"Are there any sisters who can help identify the Gucci horse mongoose you bought with the second hand?" "The authenticity of the ancient flowers in LV, friends, please help me see." Searching for the keywords related to Ershe on social platforms, you can see the "Question" posts

"Are there any sisters who can help identify the Gucci horse mongoose you bought with second-hand buy?" "The authenticity of the ancient flowers in LV, friends, please help me take a look." Searching for the keywords related to Ershe on social platforms, you can see the "Question" posts posted by many consumers. It is not difficult to find that identification has become a key issue that Ershe cannot avoid. At the same time, with the continuous development of technology and the continuous improvement of counterfeiting technology, "high imitation products" and "S-class" counterfeit goods emerge one after another, greatly increasing the difficulty of identification. Under this circumstance, introducing big data has become a new development direction in the field of luxury goods identification.

Currently, the application of big data in the luxury goods appraisal industry is mainly presented in AI appraisal and physical appraisal assisted by big data.

AI authentication uses consumers to upload product pictures and use algorithms to compare the identification points to obtain results. For consumers, they can not only get results instantly, but also save a lot of time and cost. However, AI identification also has certain limitations. In the absence of manual comprehensive and comprehensive judgment of products, factors such as photo quality, different identification points of the product, new and old quality will cause different degrees of judgment interference to the algorithm results. The physical identification of

, which is the opposite, greatly affirms the value of manual identification and explores various possibilities assisted by big data on it.

"We have been collecting luxury goods data since 2016 and have officially put into use the database in 2019," said Xu Zidi, director of Zhongshu Inspection, Testing and Certification Center. As a head third-party appraisal agency that insists on physical identification, Zhongshu detects and collects leather, metal and other data of luxury goods luggage through X-ray technology and ultra-deep field 3D modeling instruments and other high-quality instruments. To this day, its exclusive luxury goods database has more than 30 million samples of imitation samples.

database is essentially just a brain that stores massive data, and Zhongshu took the lead in combining the database with high-precision instruments to allow the brain to operate more efficiently and cross-cut, thereby greatly improving the accuracy and efficiency of identification.

For example, by detecting Chanel brooch by hand-held alloy analyzer, you can clearly know the metal elements and their content of this necklace. Just compare the detected data with the authentic data in the database, which can become an important basis for identification results. In addition to the handheld alloy analyzer , Zhongshu has currently put into use dozens of high-precision and cutting-edge instrument categories, all of which have been certified by the US EU and national metrology.

Xu Zidi said that the appraiser needs the appraiser's "back-to-back" appraisal results to be consistent with the results of the instrument appraisal in order to issue a genuine traceability signature to ensure that the appraisal results are more accurate. In addition, the combination of instruments and databases can also help filter out some "fake" products, making the entire appraisal process more efficient. The emergence of

big data has broken the traditional appraisal model of judging by the personal experience of appraisers, and extended the method of combining "big data + instruments". In the future, where the luxury appraisal industry under the support of technology will go, it will take time to verify. "Zhongrui still accounts for 70% of manual identification, and 30% of big data and instrument identification," said Xu Zidi. "This proportion may gradually change in the future."