipping place and the sample destination are set to a second weight; the second weight is greater than the first weight; The weights of the sample transportation route and the sample transportation time are set as a third weight; the third weight is greater than the first weight; The step of searching the database for headless item attribute information matching the sample headless item package based on the sample package characteristics and the sample transportation information, and outputting the sample headless item attribute information includes: Determine the comprehensive characteristics of the sample based on a first product of the sample package characteristics and the first weight, a second product of the sample sending and receiving location and the second weight, and a third product of the sample transportation information and the third weight; Based on the comprehensive characteristics of the sample, the headless item attribute information matching the sample headless item package is searched from the database, and the sample headless item attribute information is output.

[0010] On the other hand, a headless piece wrapping matching device is provided, the device comprising: A sample logistics data acquisition module is used to acquire sample logistics data of a sample headerless package in a logistics platform; the sample headerless package is marked with a sample headerless attribute label; A sample feature extraction module, used to input the sample logistics data of the sample headerless package into a machine learning model to extract package features and transportation information, and obtain sample package features and sample transportation information; The headless component attribute acquisition module is used to obtain the headless component attribute information of each headless component in the database; A sample attribute output module, used to search the database for headless item attribute information matching the sample headless item package based on the sample package characteristics and sample transportation information, and output the sample headless item attribute information; A model training module, used for training the machine learning model based on the difference between the sample headless component attribute information and the sample headless component attribute label to obtain a headless component information matching model; The attribute determination module is used to obtain the logistics data of the headless piece to be matched, input the logistics data into the headless piece information matching model, and obtain the headless piece attribute information corresponding to the headless piece to be matched.

[0011] In an exemplary embodiment, the sample logistics data acquisition module includes: An initial logistics data acquisition unit, used to acquire initial logistics data of the headerless logistics package in the logistics platform; A sample logistics data acquisition unit, used for performing data cleaning on the initial logistics data to obtain sample logistics data of the headerless logistics package; A preset time setting unit, used to set the preset time corresponding to suspected headerless items by analyzing the logistics process; The first marking unit is used to mark a headless logistics package whose cumulative unscanned time is greater than or equal to the preset time and less than the warning time as a suspected headless package; The second marking unit is used to mark the headless logistics package whose cumulative unscanned time is greater than or equal to the alert time as a headless package to be matched; A sample logistics data acquisition unit is used to construct the sample headless piece package based on the suspected headless piece and the headless piece to be matched, and to acquire the sample logistics data of the sample headless piece package. In an exemplary embodiment, the apparatus further comprises: A suspected headless file determining module, used for determining a headless file whose first cumulative unscanned time reaches a preset time as a suspected headless file; The module for determining headless items to be matched is used to determine the headless items whose second cumulative unscanned time reaches the warning time among the suspected headless items as headless items to be matched; the second cumulative unscanned time is the same as the start time corresponding to the first cumulative unscanned time, and the end time of the second cumulative unscanned time is later than the end time of the first cumulative unscanned time.

[0012] In an exemplary embodiment, the model training module includes: An initial matching model determining unit, configured to train the machine learning model based on the difference between the sample headless component attribute information and the sample headless component attribute label to obtain an initial matching model; A performance indicator determination unit, used to determine the performance indicator of the initial matching model using a cross-validation device; A first determining unit, configured to determine the initial matching model as the headless information matching model when the performance indicator satisfies a preset condition; a current matching model determining unit, configured to adjust the model parameters of the initial matching model to obtain a current matching model when the performance indicator does not meet a preset condition; The second determination unit is used to use a cross-validation device to determine the current performance index of the current matching model, and readjust the model parameters of the current matching model based on the current performance index until the current performance index of the current matching model meets the preset conditions, thereby obtaining the headless information matching model.

[0013] In an exemplary embodiment, the apparatus further comprises: An image processing module is used to obtain the headless piece image of each headless piece, and perform image cropping, scaling, and grayscale processing on the headless piece image to obtain a processed image; A key point extraction module, used to create a SIFT feature extractor and extract key points and descriptors from the processed image using a preset function; A storage module, used for storing key points and descriptors corresponding to each headless component in the database; The attribute determination module comprises: A target processed image determining unit is used to obtain a target headless piece image of the headless piece to be matched, and perform image cropping, scaling, and grayscale processing on the target headless piece image to obtain a target processed image; A target key point extraction unit, used to create a SIFT feature extractor and extract target key points and target descriptors in the target processed image using a preset function; A matching result determination unit, used for inputting the target key points and target descriptors of the headless item to be matched and the key points and descriptors corresponding to each headless item in the database into the headless item information matching model for matching analysis to obtain a matching result; an accuracy analysis unit, used to verify the accuracy of the matching result using a ratio testing device; The headless component attribute information determining unit is used to determine the headless component attribute information corresponding to the headless component to be matched according to the matching result when the accuracy of the matching result meets the preset matching condition.

[0014] In an exemplary embodiment, the accuracy analysis unit includes: An adjacent headless component acquisition subunit is used to acquire the nearest neighbor headless component and the next nearest neighbor headless component corresponding to the headless component to be matched in the matching result; A distance determination subunit, for calculating, based on the key points and the descriptors, a first distance between the to-be-matched headless piece and the nearest neighbor headless piece and a second distance between the to-be-matched headless piece and the next nearest neighbor headless piece using a ratio test device; a distance ratio determination subunit, configured to calculate a distance ratio between the first distance and the second distance; The matching subunit is used to determine that the nearest neighbor headless piece is a headless piece that matches the headless piece to be matched when the distance ratio is less than a preset threshold, and to determine that the accuracy of the matching result meets the preset matching condition.

[0015] In an exemplary embodiment, the apparatus comprises: An attribute acquisition module, used to obtain the sample package size, sample package weight, sample shipping place, sample destination, sample transportation route and sample transportation time of the sample headerless package; A first weight determination module, used to set the weight of the sample package size and the sample package weight as a first weight; A second weight determination module, used to set the weights of the sample shipping place and the sample destination as second weights; the second weight is greater than the first weight; A third weight determination module, used to set the weights of the sample transportation route and the sample transportation time as a third weight; the third weight is greater than the first weight; The sample attribute output module includes: A sample comprehensive feature determination unit, configured to determine the sample comprehensive feature based on a first product of the sample package feature and the first weight, a second product of the sample sending and receiving location and the second weight, and a third product of the sample transportation information and the third weight; The sample headless item attribute information output unit is used to search the database for headless item attribute information matching the sample headless item package based on the sample comprehensive features, and output the sample headless item attribute information.

[0016] On the other hand, an electronic device is provided, comprising a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the headless package matching method as described above.

[0017] On the other hand, a computer storage medium is provided, wherein the computer storage medium stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by a processor to implement the headerless package matching method as described above.

[0018] On the other hand, a computer program product or a computer program is provided, the computer program product or the computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes to implement the above-mentioned headerless package matching method.

[0019] The headless package matching method, device and equipment provided by the present invention have the following technical effects: The present invention obtains sample logistics data of sample headless packages in a logistics platform; the sample headless packages are marked with sample headless attribute labels; the sample logistics data of the sample headless packages are input into a machine learning model to extract package features and transportation information, and obtain sample package features and sample transportation information; the headless attribute information of each headless package in the database is obtained; based on the sample package features and sample transportation information, the headless attribute information matching the sample headless package is searched from the database, and the sample headless attribute information is output; based on the difference between the sample headless attribute information and the sample headless attribute label, the machine learning model is trained to obtain a headless information matching model; the logistics data of the headless package to be matched is obtained, and the logistics data is input into the headless information matching model to obtain the headless attribute information corresponding to the headless package to be matched. The present invention can quickly identify and process headless packages through automated anomaly detection and intelligent matching algorithms, reduce manual intervention, and improve overall logistics efficiency; reduce logistics losses caused by headless packages, and improve the delivery rate of packages; and improve customer service experience through real-time monitoring and rapid response. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions and advantages of the embodiments of this specification or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0021] Figure 1 is a schematic diagram of a headless package matching system provided in an embodiment of this specification; Figure 2 It is a flowchart of a method for matching a package without a header provided in an embodiment of this specification; Figure 3 It is a flowchart of a method for obtaining sample logistics data of a sample headerless package in a logistics platform provided by an embodiment of this specification; Figure 4 It is a flowchart of a method for training the machine learning model based on the difference between the sample headless component attribute information and the sample headless component attribute label to obtain a headless component information matching model provided by an embodiment of this specification; Figure 5 is a flow chart of a method for verifying the accuracy of the matching result using a ratio test method provided in an embodiment of this specification; Figure 6 It is a structural schematic diagram of a headless package matching device provided in an embodiment of this specification; Figure 7 It is a structural diagram of a server provided in an embodiment of this specification. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of this specification to clearly and completely describe the technical solutions in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0023] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products, or devices.

[0024] See also Figure 1 , Figure 1 is a schematic diagram of a headless package matching system provided in an embodiment of this specification, such as Figure 1 As shown, the headless package matching system may include at least a server 01 and a client 02 .

[0025] Specifically, in the embodiments of this specification, the server 01 may include an independently operated server, or a distributed server, or a server cluster composed of multiple servers, and may also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. Server 01 may include a network communication unit, a processor, a memory, and the like. Specifically, the server 01 can be used to train a headless information matching model.

[0026] Specifically, in the embodiments of this specification, the client 02 may include physical devices such as smart phones, desktop computers, tablet computers, laptop computers, digital assistants, smart wearable devices, smart speakers, vehicle terminals, smart TVs, etc., and may also include software running in physical devices, such as web pages provided by some service providers to users, or applications provided by these service providers to users. Specifically, the client 02 may be used to query the headless component attribute information corresponding to the headless component to be matched based on the headless component information matching model.

[0027] The following describes a headless package matching method of the present invention. Figure 2 It is a flowchart of a method for headerless package matching provided in an embodiment of this specification. This specification provides method operation steps as described in the embodiment or flowchart, but may include more or fewer operation steps based on conventional or non-creative labor. The order of steps listed in the embodiment is only one way of executing the steps among many orders, and does not represent the only order of execution. When the actual system or server product is executed, it can be executed in sequence or in parallel (for example, in a parallel processor or multi-threaded processing environment) according to the method shown in the embodiment or the drawings. Specifically, Figure 2 As shown, the method may include: S201: Obtain sample logistics data of a sample headerless package in a logistics platform; the sample headerless package is marked with a sample headerless attribute label. In the embodiments of this specification, all data of suspected headless items can be collected and cleaned to ensure accuracy and consistency. Sample headless item packages are marked with sample headless item attribute tags, which are attribute data missing from the sample headless item packages. The attribute data may include information such as the recipient address, the sender address, and contact information.

[0028] S203: Input the sample logistics data of the sample headerless package into the machine learning model to extract package features and transportation information, and obtain sample package features and sample transportation information.

[0029] In the embodiments of the present specification, the sample logistics data may include the package attributes and transportation information of the sample headerless package, etc. The sample logistics data may be input into the machine learning model to extract package features and transportation information to obtain sample package features and sample transportation information. The algorithm corresponding to the machine learning model may include but is not limited to algorithms such as random forest or neural network.

[0030] S205: Obtain the headless component attribute information of each headless component in the database.

[0031] In the embodiments of this specification, in order to effectively manage and monitor suspected headless items in the logistics process, it is necessary to establish a centralized database system. This system will be responsible for collecting and integrating relevant data information of all suspected headless items. Specifically, these data include but are not limited to the timestamp when the package was last scanned, the specific information of each logistics node passed, the weight of the package, the sender's detailed contact information, the specific location of the collection point, the relevant data of the transit point, and the weight of the express package and the classification of the contents and other detailed information. In order to ensure that this database can reflect the latest logistics status and changes in real time, it is necessary to ensure that the data pool can be continuously updated in real time, so that the latest developments of each package can be grasped in time, and the timeliness and accuracy of the data can be ensured; after collecting a large amount of data, data cleaning is particularly important. By cleaning the data, duplicate, erroneous or inconsistent information can be removed, thereby ensuring the accuracy and consistency of the data. This step is crucial for subsequent data analysis and decision-making; through in-depth analysis of the logistics process of express parcels, a series of preset logistics processes and target logistics durations are set. These preset processes and durations will serve as important reference standards for evaluating whether the package is transported normally. If a package is not scanned within the expected time frame, the system will automatically mark it as a suspected headless piece. Such a mark will attract the attention of relevant departments so that further investigation and handling measures can be taken in time; a preset warning duration is set. If a suspected headless piece has not been scanned or updated within this warning duration, the system can define it as a headless piece to be matched, activate a higher-level alarm mechanism, and notify relevant personnel to conduct more in-depth investigation and matching work to find the whereabouts of the package as soon as possible and ensure the smooth progress of the logistics process.

[0032] S207: Based on the sample package characteristics and sample transportation information, search the database for headless item attribute information that matches the sample headless item package, and output the sample headless item attribute information.

[0033] In the embodiments of this specification, a machine learning algorithm such as a random forest or a neural network is used to classify and match suspected headerless items, analyze package features and transportation information, and compare them with headerless item information in a database to find the most likely match.

[0034] S209: Based on the difference between the sample headless component attribute information and the sample headless component attribute label, the machine learning model is trained to obtain a headless component information matching model.

[0035] In the embodiment of the present specification, according to the difference between the sample headless item attribute information and the sample headless item attribute label, the target loss information is constructed, and the model parameters of the machine learning model are adjusted according to the target loss information until the training end condition is met to obtain the headless item information matching model. The training end condition may be that the target loss information is less than the preset data or the number of training iterations reaches the target number.

[0036] You can choose a suitable machine learning algorithm, such as random forest or neural network. If you choose random forest, you can use the RandomForestClassifier class in Python's scikit-learn library to build the model. If you choose neural network, you can use deep learning frameworks such as TensorFlow or PyTorch to build the model, use the preprocessed suspected headless data as input, and use the headless information in the database as output to train the model. During the training process, you can use methods such as cross-validation to evaluate the performance of the model, and adjust the model parameters to improve the accuracy and generalization ability of the model, and use the trained machine learning model to classify and match new suspected headless items.

[0037] S2011: Obtain the logistics data of the headless item to be matched, input the logistics data into the headless item information matching model, and obtain the headless item attribute information corresponding to the headless item to be matched.

[0038] In an embodiment of the present specification, the package characteristics and transportation information of a new suspected headless item are input into the model. The model will output the matching probability between the suspected headless item and each headless item in the database, and find the most likely match based on the matching probability; thereby determining the headless item attribute information corresponding to the headless item to be matched.

[0039] In the embodiments of this specification, Figure 3 As shown, the method of obtaining sample logistics data of a sample headerless package in a logistics platform includes: S20101: Acquire initial logistics data of the headerless logistics package in the logistics platform; S20103: Cleaning the initial logistics data to obtain sample logistics data of the headerless logistics package; S20105: Set the preset duration for suspected headless items by analyzing the logistics process; S20107: Mark the headless logistics package whose cumulative unscanned time is greater than or equal to the preset time and less than the warning time as a suspected headless logistics package; mark the headless logistics package whose cumulative unscanned time is greater than or equal to the warning time as a headless logistics package to be matched; S20109: Based on the suspected headless piece and the headless piece to be matched, construct the sample headless piece package, and obtain sample logistics data of the sample headless piece package.

[0040] In the embodiment of this specification, data of all suspected headless items are collected, and data cleaning is performed to ensure accuracy and consistency. By analyzing the logistics process, a preset time is set, and packages that are not scanned on time are marked as suspected headless items. If they are not scanned for more than the warning time, they are defined as headless items to be matched. Through in-depth analysis of the logistics process of express parcels, a series of preset logistics processes and target logistics time are set. These preset processes and time will serve as important reference standards for evaluating whether the parcel is transported normally. If a parcel is not scanned within the expected time range, the system will automatically mark it as a suspected headless item. Such a mark will attract the attention of relevant departments so that further investigation and handling measures can be taken in time; a preset warning time is set. If a suspected headless item is still not scanned or updated within this warning time, the system can define it as a headless item to be matched, activate a higher-level alarm mechanism, and notify relevant personnel to conduct more in-depth investigation and matching work to find the whereabouts of the parcel as soon as possible to ensure the smooth progress of the logistics process.

[0041] In the embodiments of this specification, the method includes: Obtaining the sample package size, sample package weight, sample shipping place, sample destination, sample transportation route and sample transportation time of the sample headerless package; Setting the weights of the sample package size and the sample package weight as the first weight; The weights of the sample shipping place and the sample destination are set to a second weight; the second weight is greater than the first weight; The weights of the sample transportation route and the sample transportation time are set as a third weight; the third weight is greater than the first weight; The step of searching the database for headless item attribute information matching the sample headless item package based on the sample package characteristics and the sample transportation information, and outputting the sample headless item attribute information includes: Determine the comprehensive characteristics of the sample based on a first product of the sample package characteristics and the first weight, a second product of the sample sending and receiving location and the second weight, and a third product of the sample transportation information and the third weight; Based on the comprehensive characteristics of the sample, the headless item attribute information matching the sample headless item package is searched from the database, and the sample headless item attribute information is output.

[0042] In the embodiment of the present specification, the sample package characteristics include the characteristics corresponding to the sample package size and the sample package weight; the sample delivery location includes the characteristics of the sample delivery location and the sample destination; the sample transportation information includes the characteristics of the sample transportation route and the sample transportation time; In the embodiments of this specification, the second weight can be set to be greater than the third weight, or less than the third weight; preset weights are set for different express data items, such as package size, weight, shape, color, place of shipment, destination, transportation route, transportation time, etc. It can be considered that the size and weight of the package are less important for matching, while the place of shipment and destination are more important for matching, so higher weights can be set for the place of shipment and destination; according to the set weights and the values ​​of the data items, the score of each suspected headless item is calculated, and the score calculation can use the weighted summation method, that is, the value of each data item is multiplied by the corresponding weight, and then the products of all data items are added to obtain the total score. For example, if the shipping location of a suspected headless item is the same as that of a headless item, and the weight of the shipping location is 0.5, then the score of the suspected headless item on the shipping location data item is 0.5; sort the scores of all suspected headless items to determine the most likely match. The higher the score of the suspected headless item, the greater the possibility of matching with the corresponding headless item; establish a real-time monitoring system to track the logistics status of express parcels in real time, obtain information such as the location, transportation time, and transportation route of the parcel. If it is found that the parcel is stagnant in a certain link, timely intervention can be made to reduce the generation of headless items; let users participate in the matching process. When the system determines the possible matching items, it can let users confirm whether they are correct. Users can provide feedback on the matching results through the express company's official website, customer service hotline, mobile application and other channels; based on the matching results and user feedback, continuously adjust and optimize the matching algorithm and weight settings. If the user feedback that a matching item is incorrect, analyze the cause and adjust the matching algorithm and weight settings to improve the accuracy of the match.

[0043] In the embodiment of this specification, the method further includes: Determine the first headless file whose cumulative unscanned time reaches a preset time as a suspected headless file; The headless item whose second cumulative unscanned time in the suspected headless item reaches the warning time is determined as the headless item to be matched; the second cumulative unscanned time is the same as the start time corresponding to the first cumulative unscanned time, and the end time of the second cumulative unscanned time is later than the end time of the first cumulative unscanned time.

[0044] In the embodiments of this specification, Figure 4As shown, the machine learning model is trained based on the difference between the sample headless component attribute information and the sample headless component attribute label to obtain a headless component information matching model, including: S2091: Based on the difference between the attribute information of the sample headless component and the attribute label of the sample headless component, the machine learning model is trained to obtain an initial matching model; S2093: Determine the performance index of the initial matching model using a cross-validation method; S2095: When the performance indicator meets a preset condition, the initial matching model is determined as the headless information matching model; S2097: If the performance indicator does not meet the preset condition, adjust the model parameters of the initial matching model to obtain a current matching model; S2099: Using a cross-validation method to determine the current performance index of the current matching model, and based on the current performance index, readjusting the model parameters of the current matching model until the current performance index of the current matching model meets a preset condition, thereby obtaining the headless information matching model.

[0045] In the embodiments of this specification, during the training process, methods such as cross-validation can be used to evaluate the performance of the model and adjust the parameters of the model to improve the accuracy and generalization ability of the model.

[0046] In the embodiment of this specification, the method further includes: Obtaining the headless piece images of each headless piece, and performing image cropping, scaling, and grayscale processing on the headless piece images to obtain processed images; Creating a SIFT feature extractor and using a preset function to extract key points and descriptors from the processed image; Storing key points and descriptors corresponding to each headless component in the database; The acquiring of the logistics data of the headless item to be matched, inputting the logistics data into the headless item information matching model, and obtaining the headless item attribute information corresponding to the headless item to be matched includes: Obtaining a target headless piece image of a headless piece to be matched, and performing image cropping, scaling, and grayscale processing on the target headless piece image to obtain a target processed image; Creating a SIFT feature extractor and using a preset function to extract target key points and target descriptors in the processed image of the target; Input the target key points and target descriptors of the headless item to be matched and the key points and descriptors corresponding to each headless item in the database into the headless item information matching model for matching analysis to obtain a matching result; Using a ratio test method to verify the accuracy of the matching result; When the accuracy of the matching result meets the preset matching condition, the headless component attribute information corresponding to the headless component to be matched is determined according to the matching result.

[0047] In the embodiments of the present specification, the SIFT algorithm is used to extract key points and descriptors in the processed image; the key points and descriptors in the image are extracted using image processing techniques such as the SIFT algorithm, the SIFT_create function in the OpenCV library is used to create a SIFT feature extractor, and then the detectAndCompute function is used to extract key points and descriptors in the image, and the extracted key points and descriptors are stored and managed; the image feature matching is performed using algorithms such as the FLANN matcher, the FlannBasedMatcher function in the OpenCV library is used to create a FLANN matcher, and then the match function is used to perform feature matching, and the matching results are stored and managed for subsequent verification and analysis; the matching accuracy is verified using methods such as ratio testing to eliminate false matches.

[0048] In the embodiments of this specification, Figure 5 As shown, the accuracy of the matching result is verified by using a ratio test method, including: S501: Obtain the nearest neighbor headless component and the next nearest neighbor headless component corresponding to the headless component to be matched in the matching results; S503: Calculate, based on the key points and the descriptors, a first distance between the to-be-matched headless piece and the nearest neighbor headless piece and a second distance between the to-be-matched headless piece and the next nearest neighbor headless piece using a ratio test method; S505: Calculate a distance ratio between the first distance and the second distance; S507: When the distance ratio is less than a preset threshold, determine that the nearest neighbor headless piece is a headless piece that matches the headless piece to be matched, and determine that the accuracy of the matching result meets the preset matching condition.

[0049] In the embodiments of this specification, the basic idea of ​​the ratio test is to compare the distance ratio of the nearest neighbor and the next nearest neighbor. If the distance ratio is less than a certain threshold, the match is considered correct; otherwise, the match is considered wrong. The ratioTest function in the OpenCV library can be used to perform the ratio test.

[0050] In an exemplary embodiment, the technical solution of the present invention is as follows: Step 1: Collect all data of suspected headerless items, clean the data to ensure accuracy and consistency, set a preset time by analyzing the logistics process, and mark the packages that are not scanned on time as suspected headerless items. If the packages are not scanned after the warning time, they are defined as headerless items to be matched; In order to effectively manage and monitor suspected headless items in the logistics process, it is necessary to establish a centralized database system. This system will be responsible for collecting and integrating relevant data information of all suspected headless items. Specifically, these data include but are not limited to the timestamp when the package was last scanned, the specific information of each logistics node passed, the weight of the package, the sender's detailed contact information, the specific location of the collection point, the relevant data of the transit point, and the weight of the express package and the classification of the contents. In order to ensure that this database can reflect the latest logistics status and changes in real time, it is necessary to ensure that the data pool can be continuously updated in real time, so that the latest developments of each package can be grasped in time, and the timeliness and accuracy of the data can be ensured; after collecting a large amount of data, data cleaning is particularly important. By cleaning the data, duplicate, erroneous or inconsistent information can be removed, thereby ensuring the accuracy and consistency of the data. This step is crucial for subsequent data analysis and decision-making; through in-depth analysis of the logistics process of express parcels, a series of preset logistics processes and target logistics durations are set. These preset processes and durations will serve as important reference standards for evaluating whether the package is transported normally. If a package is not scanned within the expected time frame, the system will automatically mark it as a suspected headless piece. Such a mark will attract the attention of relevant departments so that further investigation and handling measures can be taken in time; a preset warning duration is set. If a suspected headless piece has not been scanned or updated within this warning duration, the system can define it as a headless piece to be matched, activate a higher-level alarm mechanism, and notify relevant personnel to conduct more in-depth investigation and matching work to find the whereabouts of the package as soon as possible and ensure the smooth progress of the logistics process.

[0051] Step 2: Use machine learning algorithms such as random forest or neural network to classify and match suspected headerless items, analyze package features and transportation information, compare with headerless item information in the database, and find the most likely match; Select a suitable machine learning algorithm, such as random forest or neural network. If you choose random forest, you can use the RandomForestClassifier class in Python's scikit-learn library to build the model. If you choose neural network, you can use deep learning frameworks such as TensorFlow or PyTorch to build the model. Take the preprocessed suspected headless data as input and the headless information in the database as output to train the model. During the training process, you can use methods such as cross-validation to evaluate the performance of the model and adjust the parameters of the model to improve the accuracy and generalization ability of the model. Use the trained machine learning model to classify and match new suspected headless items. Input the package features and transportation information of the new suspected headless items into the model. The model will output the matching probability between the suspected headless item and each headless item in the database. According to the matching probability, find the most likely match. You can set a threshold. When the matching probability exceeds the threshold, it is considered that the suspected headless item is successfully matched with the corresponding headless item.

[0052] Step 3: Collect the published and user-uploaded headless pictures and the pictures taken during the operation as image data, use image processing techniques such as SIFT algorithm to extract image key points and descriptors, use algorithms such as FLANN matcher to perform image feature matching, use ratio test and other methods to verify the matching accuracy and eliminate false matches; Collect the headless pictures published and uploaded by users, as well as the pictures taken during the express delivery operation. Preprocess the collected pictures, including image cropping, scaling, grayscale and other operations, to facilitate the subsequent image processing algorithm processing. Use image processing techniques such as SIFT algorithm to extract key points and descriptors in the picture. Use the SIFT_create function in the OpenCV library to create a SIFT feature extractor, and then use the detectAndCompute function to extract key points and descriptors in the picture, and store and manage the extracted key points and descriptors; use algorithms such as FLANN matcher to match image features, use the FlannBasedMatcher function in the OpenCV library to create a FLANN matcher, and then use the match function to perform feature matching, and store and manage the matching results for subsequent verification and analysis; use methods such as ratio test to verify the matching accuracy and eliminate false matches. The basic idea of ​​the ratio test is to compare the distance ratio between the nearest neighbor and the next nearest neighbor. If the distance ratio is less than a certain threshold, the match is considered correct; otherwise, the match is considered wrong. The ratioTest function in the OpenCV library can be used to perform ratio testing.

[0053] Step 4: By setting weights for express delivery data and calculating scores, a monitoring system is implemented to track the logistics status of packages. Based on matching results and user feedback, the matching algorithm and weight settings are continuously adjusted and optimized.

[0054] Set preset weights for different express data items, such as package size, weight, shape, color, place of shipment, destination, transportation route, transportation time, etc. It can be considered that the size and weight of the package are less important for matching, while the place of shipment and destination are more important for matching, so higher weights can be set for the place of shipment and destination; calculate the score of each suspected headless item based on the set weight and the value of the data item. The score can be calculated using the weighted sum method, that is, multiply the value of each data item by the corresponding weight, and then add the products of all data items to get the total score. For example, if the shipping location of a suspected headless item is the same as that of a headless item, and the weight of the shipping location is 0.5, then the score of the suspected headless item on the shipping location data item is 0.5; sort the scores of all suspected headless items to determine the most likely match. The higher the score of the suspected headless item, the greater the possibility of matching with the corresponding headless item; establish a real-time monitoring system to track the logistics status of express parcels in real time, obtain information such as the location, transportation time, and transportation route of the parcel. If it is found that the parcel is stagnant in a certain link, timely intervention can be made to reduce the generation of headless items; let users participate in the matching process. When the system determines the possible matching items, it can let users confirm whether they are correct. Users can provide feedback on the matching results through the express company's official website, customer service hotline, mobile application and other channels; based on the matching results and user feedback, continuously adjust and optimize the matching algorithm and weight settings. If the user feedback that a matching item is incorrect, analyze the cause and adjust the matching algorithm and weight settings to improve the accuracy of the match.

[0055] It can be seen from the technical solutions provided by the above embodiments of this specification that the embodiments of this specification obtain sample logistics data of sample headless packages in the logistics platform; the sample headless packages are marked with sample headless attribute labels; the sample logistics data of the sample headless packages are input into the machine learning model to extract package features and transportation information, and obtain sample package features and sample transportation information; obtain the headless attribute information of each headless package in the database; based on the sample package features and sample transportation information, search the database for headless attribute information that matches the sample headless package, and output the sample headless attribute information; based on the difference between the sample headless attribute information and the sample headless attribute label, train the machine learning model to obtain a headless information matching model; obtain the logistics data of the headless package to be matched, input the logistics data into the headless information matching model, and obtain the headless attribute information corresponding to the headless package to be matched. Through automated anomaly detection and intelligent matching algorithms, the present invention can quickly identify and process headless items, reduce manual intervention, and improve overall logistics efficiency; reduce logistics losses caused by headless items and improve the delivery rate of packages; and enhance customer service experience through real-time monitoring and rapid response.

[0056] The embodiment of this specification also provides a headless piece wrapping matching device, such as Figure 6 As shown, the device comprises: The sample logistics data acquisition module 610 is used to acquire the sample logistics data of the sample headerless package in the logistics platform; the sample headerless package is marked with a sample headerless attribute label; A sample feature extraction module 620 is used to input the sample logistics data of the sample headerless package into a machine learning model to extract package features and transportation information, and obtain sample package features and sample transportation information; The headless component attribute acquisition module 630 is used to acquire the headless component attribute information of each headless component in the database; A sample attribute output module 640 is used to search the database for headless item attribute information matching the sample headless item package based on the sample package characteristics and the sample transportation information, and output the sample headless item attribute information; A model training module 650, configured to train the machine learning model based on the difference between the sample headless component attribute information and the sample headless component attribute label to obtain a headless component information matching model; The attribute determination module 660 is used to obtain the logistics data of the headless piece to be matched, input the logistics data into the headless piece information matching model, and obtain the headless piece attribute information corresponding to the headless piece to be matched.

[0057] In an exemplary embodiment, the sample logistics data acquisition module includes: An initial logistics data acquisition unit, used to acquire initial logistics data of the headerless logistics package in the logistics platform; A sample logistics data acquisition unit, used for performing data cleaning on the initial logistics data to obtain sample logistics data of the headerless logistics package; A preset time setting unit, used to set the preset time corresponding to suspected headerless items by analyzing the logistics process; The first marking unit is used to mark a headless logistics package whose cumulative unscanned time is greater than or equal to the preset time and less than the warning time as a suspected headless package; The second marking unit is used to mark the headless logistics package whose cumulative unscanned time is greater than or equal to the alert time as a headless package to be matched; A sample logistics data acquisition unit is used to construct the sample headless piece package based on the suspected headless piece and the headless piece to be matched, and to acquire the sample logistics data of the sample headless piece package. In an exemplary embodiment, the apparatus further comprises: A suspected headless file determining module, used for determining a headless file whose first cumulative unscanned time reaches a preset time as a suspected headless file; The module for determining headless items to be matched is used to determine the headless items whose second cumulative unscanned time reaches the warning time among the suspected headless items as headless items to be matched; the second cumulative unscanned time is the same as the start time corresponding to the first cumulative unscanned time, and the end time of the second cumulative unscanned time is later than the end time of the first cumulative unscanned time.

[0058] In an exemplary embodiment, the model training module includes: An initial matching model determining unit, configured to train the machine learning model based on the difference between the sample headless component attribute information and the sample headless component attribute label to obtain an initial matching model; A performance indicator determination unit, used to determine the performance indicator of the initial matching model using a cross-validation device; A first determining unit, configured to determine the initial matching model as the headless information matching model when the performance indicator satisfies a preset condition; a current matching model determining unit, configured to adjust the model parameters of the initial matching model to obtain a current matching model when the performance indicator does not meet a preset condition; The second determination unit is used to use a cross-validation device to determine the current performance index of the current matching model, and readjust the model parameters of the current matching model based on the current performance index until the current performance index of the current matching model meets the preset conditions, thereby obtaining the headless information matching model.

[0059] In an exemplary embodiment, the apparatus further comprises: An image processing module is used to obtain the headless piece image of each headless piece, and perform image cropping, scaling, and grayscale processing on the headless piece image to obtain a processed image; A key point extraction module, used to create a SIFT feature extractor and extract key points and descriptors from the processed image using a preset function; A storage module, used for storing key points and descriptors corresponding to each headless component in the database; The attribute determination module comprises: A target processed image determining unit is used to obtain a target headless piece image of the headless piece to be matched, and perform image cropping, scaling, and grayscale processing on the target headless piece image to obtain a target processed image; A target key point extraction unit, used to create a SIFT feature extractor and extract target key points and target descriptors in the target processed image using a preset function; A matching result determination unit, used for inputting the target key points and target descriptors of the headless item to be matched and the key points and descriptors corresponding to each headless item in the database into the headless item information matching model for matching analysis to obtain a matching result; an accuracy analysis unit, used to verify the accuracy of the matching result using a ratio testing device; The headless component attribute information determining unit is used to determine the headless component attribute information corresponding to the headless component to be matched according to the matching result when the accuracy of the matching result meets the preset matching condition.

[0060] In an exemplary embodiment, the accuracy analysis unit includes: An adjacent headless component acquisition subunit is used to acquire the nearest neighbor headless component and the next nearest neighbor headless component corresponding to the headless component to be matched in the matching result; A distance determination subunit, for calculating, based on the key points and the descriptors, a first distance between the to-be-matched headless piece and the nearest neighbor headless piece and a second distance between the to-be-matched headless piece and the next nearest neighbor headless piece using a ratio test device; a distance ratio determination subunit, configured to calculate a distance ratio between the first distance and the second distance; The matching subunit is used to determine that the nearest neighbor headless piece is a headless piece that matches the headless piece to be matched when the distance ratio is less than a preset threshold, and to determine that the accuracy of the matching result meets the preset matching condition.

[0061] In an exemplary embodiment, the apparatus comprises: An attribute acquisition module, used to obtain the sample package size, sample package weight, sample shipping place, sample destination, sample transportation route and sample transportation time of the sample headerless package; A first weight determination module, used to set the weight of the sample package size and the sample package weight as a first weight; A second weight determination module, used to set the weights of the sample shipping place and the sample destination as second weights; the second weight is greater than the first weight; A third weight determination module, used to set the weights of the sample transportation route and the sample transportation time as a third weight; the third weight is greater than the first weight; The sample attribute output module includes: A sample comprehensive feature determination unit, configured to determine the sample comprehensive feature based on a first product of the sample package feature and the first weight, a second product of the sample sending and receiving location and the second weight, and a third product of the sample transportation information and the third weight; The sample headless item attribute information output unit is used to search the database for headless item attribute information matching the sample headless item package based on the sample comprehensive features, and output the sample headless item attribute information.

[0062] The device and method embodiments in the described device embodiments are based on the same inventive concept.

[0063] An embodiment of the present specification provides an electronic device, which includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by the processor to implement the headless package matching method provided in the above method embodiment.

[0064] An embodiment of the present invention also provides a computer storage medium, which can be set in a terminal to store at least one instruction or at least one program related to a headless package matching method in a method embodiment. The at least one instruction or at least one program is loaded and executed by the processor to implement the headless package matching method provided in the above method embodiment.

[0065] The embodiment of the present invention also provides a computer program product or a computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes to implement the headless package matching method provided in the above method embodiment.

[0066] Optionally, in the embodiments of this specification, the storage medium may be located in at least one of the multiple network servers of the computer network. Optionally, in this embodiment, the storage medium may include but is not limited to: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store program codes.

[0067] The memory described in the embodiments of this specification can be used to store software programs and modules, and the processor executes various functional applications and data processing by running the software programs and modules stored in the memory. The memory may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, application programs required for functions, etc.; the data storage area may store data created according to the use of the device, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage device. Accordingly, the memory may also include a memory controller to provide the processor with access to the memory.

[0068] The embodiment of the headless package matching method provided in the embodiments of this specification can be executed in a mobile terminal, a computer terminal, a server or a similar computing device. Taking running on a server as an example, Figure 7 1 is a hardware structure block diagram of a server of a headerless package matching method provided in an embodiment of this specification. Figure 7As shown, the server 700 may have relatively large differences due to different configurations or performances, and may include one or more central processing units (CPU) 710 (the central processing unit 710 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 730 for storing data, and one or more storage media 720 (such as one or more mass storage devices) for storing application programs 723 or data 722. Among them, the memory 730 and the storage medium 720 can be short-term storage or permanent storage. The program stored in the storage medium 720 may include one or more modules, each of which may include a series of instruction operations on the server. Furthermore, the central processing unit 710 can be configured to communicate with the storage medium 720 and execute a series of instruction operations in the storage medium 720 on the server 700. The server 700 may also include one or more power supplies 760, one or more wired or wireless network interfaces 750, one or more input and output interfaces 740, and / or one or more operating systems 721, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, etc.

[0069] The input / output interface 740 may be used to receive or send data via a network. A specific example of the network may include a wireless network provided by a communication provider of the server 700. In one example, the input / output interface 740 includes a network adapter (Network Interface Controller, NIC), which may be connected to other network devices via a base station so as to communicate with the Internet. In one example, the input / output interface 740 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0070] It can be understood by those skilled in the art that Figure 7 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 7 More or fewer components as shown, or with Figure 7 Different configurations are shown.

[0071] It can be seen from the embodiments of the headless package matching method, device, electronic device or storage medium provided by the present invention that the present invention obtains sample logistics data of sample headless packages in a logistics platform; the sample headless packages are marked with sample headless attribute labels; the sample logistics data of the sample headless packages are input into a machine learning model to extract package features and transportation information, and obtain sample package features and sample transportation information; the headless attribute information of each headless package in the database is obtained; based on the sample package features and sample transportation information, the headless attribute information matching the sample headless package is searched from the database, and the sample headless attribute information is output; based on the difference between the sample headless attribute information and the sample headless attribute label, the machine learning model is trained to obtain a headless information matching model; the logistics data of the headless package to be matched is obtained, and the logistics data is input into the headless information matching model to obtain the headless attribute information corresponding to the headless package to be matched. Through automated anomaly detection and intelligent matching algorithms, the present invention can quickly identify and process headless items, reduce manual intervention, and improve overall logistics efficiency; reduce logistics losses caused by headless items and improve the delivery rate of packages; and enhance customer service experience through real-time monitoring and rapid response.

[0072] It should be noted that the above sequence of the embodiments of this specification is for description only and does not represent the advantages and disadvantages of the embodiments. The above describes specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0073] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, equipment, and storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0074] A person skilled in the art will appreciate that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the program may be stored in a computer storage medium, and the storage medium mentioned above may be a read-only memory, a disk or an optical disk, etc.

[0075] 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, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for matching headless packages, characterized in that: The method comprises: Obtaining sample logistics data of a sample headerless package in a logistics platform; the sample headerless package is marked with a sample headerless attribute label; Inputting the sample logistics data of the sample headerless package into the machine learning model to extract package features and transportation information, and obtain sample package features and sample transportation information; Get the headless item attribute information of each headless item in the database; Based on the sample package characteristics and the sample transportation information, searching the database for headless item attribute information matching the sample headless item package, and outputting the sample headless item attribute information; Based on the difference between the sample headless item attribute information and the sample headless item attribute label, the machine learning model is trained to obtain a headless item information matching model; The logistics data of the headless item to be matched is obtained, and the logistics data is input into the headless item information matching model to obtain the headless item attribute information corresponding to the headless item to be matched.

2. The method according to claim 1, characterized in that The obtaining of sample logistics data of sample headerless packages in the logistics platform includes: Obtaining initial logistics data of the headerless logistics package in the logistics platform; Performing data cleaning on the initial logistics data to obtain sample logistics data of the headerless logistics package; By analyzing the logistics process, set the preset time corresponding to suspected headless items; A headless logistics package whose cumulative unscanned time is greater than or equal to the preset time and less than the alert time is marked as a suspected headless package; Mark the headless logistics package whose accumulated unscanned time is greater than or equal to the alert time as a headless package to be matched; Based on the suspected headless piece and the headless piece to be matched, the sample headless piece package is constructed, and sample logistics data of the sample headless piece package is obtained.

3. The method according to claim 2, characterized in that The method further comprises: Determine the first headless file whose cumulative unscanned time reaches a preset time as a suspected headless file; The headless item whose second cumulative unscanned time in the suspected headless item reaches the warning time is determined as the headless item to be matched; the second cumulative unscanned time is the same as the start time corresponding to the first cumulative unscanned time, and the end time of the second cumulative unscanned time is later than the end time of the first cumulative unscanned time.

4. The method according to claim 1, characterized in that: The step of training the machine learning model based on the difference between the sample headless component attribute information and the sample headless component attribute label to obtain a headless component information matching model includes: Based on the difference between the attribute information of the sample headless component and the attribute label of the sample headless component, the machine learning model is trained to obtain an initial matching model; Using a cross-validation method to determine the performance index of the initial matching model; In the case where the performance index meets the preset conditions, determining the initial matching model as the headless information matching model; When the performance indicator does not meet the preset conditions, adjusting the model parameters of the initial matching model to obtain a current matching model; A cross-validation method is used to determine the current performance index of the current matching model, and based on the current performance index, the model parameters of the current matching model are readjusted until the current performance index of the current matching model meets the preset conditions, thereby obtaining the headless information matching model.

5. The method according to claim 1, characterized in that The method further comprises: Obtaining the headless piece images of each headless piece, and performing image cropping, scaling, and grayscale processing on the headless piece images to obtain processed images; Creating a SIFT feature extractor and using a preset function to extract key points and descriptors from the processed image; Storing key points and descriptors corresponding to each headless component in the database; The acquiring of the logistics data of the headless item to be matched, inputting the logistics data into the headless item information matching model, and obtaining the headless item attribute information corresponding to the headless item to be matched includes: Obtaining a target headless piece image of a headless piece to be matched, and performing image cropping, scaling, and grayscale processing on the target headless piece image to obtain a target processed image; Creating a SIFT feature extractor and using a preset function to extract target key points and target descriptors in the processed image of the target; Input the target key points and target descriptors of the headless item to be matched and the key points and descriptors corresponding to each headless item in the database into the headless item information matching model for matching analysis to obtain a matching result; Using a ratio test method to verify the accuracy of the matching result; When the accuracy of the matching result meets the preset matching condition, the headless component attribute information corresponding to the headless component to be matched is determined according to the matching result.

6. The method according to claim 5, characterized in that The use of a ratio test method to verify the accuracy of the matching result includes: Obtain the nearest neighbor headless component and the next nearest neighbor headless component corresponding to the headless component to be matched in the matching results; According to the key points and the descriptors, a ratio test method is used to calculate a first distance between the headless piece to be matched and the nearest neighbor headless piece and a second distance between the headless piece to be matched and the next nearest neighbor headless piece; calculating a distance ratio between the first distance and the second distance; When the distance ratio is less than a preset threshold, the nearest neighbor headless piece is determined to be a headless piece that matches the headless piece to be matched, and it is determined that the accuracy of the matching result meets the preset matching condition.

7. The method according to claim 1, characterized in that The method comprises: Obtaining the sample package size, sample package weight, sample shipping place, sample destination, sample transportation route and sample transportation time of the sample headerless package; Setting the weights of the sample package size and the sample package weight as the first weight; The weights of the sample shipping place and the sample destination are set to a second weight; the second weight is greater than the first weight; The weights of the sample transportation route and the sample transportation time are set as a third weight; the third weight is greater than the first weight; The step of searching the database for headless item attribute information matching the sample headless item package based on the sample package characteristics and the sample transportation information, and outputting the sample headless item attribute information includes: Determine the comprehensive characteristics of the sample based on a first product of the sample package characteristics and the first weight, a second product of the sample sending and receiving location and the second weight, and a third product of the sample transportation information and the third weight; Based on the comprehensive characteristics of the sample, the headless item attribute information matching the sample headless item package is searched from the database, and the sample headless item attribute information is output.

8. A headless piece wrapping matching device, characterized in that: The device comprises: A sample logistics data acquisition module is used to acquire sample logistics data of a sample headerless package in a logistics platform; the sample headerless package is marked with a sample headerless attribute label; A sample feature extraction module, used to input the sample logistics data of the sample headerless package into a machine learning model to extract package features and transportation information, and obtain sample package features and sample transportation information; The headless component attribute acquisition module is used to obtain the headless component attribute information of each headless component in the database; A sample attribute output module, used to search the database for headless item attribute information matching the sample headless item package based on the sample package characteristics and sample transportation information, and output the sample headless item attribute information; A model training module, used for training the machine learning model based on the difference between the sample headless component attribute information and the sample headless component attribute label to obtain a headless component information matching model; The attribute determination module is used to obtain the logistics data of the headless piece to be matched, input the logistics data into the headless piece information matching model, and obtain the headless piece attribute information corresponding to the headless piece to be matched.

9. An electronic device, characterized in that: The device includes: a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by the processor to implement the headless package matching method as described in any one of claims 1-7.

10. A computer storage medium, characterized in that: The computer storage medium stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by the processor to implement the headless package matching method as described in any one of claims 1-7.

Images