增强壁画图像艺术性的超分辨率重建

Super-resolution reconstruction of enhanced artistic mural images

针对现有壁画图像中分辨率低、图像细节粗糙导致壁画图像艺术价值降低等问题,提出一种增强艺术性壁画图像的超分辨率重建算法优化壁画图像。以生成对抗网络为框架,使用生成网络输出重建的高分辨率图像,在判别网络中提取壁画图像的特征,用于判断生成网络的输出是否为真实高分辨率壁画图像。采用迁移学习思想、去除批标准化、将网络模型分阶段构建的方法实现更好的壁画图像优化。与现有几种图像超分辨率算法对比,在客观实验指标上PSNR值平均提高了1.2 dB-3.3 dB,SSIM值增加了0.04-0.13;在主观评分方面也优于其它算法,验证了该算法在壁画图像的超分辨率重建上具有有效性和卓越型。

Aiming at the problems of low resolution and rough image details in the existing mural images,which reduce the artistic value of mural images,a super-resolution reconstruction algorithm to enhance artistic mural images was proposed to optimize mural images.The generative adversarial networks were used as the framework.The generated network was used to output the reconstructed high-resolution image,and the features of the mural image in the discriminant network were extracted to determine whether the output of the generated network was a true high-resolution mural image.Transfer learning ideas,remove batch normalization,and constructing the network model in stages were adopted to achieve better mural image optimization.Compared with several existing image super-resolution algorithms,the PSNR value is increased by 1.2 dB-3.3 dB on average,and the SSIM value is increased by 0.04-0.13.It is also superior to other algorithms in subjective scoring.It verifies that the algorithm is effective and superior in super-resolution reconstruction of mural images.