Research on Image Recognition of Supermarket Commodity Based on Convolutional Neural Network
Yanmei Liu, Ben Liu, Yuda Chen · 2019
For the actual shopping scenes in supermarkets, there will be more complex background environment, more diverse detection objects, and more changeable actual scenes, so the traditional image recognition network cannot meet the existing needs. In this paper, Convolutional Neural Network technology and YOLOv3 target detection algorithm are adopted to realize the recognition and detection of target objects in images, automated analysis and processing of these product image data with computer vision algorithms. The convolution neural network and COCO data set are used to train the neural network. Finally, the well-trained network is tested. Tested, the effect is good.