Image Super-Resolution Using Multi-Scale Space Feature and Deformable Convolutional Network

Abstract

Recent researches on image super-resolution (SR) have achieved great progressing with the great development of convolutional neural networks (CNNs). However, existing CNNs usually adopt fixed filter structures and the convolutions just rely on the local information contained in the fixed receptive field. Above phenomena prevent high-level convolution layers from encoding semantics over spatial locations and largely limits the learning capacity of CNNs. What's more, many methods simply used a single-size feature map and failed to consider the spatial information, thereby these results also are unsatisfactory. To address these problems, in this paper, a network with multi-scale space features and deformable convolutional (MulSSD) is presented to further improve the reconstruction accuracy. Specifically, a multi-scale space features compressed block containing the deformable convolutional layer is proposed, which can augment the spatial sampling locations and incorporate the multi-scale space compression features and adaptively adjust the sampling grid and receptive fields. In addition, the design of symmetrical combinations make the information can be smoothly propagated through multiple channels during the training, which effectively improves the training efficiency. Extensive experiments on benchmark datasets validate that the proposed method achieves outperforming quantitative and qualitative performance. And the experimental results also proved that our proposed MulSSD can reconstruct high-quality high-resolution (HR) images at a relatively fast speed and outperform other methods by a large margin.

Psoriasis severity classification based on adaptive multi-scale features for multi-severity disease

“…In convolution, a filter of a fixed size and shape extracts features that have limited spatial information, i.e., local information in a fixed receptive field. The use of single-size filters is also limited in the extraction of various hierarchical representations of convolutional networks 39 , 40 . In addition, for accurate evaluation of psoriasis diseases, including multiple-severity diseases, severity features such as texture information of scaling and semantic information on the representative region should be simultaneously extracted.…”

Cross-domain fault diagnosis of rotating machinery based on graph feature extraction

“…Compared with the two-dimensional network structure, the structure of the graph is unstable. Therefore, the convolution kernel cannot be used to deal with the points of graph in the spatial domain directly, and the points in the graph need to be transformed into the coordinates of the spectral domain by Fourier transform [29][30][31]. Convolution in the spatial domain is equivalent to the product in the spectral domain.…”