Garment image classification model establishing method based on deep learning

The invention discloses a garment image classification model establishing method based on deep learning. A VGG-16 network is adopted, a clothing image classification model is established, attribute labeling is performed on a large number of unlabeled clothing images by using the model, and labeled clothing data sets are generated in batches, so that the situation that a large number of unlabeled clothing image data cannot be effectively utilized is changed; the generated clothing data set is applied to a personalized clothing retrieval system, so that an idea is provided for the application ofthe labeled clothing images, the application value of the clothing image classification model in clothing e-commerce is displayed, and powerful support is further provided for the development of clothing e-commerce. 本发明公开了一种基于深度学习的服装图像分类模型建立方法，采用VGG-16网络，建立了服装图像分类模型模型，运用该模型对大量未标注服装图像进行属性标注并批量生成已标注服装数据集，改变了大量未标注服装图像数据无法被有效利用的状况。并将生成的服装数据集应用于个性化服装检索系统，为已标注的服装图像的应用提供了思路，展示了服装图像分类模型在服装类电商的应用价值，进一步为服装类电子商务的发展提供有力支持。