自适应超分辨率图像重构

超分辨率图像重构是利用关于同一场景的多帧低分辨率图像重构出一幅具有更高分辨率图像的过程。已有的MAP超分辨率图像重构算法由于没有对图像中的不同区域分别处理，所以无法在重构出清晰的边界和更多细节的同时抑制边缘振荡效应的产生。提出了针对图像中的不同区域进行自适应处理的超分辨率图像重构算法，与已有的算法相比，新算法所重构出的图像质量，不论在定量指标上还是人眼主观评判上都有了很大的提高。

Adaptive super-resolution image reconstruction

Super-resolution image reconstruction produces a high-resolution image from a set of low-resolution images. Due to the disability of handling different regions of the image, previously published MAP super-resolution image reconstruction techniques generally produced severe blurred image or image with severe ringing artifacts. A new adaptive super-resolution image reconstruction algorithm was proposed, which could treat different regions of the image adaptively. Experimental tests showed that, with respect to normal MAP algorithms, a decent increases of reconstruction quality, both visually and quantitatively, is achieved by the proposed algorithm.