Feng Hesen, Ma Lihong, Tian Jing
School of Electronics & Information Engineering, South China University of Technology, Guangzhou 510640, China.
National Research Center for Mobile Ultrasonic Detection, Guangzhou 510640, China.
Sensors (Basel). 2022 Jun 1;22(11):4231. doi: 10.3390/s22114231.
Image super-resolution aims to reconstruct a high-resolution image from its low-resolution counterparts. Conventional image super-resolution approaches share the same spatial convolution kernel for the whole image in the upscaling modules, which neglect the specificity of content information in different positions of the image. In view of this, this paper proposes a regularized pattern method to represent spatially variant structural features in an image and further exploits a dynamic convolution kernel generation method to match the regularized pattern and improve image reconstruction performance. To be more specific, first, the proposed approach extracts features from low-resolution images using a self-organizing feature mapping network to construct regularized patterns (RP), which describe different contents at different locations. Second, the meta-learning mechanism based on the regularized pattern predicts the weights of the convolution kernels that match the regularized pattern for each different location; therefore, it generates different upscaling functions for images with different content. Extensive experiments are conducted using the benchmark datasets Set5, Set14, B100, Urban100, and Manga109 to demonstrate that the proposed approach outperforms the state-of-the-art super-resolution approaches in terms of both PSNR and SSIM performance.
图像超分辨率旨在从低分辨率图像重建高分辨率图像。传统的图像超分辨率方法在放大模块中对整个图像使用相同的空间卷积核，这忽略了图像不同位置内容信息的特异性。鉴于此，本文提出一种正则化模式方法来表示图像中的空间变化结构特征，并进一步开发一种动态卷积核生成方法来匹配正则化模式并提高图像重建性能。具体而言，首先，所提出的方法使用自组织特征映射网络从低分辨率图像中提取特征以构建正则化模式（RP），该模式描述不同位置的不同内容。其次，基于正则化模式的元学习机制预测与每个不同位置的正则化模式相匹配的卷积核的权重；因此，它为具有不同内容的图像生成不同的放大函数。使用基准数据集Set5、Set14、B100、Urban100和Manga109进行了大量实验，以证明所提出的方法在PSNR和SSIM性能方面均优于现有最先进的超分辨率方法。
