Method for processing commodity in picture

The invention discloses a method for processing commodities in a picture, which adopts a machine vision technology based on deep learning to identify, segment and automatically classify commodity objects in the picture, and highlights and marks commodity position information, range information and t...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: ZHENG YIMOU, GUO NAICHENG, JI QING, ZHONG YAOZHU
Format: Patent
Sprache:chi ; eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The invention discloses a method for processing commodities in a picture, which adopts a machine vision technology based on deep learning to identify, segment and automatically classify commodity objects in the picture, and highlights and marks commodity position information, range information and the like. Commodity SKU objectification management is carried out on commodities contained in any picture, and online ordering processing can be further carried out on each identified and labeled commodity SKU object. According to the invention, commodity objects in a large number of live pictures are firstly separated, so that a purchasing end user selects commodities as personally going to a real spot warehouse, an interactive transaction mode that the purchasing end user selects the commodities one by one for batch purchasing is provided, and real online reappearing is realized at any place. The interactive experience of wholesale purchase can be carried out on a large-scale centralized warehouse site, meanwhile,