Second-hand commodity transaction system

Abstract

The invention relates to a second-hand commodity transaction system, which provides a digital resume of an object provided by the system during online transaction of a buyer and a seller, provides a service for'instant verification 'whether the source and content of the object are true or not in the process of delivering the object, and comprises a commodity transaction platform module, a digital resume management module and a feature image comparison module, the system comprises a digital resume certificate output module and a database, and the modules and the database are connected to perform two-way communication of information and data exchange so as to form a transaction system.

Description

Technical Field

[0001] The technical means of the present invention mainly relates to the field of online transactions of network cables, especially a trading system for second-hand goods that can be verified online during transactions. Background Art

[0002] According to a study by the Boston Consulting Group (BCG), the second-hand goods trading market is attracting attention. More and more people are reluctant to buy new products, mainly in response to the concept of "sustainable fashion" and in line with the ethical consumption concept of cherishing things, making the use of items more sustainable and environmentally friendly. Among them, there are hundreds of women's brands, and the second-hand goods market for men is also constantly expanding. In addition, consumers are gradually starting to reject "fast fashion" which is fast but of poor quality. Even though many commodity trading websites promote second-hand luxury goods of well-known designs at a super-high discount of 30% of the original retail price, and even include almost brand-new second-hand goods that have hardly been used, they may not necessarily be favored by consumers because there is a lot of false information and deception, resulting in many disputes in transactions by end consumers and generating a sense of distrust. Although general luxury goods come with the original factory certificate or warranty when sold, when such luxury goods are resold as second-hand goods to others, even if the seller provides the original factory certificate, there may still be a situation where the item is swapped because ordinary consumers lack the ability to identify whether the item is genuine. Even if it is agreed to entrust a third party to participate in the verification, it takes a long time and an additional expensive fee has to be paid. This is also the main reason why many second-hand luxury goods trading markets, although hiding huge business opportunities, have been unable to be greatly promoted.

[0003] Existing platforms participating in online transactions of second-hand goods, including the technical content disclosed in the patent document of "Second-hand Goods Trading System and Its Method" No. 201117124 in Taiwan, China, are mostly aimed at providing a convenient and safe mode that expects to protect both buyers and sellers during the "process" of transactions. The so-called "safety" here refers to the "process" of online platform transactions, that is, the design of the platform is to protect consumers as much as possible to complete transactions according to their wishes and instructions, rather than verifying the second-hand items themselves, especially "instant" verification to protect the "result" of consumers obtaining the items. Assuming the second-hand item is an inexpensive commodity, perhaps consumers won't care particularly, but if it is a high-priced luxury item such as a well-known brand leather bag, the resulting losses may be difficult to estimate. Summary of the Invention

[0004] The present invention is a second-hand commodity trading system, which provides the process of online trading between buyers and sellers to obtain the digital resume of the items provided by the system as a reference for whether to purchase, and provides "instant verification" during the process of delivering the items to ensure whether the source and content of the items are true, so as to guarantee the security of the trading result of second-hand commodities, which is the main purpose of the present invention.

[0005] To achieve the invention purpose, in addition to including the existing C2C (Consumer To Consumer) trading platform, the trading system of the present invention further introduces AI (Artificial Intelligence) artificial intelligence identification technology. Through continuous training and accumulation, the system helps to identify the authenticity of the content of the items and makes an instant judgment to feedback the information to the trading platform to enable safe trading between buyers and sellers. Its main technical content is realized by a digital resume management module, a feature image comparison module, a digital resume certificate generation module, and a database; among them, The digital resume management module communicates and connects with the commodity trading platform, and is used to provide identification, establish a resume, analyze and manage the second-hand commodities with trading requirements. After receiving the information of the second-hand items from the seller on the trading platform, it arranges in advance to identify and authenticate whether the second-hand items are genuine products from the original factory through an appraisal institution, and records resume items including but not limited to the "item name", "brand name", "factory information (information on the purchase certificate)", "certification date", "feature description (including scratches, old marks, etc.)", and "certification result" of the second-hand items. At the same time, through an imaging device, "feature images" of different angles of the second-hand items are obtained by using technologies including but not limited to photography and scanning, which are used as the basis for verifying the second-hand items in the future.

[0006] The digital resume certificate generation module is connected between the digital resume management module and the database, and is used to generate a digital resume certificate for the resume items of the second-hand items recorded by the digital resume management module, including the "item name", "brand name", "factory information (information on the purchase certificate)", "certification date", "feature description (including scratches, old marks, etc.)", and "certification result", etc. A unique digital code is assigned to the digital resume certificate, and the "feature image" selected from the second-hand items is paired with the digital code and stored in the database.

[0007] The feature image comparison module is connected between the digital resume management module and the database, and incorporates AI (Artificial Intelligence) artificial intelligence identification technology to identify the "feature images" of the second-hand items stored in the database.

[0008] This database is used to store the digital resume certificates produced by the digital resume certificate generation module and the characteristic images of the second-hand goods, and provide access to the digital resume certificate generation module and the characteristic image comparison module for resume modification of the digital resume certificate and image comparison of the second-hand goods.

[0009] In this way, when the system user of the buyer selects an item on the trading platform, they can preview the items recorded on the digital resume certificate as a reference for purchasing the commodity, so as to fully understand the actual situation of the second-hand goods. After the purchase process is confirmed, the trading platform will send the encrypted digital resume certificate to the buyer as a guarantee corresponding to the item. After the item holder of the buyer has obtained a digital resume certificate with a unique digital code, if they want to resell the item to a third party, in addition to delivering the item to the third party for inspection, they can also pre-check the authenticity and resume of the item through the trading system, and enter the unique digital code on the digital resume certificate they hold to confirm whether the source of the second-hand item comes from this trading system. After confirmation, they need to capture the "characteristic images" of each angle of the item according to the requirements of the digital resume management module of the system. The characteristic image comparison module of the system will compare each of the input "characteristic images" with the "characteristic images" of the item originally stored in the database to check if they match. If they match, the digital resume management module will send the comparison result to the commodity trading platform module and feedback it to the platform user to verify the source and resume content of the second-hand item, so as to prevent the seller from delivering a false item to the buyer, or providing the buyer with the ability to "instantly" confirm the authenticity of the item through such a verification mechanism after receiving the second-hand item to protect the rights and interests of the buyer. Brief Description of the Drawings

[0010] Figure 1 It is the structural diagram of the trading system of the present invention.

[0011] Figure 2 It is the reference diagram of the digital resume certificate of the present invention.

[0012] Figure 3 It is the flow chart of the second-hand item holder verifying the commodity through the trading system of the present invention.

[0013] Figures 4a to 4d It is the content of the characteristic image selection of the example item of the present invention.

[0014] In the figure: 1 Trading system 10 Commodity trading platform module 100 User 101 Administrator 11 Digital resume management module 12 Characteristic image comparison module 13 Digital Resume Certificate Generation Module 14 Database 15 Digital Encryption Module 2 Digital Resume Certificate 20 Digital Coding. Detailed Implementation Manner

[0015] The present invention is an online trading system for trading second-hand goods, especially second-hand "boutiques", which provides verification and resumes provided by the system to the buyer and seller during the trading process to distinguish authenticity, so as to protect the rights and interests of the buyer and seller of second-hand goods in online trading and avoid many unnecessary disputes in the future.

[0016] As Figure 1 and Figure 2 shown, the trading system 1 described in the present invention mainly includes: a commodity trading platform module 10, a digital resume management module 11, a feature image comparison module 12, a digital resume certificate generation module 13, and a database 14; information and data can be communicated bidirectionally between the modules through connections to form a trading system 1.

[0017] In the embodiment disclosed in the present invention, the commodity trading platform module 10 can be but is not limited to a C2C (Consumer To Consumer) platform, which is used to provide front-end users 100, that is, consumers of both the buyer and the seller, to conduct transactions online, and provide services and system maintenance for the back-end manager 101.

[0018] The digital resume management module 11 is communicatively connected to the commodity trading platform module 10, and is used to identify, establish a resume, analyze and manage the traded commodities; after receiving the seller's second-hand item information from the trading platform module 10, it arranges in advance to identify and authenticate whether the second-hand item is a genuine product from the original factory through an appraisal agency, and records resume items including but not limited to the "item name", "brand name", "factory information (information on the purchase certificate)", "certification date", "feature description (including scratches, old marks... etc.)", and "certification result" of the second-hand item. At the same time, through an imaging device, "feature images" of different angles of the second-hand item are obtained by using technologies including but not limited to photography and scanning, as the basis for authenticating the second-hand item in the future.

[0019] The digital resume certificate generation module 13 is connected between the digital resume management module 11 and the database 14, and is used to output the resume items of the "item name", "brand name", "factory information (information on the purchase certificate)", "certification date", "feature description (including scratches, old marks... etc.)", and "certification result" of the second-hand item recorded by the aforementioned digital resume management module 11 as follows Figure 2The digital resume certificate 2 shown, and a unique digital code 20 is assigned to the digital resume certificate 2, and the "feature image" selected from the second-hand item is paired with the digital code 20 and stored in the database 14; wherein, the digital code 20 can be but is not limited to a type such as Figure 2 the QR code shown. Preferably, the digital resume certificate generation module 13 is further connected to a digital encryption module 15, and the digital resume certificate 2 is encrypted by the digital encryption module 15 to prevent it from being tampered with or stolen without permission.

[0020] The feature image comparison module 12 is connected between the digital resume management module 11 and the database 14, and is built with AI (Artificial Intelligence) artificial intelligence identification technology to identify whether the "feature image" of the second-hand item stored in the database 14 matches the information provided by the user 100.

[0021] The database 14 is used to store the digital resume certificate 2 generated by the digital resume certificate generation module 13, as well as the feature image of the second-hand commodity, and provides access to the digital resume certificate generation module 13 and the feature image comparison module 12 to modify the resume of the digital resume certificate 2 and compare the images of the second-hand commodity.

[0022] When trading, the seller of the second-hand commodity sends the item to the administrator 101 of the trading system 1 of the present invention. The administrator 101 arranges an appraisal agency in advance to authenticate the authenticity of the item, and records the item through the aforementioned digital resume management module 11, including but not limited to the "item name", "brand name", "factory information (information on the purchase certificate)", "certification date", "feature description (including scratches, old marks... etc.)", and "certification result" and other resume items of the item, and uses an imaging device including but not limited to technologies such as photography and scanning to obtain the "feature images" of different angles of the item, and then transmits these information and data to the digital resume certificate generation module 13 to generate such as Figure 2 the digital resume certificate 2 shown, and a unique digital code 20 is assigned to the digital resume certificate 2, and the "feature image" of the item is paired with the digital code 20 and stored in the database 14.

[0023] When the buyer user 100 selects an item on the commodity trading platform module 10 of the present invention, the items recorded on the digital resume certificate 2 can be previewed as a reference for purchasing the commodity, so as to fully understand the actual situation of the second-hand commodity, and after the purchase process is determined to be completed, the encrypted digital resume certificate 2 is sent by the commodity trading platform module 10 to the buyer as a guarantee for the item.

[0024] The object holder of the buyer also holds a digital resume certificate 2 with a unique digital code 20. If the second-hand commodity object is to be resold to a third party, in addition to delivering the object to the third party for inspection, the authenticity and resume of the object can be pre-checked through the trading system 1 of the present invention. Please also refer to Figure 3 The flowchart for the second-hand object holder of the present invention to verify the authenticity of the commodity through the trading system 1. Here, whether it is the object holder of the original buyer (i.e., the current seller), or the object holder who has delivered the object to the third party (i.e., the current buyer), can enter the commodity trading platform module 10 of the trading system 1 of the present invention for operation. Enter the unique digital code 20 on the digital resume certificate 2 held to confirm whether the source of the second-hand object comes from this trading system 1. After confirmation, according to the requirements of the digital resume management module 11 of the system, "feature images" of each angle of the object are captured. Subsequently, the feature image comparison module 12 of the system compares whether the "feature images" input are consistent with the "feature images" originally stored in the database 14 for the object. If they are consistent, the digital resume management module 11 will send the comparison result to the commodity trading platform module 10 and feedback it to the platform user 100 (i.e., the current seller object holder, or the object holder who has delivered the object to the current buyer object holder) to authenticate the source and resume content of the second-hand object; conversely, if the feature image comparison module 12 compares the "feature images" input with the "feature images" originally stored in the database 14 for the object and they are not consistent, the digital resume management module 11 will require the platform user 100 to re-enter or re-capture the "feature images" of each angle of the second-hand object for comparison again, so as to prevent the seller from delivering a false object to the buyer, or providing the buyer with the ability to independently confirm the authenticity of the object through such a verification mechanism after receiving the second-hand object, in order to protect the rights and interests of the buyer.

[0025] In the above, the selected parts and angles of the "feature images" of second-hand objects of different styles are different. The AI (Artificial Intelligence) artificial intelligence recognition technology built into the feature image comparison module 12 needs to establish norms for a long time for the machine to learn. The following briefly describes the selectable feature images using luxury handbags such as Hermès and Chanel as examples.

[0026] I. For example, for Hermès handbags, please also refer to Figures 4a to 4d, the selection of characteristic images of the object may include but is not limited to: metal engraving on small ears, engraving depth, font type of the engraving, metal thickness, metal brightness, commonly used metal materials, workmanship of having four screws on the metal, etc. There are specific films that cannot be counterfeited. In addition, the glue on the leather can also be used as a characteristic; Hermès has internal code star engraving, horseshoes, numbers, boxes... etc. Among them, in 2021, it starts with Y and Z, followed by numbers to form the internal code, and the stitches are sewn manually by the master.

[0027] Second, for example, Chanel handbags. Among the objects of this brand, the most obvious selection of characteristic images is the texture performance of the leather, including the leather on the front of the object, the leather on the side, and the leather on the back. Even for the same style and the same color of leather, the performance will be different; furthermore, regarding the presentation of the internal code, the numbers of the internal code are naturally different, and the engraving thickness, font presentation, and font color of the internal code can all be the key to the characteristics.

[0028] Regarding the AI (Artificial Intelligence) artificial intelligence recognition technology, Yann LeCun proposed the Convolutional Neural Networks (CNN) model in 1998, which is widely used in the recognition of handwritten postal codes. Among them, the concept of the convolutional layer is the most common neural network architecture in deep learning, which is composed of a convolutional layer, a pooling layer, and a fully connected layer. Among them, the Convolutional Layer performs operations on the image matrix and the weight matrix to extract features with different weights; the Pooling Layer extracts the key features after the operation and has the function of filtering noise; the Fully Connected layer flattens the features after pooling and then conducts the final classification. In 2012, Alex Krizhevsky proposed a convolutional neural network model with a certain depth. The input layer is an RGB image. The special feature of this model is that there are seven hidden layers in the middle, among which the first five layers are convolutional layers and the last two layers are fully connected layers. Finally, the softmax function is used as the output layer. AlexNet participated in the ILSVRC 2012 competition and won the championship, with a huge gap from the second place. Since then, deep learning has become the well-known main development direction of AI.

[0029] The second place in the ILSVRC 2014 classification task competition was achieved by VGGNet proposed by the University of Oxford. VGGNet uses smaller convolutional kernels and deepens the network depth, reducing the image field of view (FOV) of the front convolutional layers. The smaller convolutional kernels and deeper network improve the efficiency of parameter training and obtain better model generalization performance, so it is more commonly used in tasks such as feature image selection.

[0030] Due to the characteristic of shared weights in the convolutional layers of the convolutional neural network, and the pooling layer can perform downsampling on features, reducing the sensitivity to deformation. Coupled with the data-augmented images mentioned above, it further makes the extracted features have strong invariance to rotation, displacement, scaling, etc. Using the model weights of VGGNet that have been trained on the ImageNet image dataset as the initialization parameters for pre-training, typically the most representative features will be formed on the network layers at the back end of the network architecture. For example, the front convolutional layer will extract the edge features of the object, the middle convolutional layer will obtain the corners and contours of the object, the back convolutional layer will obtain a certain part of the object, and the fully connected layer can make a global consideration of the features extracted by the convolution and refine the object features with higher representativeness. In the present invention, the pre-trained model of VGG16 will be used as the initial weight, and the image dataset after part augmentation will be input to further train VGG16. After the training is completed, the FC2 layer of VGG16 will be used as the extracted part features.

[0031] The AI (Artificial Intelligence) artificial intelligence identification technology is built into the feature image comparison module 12. By continuously deepening the learning of the "feature images" of a large number of objects stored in the database 14, it is used to make an immediate judgment on the "feature image" of the object input by the transaction buyer, and to provide the security of the second-hand commodity transaction through the system by scientific methods to ensure the result of the online transaction.

[0032] The above-described embodiments are only preferred embodiments cited to fully illustrate the present invention, and the protection scope of the present invention is not limited thereto. Equivalent substitutions or transformations made by those skilled in the art on the basis of the present invention, such as defining the items recorded in the "digital resume certificate" or the angle or content of the selection of the "feature image" of the object, are all within the protection scope of the present invention. The protection scope of the present invention is subject to the claims.

Claims

1. A second-hand goods trading system for trading second-hand items, characterized in that, Including: A commodity trading platform module, a digital resume management module, a feature image comparison module, a digital resume certificate generation module, and a database. Information and data are exchanged bidirectionally between these modules through connections. Among them, The digital resume management module is communicatively connected to the commodity trading platform module, used to establish a resume for the second-hand item in the transaction, and obtain the feature image of the second-hand item through an imaging device. The digital resume certificate generation module is connected between the digital resume management module and the database, used to generate a digital resume certificate for the resume of the second-hand item, and assign a unique digital code to the digital resume certificate. The feature image selected from the second-hand item is paired with this digital code and stored in the database together. The feature image comparison module is connected between the digital resume management module and the database, and built-in with AI artificial intelligence recognition technology, used to compare whether the feature image of the second-hand item stored in the database matches the information provided by the item holder. The database is used to store the digital resume certificate and the feature image of the second-hand item. Thus, the item holder inputs the unique digital code on the digital resume certificate into the system to confirm whether the source of the second-hand item comes from this trading system. After confirmation, the item holder then captures and uploads the feature images of each angle of the item according to the requirements of the digital resume management module. The feature image comparison module compares whether each of the uploaded feature images is consistent with the existing images paired with the unique digital code stored in the database.

2. The second-hand commodity trading system according to claim 1, characterized in that, The unique digital code on the digital resume certificate is a QR code.

3. The second-hand commodity trading system according to claim 1, characterized in that, If the comparison between each of the uploaded feature images and the existing images paired with the unique digital code stored in the database is consistent, the digital resume management module will send the comparison result to the commodity trading platform module and feedback it to the user of the platform; otherwise, if it is inconsistent, the digital resume management module will feedback the information of inconsistent comparison to the user of the platform and require the user to re-enter or re-capture the feature images of each angle of the second-hand item for comparison again.

4. The second-hand commodity trading system according to claim 1, wherein The resume recorded on the digital resume certificate at least includes the factory information, certification date, and feature description and / or certification result items of the second-hand item.

5. The second-hand commodity trading system according to any one of claims 1 to 4, characterized in that The digital resume certificate generation module is also connected to a digital encryption module, and the digital resume certificate is encrypted through this digital encryption module.

6. The second-hand commodity trading system according to any one of claims 1 to 4, characterized in that The feature image is a texture of leather.

7. The second-hand goods trading system according to any one of claims 1 to 4, characterized in that, The feature image is a metal engraving.

Images