Online lease item return intelligent quality inspection monitoring method and system

Abstract

The invention discloses an intelligent quality inspection monitoring method and system for online lease item return, and relates to the technical field of information. The method comprises the following steps: acquiring online lease order information, including the historical order of a tenant, the lease duration, the postal road duration, the mailing mode and the quality state of a lease item; receiving a pre-box-entry equipment state photo and a returned package outer package multi-angle photo which are shot by the tenant before returning; analyzing the appearance state of the rental item before the rental item is sent back through an image recognition model, and calculating the probability that the parcel is violently transported by using an intelligent model; carrying out risk assessment by combining order information, appearance state and violent transportation probability, and dividing the returned parcels into a first class or a second class; and respectively executing quality inspection processes with different strict degrees. The system comprises corresponding function modules. According to the invention, automatic and intelligent quality inspection monitoring is realized, the quality inspection efficiency and accuracy are improved, the labor cost is reduced, and the resource allocation is optimized.

Description

Technical Field

[0001] This application relates to the field of information technology, specifically to an intelligent quality inspection and monitoring method and system for online rental item return. Background Technology

[0002] With the development of the sharing economy and the circular economy, online equipment leasing services are becoming increasingly popular. At the end of the lease term, the lessee must return the leased item to the leasing platform. Upon receiving the returned package, the platform must conduct a rigorous quality inspection of the leased item to assess its condition, determine depreciation costs, or assign liability. This is a crucial step in ensuring asset security and maintaining the sustainable operation of the business model.

[0003] Traditional quality inspection methods rely entirely on manual unpacking and inspection. Inspectors need to check the appearance and function of each piece of equipment one by one and determine whether the damage was present before the rental, caused during use, or occurred during transportation. The entire process is time-consuming and labor-intensive, creating a bottleneck when business volume is high and increasing the time cost of equipment turnaround. Manual quality inspection is highly subjective, with inconsistent standards for judging minor scratches and signs of use, which can easily lead to disputes between the platform and the lessee regarding liability for damage, affecting user experience and increasing after-sales costs. When rented items are damaged after being returned, it is difficult to determine the responsible party. Due to the lack of effective and reliable records of the equipment's condition before return and the lack of quantitative assessment of the risks of outer packaging transportation, the platform often falls into a dilemma of liability determination, bearing losses that should not be borne by itself. The existing process usually adopts a "one-size-fits-all" comprehensive quality inspection strategy for all returned packages without risk classification. This means that a large number of devices in good condition and with low risk also undergo complex quality inspection processes, resulting in a waste of human resources and equipment turnaround time.

[0004] Therefore, existing technologies lack a solution capable of automating and intelligently assessing the pre-shipment condition and predicting transportation risks of returned leased goods, and based on this, achieving precise and differentiated quality inspection to overcome the aforementioned deficiencies, improve inspection efficiency and accuracy, optimize resource allocation, and effectively clarify responsibilities. Summary of the Invention

[0005] This specification describes a method and system for intelligent quality inspection and monitoring of online rental item return shipments through several embodiments. The specific technical solutions adopted are as follows:

[0006] Firstly, this specification provides an online intelligent quality inspection and monitoring method for the return of leased items, comprising the following steps:

[0007] Obtain online rental order information, including the lessee's historical order information, rental duration, postal route duration, mailing method, and the quality and condition of the rented item;

[0008] Photos of the equipment status before it is placed in the box and multi-angle photos of the outer packaging of the returned parcel, taken before the lessee sends the parcel back;

[0009] Based on the photos of the equipment status before it was placed in the box, the appearance of the leased item before it was returned was analyzed using an image recognition model.

[0010] Based on multi-angle photos of the outer packaging of the returned parcel, an intelligent model is used to identify and calculate the probability that the parcel was subjected to rough handling during transportation.

[0011] Based on the order information, the appearance condition of the leased item before return, and the probability of rough handling, a risk assessment is conducted on the returned parcels, and the returned parcels are classified into Class I or Class II parcels accordingly.

[0012] A first-class quality inspection process is performed on the first type of package, and a second-class quality inspection process is performed on the second type of package. The second-class quality inspection process has more inspection items than the first-class quality inspection process.

[0013] Secondly, embodiments of this specification provide an online intelligent quality inspection and monitoring system for the return of leased items, including:

[0014] The reading module retrieves online rental order information, which includes the lessee's historical order information, rental duration, postal route duration, mailing method, and the quality status of the rented item.

[0015] The receiving module receives photos of the equipment status before it is placed in the box and multi-angle photos of the outer packaging of the returned parcel taken by the lessee before the parcel is returned.

[0016] The analysis module, based on the photos of the equipment status before it was placed in the box, analyzes the appearance of the leased item before it is returned using an image recognition model;

[0017] The identification module, based on multi-angle photos of the outer packaging of the returned parcel, uses an intelligent model to identify and calculate the probability that the parcel was subjected to rough handling during transportation;

[0018] The assessment module, combining the order information, the appearance condition of the leased item before return, and the probability of rough handling, conducts a risk assessment of the returned package and classifies the returned package into Class I or Class II packages accordingly.

[0019] The classification module performs a first-class quality inspection process on the first-class packages and a second-class quality inspection process on the second-class packages. The second-class quality inspection process has more inspection items than the first-class quality inspection process.

[0020] Thirdly, embodiments of this specification provide an electronic device, including a processor and a memory;

[0021] The processor is connected to the memory;

[0022] The memory is used to store executable program code;

[0023] The processor runs a program corresponding to the executable program code stored in the memory to perform the method described in any of the above aspects.

[0024] Fourthly, embodiments of this specification provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the methods described in any of the above aspects.

[0025] Fifthly, embodiments of this specification provide a computer program product, including a computer program that, when executed by a processor, implements the methods described in any of the above aspects.

[0026] The beneficial effects of the technical solutions provided in some embodiments of this specification include at least the following:

[0027] This specification provides an intelligent quality inspection and monitoring method for the return of online rental items. By acquiring rental order information and photos of the equipment's condition and outer packaging taken by the lessee before return, and integrating image recognition and intelligent prediction models, it achieves automated and accurate risk assessment of returned packages. Based on the analysis of the rental item's appearance before return, the calculation of the probability of rough handling during transportation, and combined with multi-dimensional information such as the lessee's credit and rental duration, a comprehensive risk index is constructed. This intelligently classifies packages into "Category I" or "Category II" and executes differentiated quality inspection processes. This process improves quality inspection efficiency, optimizes human resource allocation, and reduces operating costs. Simultaneously, through full-process data recording and objective analysis, it effectively defines responsibility for equipment damage, reduces potential disputes between the lessor and lessee and the logistics carrier, and enhances process transparency and user trust. The various models used in the system possess self-learning and continuous optimization capabilities, can adapt to complex and ever-changing real-world application scenarios, and demonstrate strong scalability and practicality, providing an efficient, reliable, and scalable intelligent solution for the quality inspection of returned items in the online rental industry.

[0028] Other features and advantages of various embodiments of this specification will be further revealed in the following detailed description and accompanying drawings. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the embodiments of this specification, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0030] Figure 1 This document provides a flowchart of an online intelligent quality inspection and monitoring method for the return of leased items.

[0031] Figure 2 This is a flowchart illustrating the method for analyzing the appearance of the leased item before its return, as provided in this manual.

[0032] Figure 3 This is a schematic diagram of an online intelligent quality inspection and monitoring system for the return of leased items, as provided in this manual.

[0033] Figure 4 A schematic diagram of an electronic device provided in an embodiment of this specification. Detailed Implementation

[0034] The technical solutions of the embodiments of this specification will be explained and described below with reference to the accompanying drawings. However, the following embodiments are only preferred embodiments of this specification and not all of them. Other embodiments obtained by those skilled in the art based on the embodiments in the implementation methods without creative effort are all within the protection scope of this specification.

[0035] The terms "first," "second," "third," etc., in the description, claims, and accompanying drawings are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such processes, methods, products, or apparatus.

[0036] In the following description, terms such as “inner,” “outer,” “upper,” “lower,” “left,” and “right” are used only to facilitate the description of the embodiments and to simplify the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this specification.

[0037] All data involved in this application are information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0038] Please see Figure 1 The flowchart shown is a method for intelligent quality inspection and monitoring of online rental item return provided in an embodiment of this specification.

[0039] With the booming development of online leasing businesses, the return and quality inspection of leased items after the lease term has become a critical stage affecting operational efficiency and user experience. Traditional quality inspection models heavily rely on manual unpacking and inspection, which has the following significant drawbacks: First, it is inefficient and cannot handle large-scale return shipments, easily leading to warehouse backlogs; second, human judgment is highly subjective, easily causing disputes with lessees, such as the inability to determine whether equipment damage occurred during user use or during return transport; third, it lacks the ability to anticipate transportation risks, often discovering damage only after unpacking, resulting in lengthy and costly claims processes. Some solutions attempt to introduce user-taken photos as evidence, but these photos are mostly isolated evidence, only serving as a reference for post-event accountability, and cannot be used for intelligent analysis and risk warning. The entire return quality inspection process remains in a passive response state, lacking a comprehensive, intelligent risk assessment and classification solution that spans the entire chain from "before user return," "during transport," and "after warehouse receipt."

[0040] like Figure 1 As shown in the embodiments of this specification, an online intelligent quality inspection and monitoring method for the return of leased items includes the following steps:

[0041] Step S1: Obtain online rental order information, which includes the lessee's historical order information, rental duration, postal route duration, mailing method, and the quality status of the rented item.

[0042] Obtaining online rental order information is the starting point of the entire intelligent quality inspection and monitoring process. The obtained order information includes the following key elements:

[0043] The lessee's historical order information includes the number of rentals, the type of rented items, historical damage records, credit score, etc., which are used to assess the lessee's creditworthiness and risk level;

[0044] Rental duration refers to the length of time the item is currently rented, used to determine whether the item is in a vulnerable or high-wear stage;

[0045] Postal route length refers to the estimated mailing time from the lessee's location to the leasing company's warehouse. The longer the postal route, the more transportation risks may be involved.

[0046] The mailing method refers to the type of courier service used; different mailing methods have different risk levels.

[0047] The condition of the leased item refers to its condition during the lease period, used to determine the item's fragility.

[0048] This information collectively forms the basic data for risk assessment, helping the system determine the overall risk level of returned parcels.

[0049] Step S2: Receive photos of the equipment status before it is placed in the box and multi-angle photos of the outer packaging of the returned package taken by the lessee before returning it.

[0050] Photos of the equipment before it was placed in the container:

[0051] Photos taken by the lessee before packing the rented item into a parcel; used to record the original appearance of the rented item before it is returned; to provide raw data for subsequent image recognition analysis; the photos must include key parts of the rented item for detailed inspection.

[0052] Multiple photos of the outer packaging of the return parcel from various angles:

[0053] Multiple photos of the package's outer packaging taken by the lessee from various angles are used to analyze the potential impact of external forces on the package during transportation. These multi-angle photos allow for a comprehensive assessment of the integrity of the package's outer packaging and provide an image data foundation for subsequent calculations of the probability of rough handling during transportation.

[0054] These photos are key inputs for the system's intelligent quality inspection. Image recognition technology can analyze the initial state of the rented items and the potential damage to the packages during transportation, thus providing a basis for subsequent risk assessment and quality inspection processes.

[0055] Step S3: Based on the photos of the equipment status before it was placed in the box, analyze the appearance of the leased item before it is returned using an image recognition model.

[0056] Please refer to the appendix for details. Figure 2 The steps for analyzing the appearance of leased items before return using image recognition models include:

[0057] Step S31: Read the preset key part marking information of the rental property, the marking information including part name, location coordinates and appearance feature template;

[0058] Step S32: Perform image preprocessing on the equipment status photo before it is placed in the box, and locate the key parts of the rental item in the preprocessed photo based on the key part marking information;

[0059] Step S33: Compare the images of each key area with the corresponding appearance feature template to identify whether there are defects;

[0060] Step S34: Generate a description of the appearance of the leased item before it is returned based on the recognition results, and output the damage prediction score for each key part.

[0061] The system predefines the marking information for each key part of the leased property; this information forms the basis for the system's analysis. The marking information is as follows:

[0062] Part names: such as "screen", "bezel", "camera", "interface", and other key components of the leased item;

[0063] Location coordinates: The relative location coordinates of these parts in the standard photograph, used for subsequent positioning;

[0064] Appearance feature template: Normal state image or feature description of each key part, used for comparison with the actual photo.

[0065] The system optimizes photos of the equipment's condition before it is placed in the container, taken by the lessee. This includes adjusting brightness and contrast to improve image quality; removing noise to reduce interference; enhancing the image to highlight key features; and performing image cropping and rotation correction. Using preset location coordinates, the system accurately locates key areas of the rented item in the pre-processed photos, preparing for subsequent comparisons.

[0066] The system will accurately compare the images of the key areas located with the preset appearance feature templates. Based on the recognition results, the system will automatically generate a description of the appearance of the rented item before it is returned. The score will serve as important input data for subsequent risk assessment to determine the risk level of the returned package.

[0067] Step S4: Based on the multi-angle photos of the outer packaging of the returned parcel, use an intelligent model to identify and calculate the probability that the parcel was subjected to rough handling during transportation.

[0068] The steps involved in using intelligent models to identify and calculate the probability that the package was subjected to rough handling during transportation include:

[0069] The multi-angle photos are stitched together or the viewpoints are fused to generate a three-dimensional surface reconstruction map of the outer packaging of the package.

[0070] Extract deformation features and damage information of the outer packaging of the package, including damaged areas, indentations, tears, tape abnormalities, and improper sealing information;

[0071] The deformation features and damage information are input into a pre-trained violent transportation risk prediction model to obtain a violent transportation probability estimate.

[0072] A correction weight is generated based on the mailing method, postal route duration, and the historical performance of the carrier logistics service provider.

[0073] Based on the corrected weights and the estimated probability of violent transportation, the probability that a package will be subjected to violent transportation during the transportation process is obtained.

[0074] By combining multi-angle image stitching and 3D reconstruction technology with deep learning models, the system accurately identifies the deformation and damage characteristics of package packaging, enabling a quantitative assessment of violent behavior during transportation. It can capture physical damage traces of packages from multiple perspectives and generate dynamic correction weights based on historical logistics data and carrier performance, thus scientifically calibrating prediction results by comprehensively considering transportation route characteristics and service reliability. Through the deep integration of image recognition, behavioral modeling, and logistics big data, the system provides quantifiable decision-making basis for the full-process safety monitoring of express parcels, reducing disputes and economic losses caused by violent transportation, and has significant practical value for building an efficient and transparent modern logistics system.

[0075] The methods for training violent transportation risk prediction models include:

[0076] Collect historical data containing various types of damage to the outer packaging of packages. The historical data includes multi-angle photos, deformation features, damage information, and records of damage during rough transportation.

[0077] The collected historical data is labeled, including whether it was subjected to violent transportation and the severity level;

[0078] Construct a deep learning model architecture by dividing the labeled historical data into a training set, a validation set, and a test set. Use the training set to train the model and use the validation set to adjust the model parameters.

[0079] Based on the trained deep learning model, a violent transportation risk prediction model is obtained.

[0080] By collecting and labeling historical data containing various types of package packaging damage, the diversity and authenticity of the training data are ensured. This allows the model to learn the characteristics of rough handling in various real-world scenarios, rather than being limited to a few cases. This enables more accurate and reliable probability predictions when faced with new and unseen packages. By dividing the data into training, validation, and test sets, and using the validation set to adjust model parameters, the model is ensured not to merely "memorize" the training data, but to truly learn generalization rules, possessing the ability for continuous optimization and iterative improvement. Labeling historical data with "whether it suffered rough handling and the severity level" transforms the subjective experience and judgment standards of quality inspectors or logistics experts into objective and unified labels suitable for machine learning. The high-precision model provides high-quality input data for subsequent comprehensive risk assessment, ensuring a more scientific and accurate classification decision between "Category I" and "Category II" packages, thus achieving optimal resource allocation.

[0081] Step S5: Based on the order information, the appearance condition of the leased item before return, and the probability of rough handling, conduct a risk assessment of the returned package and classify it into Class I or Class II packages accordingly.

[0082] The steps for classifying return parcels into Category I or Category II parcels include:

[0083] A credit score is assigned to the lessee based on their historical order information. If the credit score is lower than a preset threshold, the risk weight is increased.

[0084] Based on the lease duration and the condition of the leased item, determine whether the equipment is in a vulnerable period or a high-wear stage. If it is in a high-wear stage, increase the risk weight.

[0085] The damage prediction scores of each key part of the leased item are weighted and summarized before the item is returned to generate an overall appearance risk score for the equipment.

[0086] The probability of violent transportation, the risk score of the overall appearance of the equipment, the postal route duration, the reliability of the mailing method, and the lessee's credit score are input into a preset risk fusion model to calculate a comprehensive risk index.

[0087] A preset risk threshold is set. When the comprehensive risk index is lower than the first threshold, the package is classified as a Class I package; when the comprehensive risk index is higher than or equal to the first threshold, the package is classified as a Class II package.

[0088] The system calculates a credit score based on the lessee's historical order data, such as on-time return rate, equipment damage rate, complaint records, and payment timeliness.

[0089] Preset threshold: For example, set to 75 points out of 100 points. Tenants with scores below this threshold have a lower credit rating.

[0090] Increased risk weight: If the credit score is below the threshold (e.g., 65 points), the system will automatically add an extra risk weight to the package (e.g., increase the weighting coefficient by 0.15).

[0091] The system determines the rental duration and the condition of the leased item.

[0092] The system assigns different weights to the damage prediction scores of each key part according to their importance, generating a quantitative index that comprehensively reflects the appearance condition of the equipment before it is returned, which is used for risk assessment.

[0093] Input multi-dimensional risk factors into a pre-defined risk fusion model to calculate a comprehensive risk index.

[0094] Input factors: probability of violent transportation; risk score of overall equipment appearance; postal route duration; reliability score of mailing method; credit score of lessee.

[0095] The steps involved in setting up the risk fusion model include:

[0096] Identify key factors affecting the risk of returned parcels, including lessee credit score, overall equipment appearance risk score, probability of rough handling, postal route duration, and reliability score of mailing method;

[0097] Initial weights are assigned to each key factor, and these initial weights are determined based on correlation analysis between each factor and the actual damage results in historical return data.

[0098] A comprehensive risk index is calculated based on the lessee's credit score, the risk score of the overall appearance of the equipment, the probability of violent transportation, the postal route duration, and the reliability score of the mailing method.

[0099] The comprehensive risk index is correlated with historical quality inspection results as sample data, and the risk fusion model is established and trained using the sample data.

[0100] The final risk fusion model is obtained based on the trained risk fusion model.

[0101] The risk fusion model identified five key risk factors that collectively influence the risk assessment of returned parcels.

[0102] Lessee credit score: A credit rating calculated based on the lessee's historical order data, reflecting the lessee's potential tendency to cause damage;

[0103] Overall equipment appearance risk score: a summary of damage prediction scores for the leased item before return, obtained through the preceding image analysis;

[0104] Probability of rough handling during transportation: The probability of a package being subjected to rough handling during transportation, obtained through analysis of its outer packaging;

[0105] Postal route duration: The estimated transit time for a package from the lessee's location to the leasing company's warehouse;

[0106] Mailing method reliability rating: Reliability assessment of different mailing methods.

[0107] Based on the correlation analysis between each factor and the actual damage results in historical return data, statistical analysis methods were used to determine the importance of each factor.

[0108] Comprehensive Risk Index = (Lessee Credit Score × Credit Score Weight) + (Overall Equipment Appearance Risk Score × Appearance Risk Weight) + (Probability of Violent Transportation × Violent Transportation Weight) + (Postal Route Duration × Postal Route Duration Weight) + (Reliability Score of Mailing Method × Reliability Weight).

[0109] By linking the comprehensive risk index of historical returned parcels with actual quality inspection results as a sample, the model will continuously adjust the weights and calculation methods of each factor through iterative training, so that the prediction results are more consistent with the actual historical results.

[0110] Step S6: Perform a first-class quality inspection process on the first type of package and a second-class quality inspection process on the second type of package. The second-class quality inspection process has more inspection items than the first-class quality inspection process.

[0111] On the other hand, please refer to Appendix 3. This manual provides an online intelligent quality inspection and monitoring system for the return of leased items, including:

[0112] The reading module 100 actively retrieves online rental order information from the rental platform's order database. The information it retrieves includes, but is not limited to: the lessee's historical order information (used for credit assessment), the rental duration of this lease, the estimated postal delivery time, the user's chosen mailing method, and the initial quality condition of the equipment at the time of rental. The reading module forms the data foundation for the system's risk assessment.

[0113] The receiving module 200 is responsible for receiving and temporarily storing two types of key image data that the lessee takes and uploads according to the prescribed procedures before returning the item via a mobile application: one is "equipment status photos before being placed in the box", which is used to record the physical condition of the rented item before it is handed over to the logistics company; the other is "multi-angle photos of the outer packaging of the return package", which is used to assess the integrity of the packaging.

[0114] The analysis module 300, connected to an image recognition model library, is specifically designed to process photos of the equipment's condition before it is placed in the container. It calls upon pre-stored key component marker information of the leased item, such as component name, coordinates, and standard appearance templates, to preprocess the uploaded photos. Then, it accurately locates the key component areas and uses a comparison algorithm to identify any defects such as scratches or dents. Finally, the analysis module generates a structured appearance condition description report and outputs a quantified damage prediction score for each key component.

[0115] The identification module 400 focuses on assessing transportation risks. It first processes multi-angle photographs of the outer packaging of return parcels to extract deformation features and damage information. Then, this feature data is input into a pre-trained model for predicting the risk of rough handling during transportation, obtaining an initial probability estimate. This module also integrates external data, generating a correction weight based on the mailing method, delivery time, and the logistics provider's historical performance to calibrate the initial probability, ultimately outputting a more accurate "probability of rough handling during transportation."

[0116] The assessment module 500 receives order information from the reading module, appearance status reports from the analysis module (especially the overall equipment appearance risk score), and the probability of rough handling calculated by the identification module. It incorporates a risk fusion model that integrates the aforementioned multi-source data, such as lessee credit scores, equipment appearance risk, transportation probability, and postal route duration, to derive a comprehensive risk index. Based on preset risk thresholds, this module intelligently classifies each return parcel into either Category I or Category II parcels.

[0117] The classification module 600, which integrates with the warehouse management system, triggers different quality inspection instructions based on the classification results from the evaluation module. For return parcels classified as Category 1, the system instructs the execution of a standardized Category 1 quality inspection process, which may only involve visual inspection and basic functional testing. For Category 2 parcels, a more stringent and comprehensive Category 2 quality inspection process is instructed, with more inspection items, potentially including in-depth performance diagnostics and internal structural checks, thereby achieving precise allocation of quality inspection resources. The Category 1 quality inspection process is executed for Category 1 parcels, and the Category 2 quality inspection process is executed for Category 2 parcels, with the Category 2 process involving more inspection items than the Category 1 process.

[0118] This embodiment of an online rental item return intelligent quality inspection and monitoring system constructs a closed loop from data entry to intelligent decision-making and process execution through the above modules. It effectively transforms the traditional passive and manual quality inspection mode into an active and intelligent monitoring and disposal mode, providing strong technical support for asset management in the online rental industry.

[0119] Please see Figure 3 The diagram shows a structural schematic of an electronic device provided in an embodiment of this specification.

[0120] like Figure 3As shown, the electronic device 1100 may include: at least one processor 1101, at least one network interface 1104, a user interface 1103, a memory 1105, and at least one communication bus 1102. The communication bus 1102 can be used to connect and communicate with the various components mentioned above. The user interface 1103 may include buttons, and optionally may include standard wired or wireless interfaces. The network interface 1104 may include, but is not limited to, a Bluetooth module, an NFC module, or a WiFi module. The processor 1101 may include one or more processing cores. The processor 1101 connects to various parts within the electronic device 1100 using various interfaces and lines, and performs various functions of the routing device and processes data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 1105, and by calling data stored in the memory 1105. Optionally, the processor 1101 may be implemented using at least one hardware form of DSP, FPGA, or PLA. The processor 1101 may integrate one or more combinations of CPU, GPU, and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content that the display screen needs to show; and the modem is used for wireless communication.

[0121] It is understandable that the aforementioned modem may not be integrated into the processor 1101, but may be implemented using a separate chip.

[0122] The memory 1105 may include RAM or ROM. Optionally, the memory 1105 may include a non-transitory computer-readable medium. The memory 1105 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 1105 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory 1105 may also be at least one storage device located remotely from the aforementioned processor 1101. As a computer storage medium, the memory 1105 may include an operating system, a network communication module, a user interface module, and application programs. The processor 1101 may be used to call the application programs stored in the memory 1105 and execute the methods in the above-described embodiments.

[0123] This specification also provides a computer-readable storage medium storing instructions that, when executed on a computer or processor, cause the computer or processor to perform multiple steps as described in the above embodiments. If the constituent modules of the above-described electronic device are implemented as software functional units and sold or used as independent products, they can be stored in the computer-readable storage medium.

[0124] This specification also provides a computer program product, including a computer program that, when executed by a processor, implements the multiple steps described in the above embodiments.

[0125] Where there is no conflict, the technical features in this embodiment and implementation scheme can be combined arbitrarily.

[0126] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes multiple computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this specification are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted through the computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center integrating multiple available media. The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., digital versatile discs (DVDs)), or semiconductor media (e.g., solid-state drives (SSDs)).

[0127] When implemented through hardware or firmware, the aforementioned method flow is programmed into the hardware circuit to obtain the corresponding hardware circuit structure and achieve the corresponding function. For example, a Programmable Logic Device (PLD) (such as a Field Programmable Gate Array (FPGA)) is such an integrated circuit, whose logic function is determined by the user programming the device. Designers can program a digital system onto a PLD themselves, eliminating the need for chip manufacturers to design and fabricate dedicated integrated circuit chips. Furthermore, nowadays, instead of manually fabricating integrated circuit chips, this programming is mostly implemented using "logic compiler" software, similar to the software compiler used in program development. The original code before compilation must also be written in a specific programming language, called a Hardware Description Language (HDL). There is not just one HDL, but many. Those skilled in the art should understand that by simply performing some logic programming on the method flow using one of the aforementioned hardware description languages ​​and programming it into an integrated circuit, the hardware circuit implementing the logic method flow can be easily obtained.

[0128] The embodiments described above are merely preferred embodiments of this specification and are not intended to limit the scope of this specification. Any modifications and improvements made by those skilled in the art to the technical solutions of this specification without departing from the spirit of this specification should fall within the protection scope defined by the claims of this specification.

Claims

1. A method for intelligent quality inspection and monitoring of online rental item return, characterized in that, Includes the following steps: Obtain online rental order information, including the lessee's historical order information, rental duration, postal route duration, mailing method, and the quality and condition of the rented item; Photos of the equipment status before it is placed in the box and multi-angle photos of the outer packaging of the returned parcel, taken before the lessee sends the parcel back; Based on the photos of the equipment status before it was placed in the box, the appearance of the leased item before it was returned was analyzed using an image recognition model. Based on multi-angle photos of the outer packaging of the returned parcel, an intelligent model is used to identify and calculate the probability that the parcel was subjected to rough handling during transportation. Based on the order information, the appearance condition of the leased item before return, and the probability of rough handling, a risk assessment is conducted on the returned parcels, and the returned parcels are classified into Class I or Class II parcels accordingly. A first-class quality inspection process is performed on the first type of package, and a second-class quality inspection process is performed on the second type of package. The second-class quality inspection process has more inspection items than the first-class quality inspection process.

2. The intelligent quality inspection and monitoring method for online rental item return according to claim 1, characterized in that, Based on the photograph of the equipment's condition before being placed in the box, the method for analyzing the appearance of the leased item before its return using an image recognition model includes: Read the preset key part marking information of the leased property, the marking information including part name, location coordinates and appearance feature template; The equipment status photos before being placed in the box are preprocessed, and the key parts of the rental item are located in the preprocessed photos based on the key part marking information. The images of each key area are compared with the corresponding appearance feature templates to identify whether defects exist; Based on the identification results, a description of the appearance of the leased item before its return is generated, and damage prediction scores for each key part are output.

3. The intelligent quality inspection and monitoring method for online rental item return according to claim 1, characterized in that, Based on multi-angle photos of the outer packaging of the returned parcel, a method for identifying and calculating the probability that the parcel was subjected to rough handling during transportation using an intelligent model includes: The multi-angle photos are stitched together or the viewpoints are fused to generate a three-dimensional surface reconstruction map of the outer packaging of the package. Extract deformation features and damage information of the outer packaging of the package, including damaged areas, indentations, tears, tape abnormalities, and improper sealing information; The deformation features and damage information are input into a pre-trained violent transportation risk prediction model to obtain a violent transportation probability estimate. A correction weight is generated based on the mailing method, postal route duration, and the historical performance of the carrier logistics service provider. Based on the corrected weights and the estimated probability of violent transportation, the probability that a package will be subjected to violent transportation during the transportation process is obtained.

4. The online rental item return intelligent quality inspection and monitoring method according to claim 3, characterized in that, Methods for training violent transportation risk prediction models include: Collect historical data containing various types of damage to the outer packaging of packages. The historical data includes multi-angle photos, deformation features, damage information, and records of damage during rough transportation. The collected historical data is labeled, including whether it was subjected to violent transportation and the severity level; Construct a deep learning model architecture by dividing the labeled historical data into a training set, a validation set, and a test set. Use the training set to train the model and use the validation set to adjust the model parameters. Based on the trained deep learning model, a violent transportation risk prediction model is obtained.

5. The intelligent quality inspection and monitoring method for online rental item return according to claim 1, characterized in that, The method for assessing the risk of returned parcels based on the order information, the physical condition of the rented item before return, and the probability of rough handling during transport, and classifying returned parcels into Class I or Class II parcels accordingly, includes: A credit score is assigned to the lessee based on their historical order information. If the credit score is lower than a preset threshold, the risk weight is increased. Based on the lease duration and the condition of the leased item, determine whether the equipment is in a vulnerable period or a high-wear stage. If it is in a high-wear stage, increase the risk weight. The damage prediction scores of each key part of the leased item are weighted and summarized before the item is returned to generate an overall appearance risk score for the equipment. The probability of violent transportation, the risk score of the overall appearance of the equipment, the postal route duration, the reliability of the mailing method, and the lessee's credit score are input into a preset risk fusion model to calculate a comprehensive risk index. A preset risk threshold is set. When the comprehensive risk index is lower than the first threshold, the package is classified as a Class I package; when the comprehensive risk index is higher than or equal to the first threshold, the package is classified as a Class II package.

6. The intelligent quality inspection and monitoring method for online rental item return according to claim 1, characterized in that, Methods for pre-setting risk fusion models include: Identify key factors affecting the risk of returned parcels, including lessee credit score, overall equipment appearance risk score, probability of rough handling, postal route duration, and reliability score of mailing method; Initial weights are assigned to each key factor, and these initial weights are determined based on correlation analysis between each factor and the actual damage results in historical return data. A comprehensive risk index is calculated based on the lessee's credit score, the risk score of the overall appearance of the equipment, the probability of violent transportation, the postal route duration, and the reliability score of the mailing method. The comprehensive risk index is correlated with historical quality inspection results as sample data, and the risk fusion model is established and trained using the sample data. The final risk fusion model is obtained based on the trained risk fusion model.

7. An intelligent quality inspection and monitoring system for the return of online leased items, characterized in that, include: The reading module retrieves online rental order information, which includes the lessee's historical order information, rental duration, postal route duration, mailing method, and the quality status of the rented item. The receiving module receives photos of the equipment status before it is placed in the box and multi-angle photos of the outer packaging of the returned parcel taken by the lessee before the parcel is returned. The analysis module, based on the photos of the equipment status before it was placed in the box, analyzes the appearance of the leased item before it is returned using an image recognition model; The identification module, based on multi-angle photos of the outer packaging of the returned parcel, uses an intelligent model to identify and calculate the probability that the parcel was subjected to rough handling during transportation; The assessment module, combining the order information, the appearance condition of the leased item before return, and the probability of rough handling, conducts a risk assessment of the returned package and classifies the returned package into Class I or Class II packages accordingly. The classification module performs a first-class quality inspection process on the first-class packages and a second-class quality inspection process on the second-class packages. The second-class quality inspection process has more inspection items than the first-class quality inspection process.

8. An electronic device, characterized in that, Including the processor and memory; The processor is connected to the memory; The memory is used to store executable program code; The processor runs a program corresponding to the executable program code stored in the memory to perform the method as described in any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-6.

Images