No-reference image quality evaluation method based on deep learning

The invention discloses a non-reference image quality evaluation method based on deep learning, and the method comprises the steps: extracting the multi-level features of a distorted image from a data set image of which the image quality is to be predicted through a bottom texture feature extraction network, a low-level contour feature extraction network and a high-level global semantic feature extraction network from shallow to deep; secondly, fusing the extracted multi-level features by adopting a feature fusion method based on an attention mechanism so as to enhance the expression ability of the multi-level features to the image content; and finally, according to the extracted multi-level features and a feature fusion method, designing a loss function adapted to the overall model so as to obtain an optimal quality evaluation prediction score. According to the method, the problem that the image quality score cannot be effectively evaluated when a traditional method faces multiple distortion types is solved,