多尺度關(guān)鍵信息融合的輕量級(jí)圖像超分辨重建

摘要： 針對(duì)基于卷積神經(jīng)網(wǎng)絡(luò)(convolutional neural network, CNN)的圖像超分辨重建(super-resolution, SR)模型存在特征提取不充分、網(wǎng)絡(luò)太深導(dǎo)致的參數(shù)量大以及冗余信息對(duì)網(wǎng)絡(luò)最終重建性能影響等問(wèn)題，設(shè)計(jì)了一種輕量級(jí)密集連接圖像超分辨網(wǎng)絡(luò)(lightweight densely connected image super-resolutio...