A Store Entity Identification Method Based on Deep Learning

Abstract

In recent years, the development of deep learning has led to unprecedented advances in computer vision, making it possible to use artificial intelligence to identify shop names. This paper proposes a store entity identification system based on deep learning, which consists of three modules. The text detection module adopts Cascade Mask R-CNN. The text recognition module adopts Attention LSTM. The named entity recognition module adopts Bert-BiLSTM-CRF. It can recognize all the text information in the picture and extract the shop name accurately. The final score can reach 91.12%. Our research saves the time of traditional manual extraction of the shop name in the picture, and provides technical support for the intelligent development of the shop management system.