Image super-resolution reconstruction of multi-scale deep feature distillation

Aiming at the problem that existing super-resolution reconstruction algorithms were difficult to fully utilize multi-scale information and deep features of images，an image super-resolution reconstruction method based on multi-scale deep feature distillation（MSDFDN）was proposed. First，ConvNeXt convo⁃ lution was used to replace traditional convolution layers，increasing network depth with lower computation⁃ al cost to improve performance. Second，a multi-scale deep feature distillation module was designed. By constructing ConvNeXt convolution layers of different scales and combining them with a residual feature distillation mechanism，multi-scale deep features in residual blocks were extracted while bypassing rich low-frequency information. Finally，an attention mechanism was introduced at the end of the module to adaptively weight extracted features，enabling the network to focus more on high-frequency information. Compared with other advanced lightweight super-resolution reconstruction algorithms on benchmark datas⁃ ets and the self-built PDC bit composite dataset，the peak signal-to-noise ratio and structural similarity quantitative metrics of images obtained by this method showed improvement. Especially on the Urban100 dataset with more detailed information，the peak signal-to-noise ratio of the four-fold reconstructed image reaches 26. 49 dB，and the structural similarity reaches 0. 797 6. Experimental results show that the pro⁃ posed method has better objective and subjective measurement results. © 2025 Chinese Academy of Sciences. All rights reserved.