Research on garment image classification and detection algorithm based on improved deep learning

With the development of e-commerce economy, online shopping has become an indispensable part of people's life. Therefore, the study of clothing image classification is of great significance to realize the effective operation of online clothing change and clothing recommendation system. This paper uses two classical deep learning algorithms: LeNet and ResNet to construct the clothing image classification model and realize the classification and detection of. At the same time, on the basis of Yolov5, combined with ResNet to improve its backbone network and further improve the accuracy of clothing image classification and detection. Experiments show that on the FashionMnist dataset, the accuracy and precision of LeNet and ResNet are about 93%, the accuracy and precision of improved Yolov5 are more than 98%, which is much higher than the former.