The topic of this work is "taking Chinese painting as the theme, training the style of a specific picture based on deep learning convolutional neural network VGG-16, and then applying the corresponding style to the development of image processing software for any picture". The name of the work is "painting bones". the main functions of this APP we designed are: picture processing includes national style filter, three primary colors, brightness and saturation adjustment, popular science of Chinese style writers and works, and community content release and display. We believe that the use of convolutional neural network to VGG-16 this trained image classification model can apply the style of some artistic treasures to their own works, to achieve the role of artistic style spread, but also to improve the social awareness of artistic works, reflecting the main requirements of creativity to improve quality.