Guided filter‐based multi‐scale super‐resolution reconstruction. Issue 2 (11th May 2020)

• Record Type: Journal Article
• Authors: Feng, Xiaomei; Li, Jinjiang; Hua, Zhen
• Is Part Of: CAAI transactions on intelligence technology. Volume 5:Issue 2(2020)
• Page Start: 128
• Page End: 140
• Publication Date: 2020-05-11
• Subjects: image resolution -- image texture -- learning (artificial intelligence) -- image reconstruction -- image filtering; high‐resolution image -- low‐resolution image loss -- super‐resolution reconstruction effect -- guided filter‐based multiscale super‐resolution reconstruction -- learning‐based super‐resolution reconstruction method -- multiscale super‐resolution reconstruction network -- end‐to‐end super‐resolution reconstruction task -- multiscale super‐resolution reconstruction method -- guided image filtering -- multiscale super‐resolution network -- guide image filter -- newly generated image -- super‐resolution reconstruction scheme; B0290F Interpolation and function approximation (numerical analysis) -- B6135 Optical, image and video signal processing -- B6140B Filtering methods in signal processing -- C5260B Computer vision and image processing techniques -- C6170K Knowledge engineering techniques; Artificial intelligence -- Periodicals; Computer science -- Periodicals; Artificial intelligence; Computer science; Electronic journals; Periodicals; 006.305
• DOI: 10.1049/trit.2019.0065
• Languages: English
• ISSNs: 2468-6557

Abstract

Abstract : The learning‐based super‐resolution reconstruction method inputs a low‐resolution image into a network, and learns a non‐linear mapping relationship between low‐resolution and high‐resolution through the network. In this study, the multi‐scale super‐resolution reconstruction network is used to fuse the effective features of different scale images, and the non‐linear mapping between low resolution and high resolution is studied from coarse to fine to realise the end‐to‐end super‐resolution reconstruction task. The loss of some features of the low‐resolution image will negatively affect the quality of the reconstructed image. To solve the problem of incomplete image features in low‐resolution, this study adopts the multi‐scale super‐resolution reconstruction method based on guided image filtering. The high‐resolution image reconstructed by the multi‐scale super‐resolution network and the real high‐resolution image are merged by the guide image filter to generate a new image, and the newly generated image is used for secondary training of the multi‐scale super‐resolution reconstruction network. The newly generated image effectively compensates for the details and texture information lost in the low‐resolution image, thereby improving the effect of the super‐resolution reconstructed image.Compared with the existing super‐resolution reconstruction scheme, the accuracy and speed of super‐resolution reconstruction are improved.