並列摘要
This study used a deep learning convolutional neural network to classify large-scale image in commodity datasets with a certain degree of effectiveness. It is much more precise than the traditional way of artificially customizing the features of objects. Deep learning is a part of machine learning technology. In recent years, deep learning of convolutional neural network technology has led in many fields, such as image classification, speech recognition, natural language processing, and so on. Convolutional neural networks are a class of the neural networks, but it shared the weights, which greatly reduce network parameters and reduce many over-fitting situations than other types of neural networks. So this kind of neural networks are easier to train than other neural networks.