Fuzzy photo detection method based on deep learning

The invention discloses a fuzzy photo detection method based on deep learning, and the method comprises the steps: collecting photos shot by a photographer and a user, and obtaining the data of an original sampling image; performing data preprocessing on the obtained original sampling image data; obtaining local features and global features by using the extended full convolutional neural network; using a convolutional neural network for feature extraction; carrying out joint training by utilizing the convolutional neural network, and carrying out end-to-end feature learning on the preprocessed data; classifying the learned features to obtain a deep learning model; and utilizing the deep learning model to automatically detect the photographer and the photo actually shot by the user.The image blurring type recognition rate can be effectively improved, higher-precision blurring image type recognition can be realized, and the average accuracy rate is improved to 89.31%. 本发明公开了一种基于深度学习的模糊照片检测方法，包括：采集摄影师与用户真实拍摄的照片，