Image super-resolution reconstruction (SRR) technology can improve image spatial resolution on the basis of existing imaging systems. Projection onto convex sets (POCS) is one of the main methods of SRR. POCS is improved in this paper, which mainly includes two aspects: (1) Improving the quality of the initial estimate by kernel regression interpolation;(2) Replacing Gaussian kernel with steering kernel in the point spread function (PSF) to reduce edge ringing artifacts of the reconstructed image. The proposed algorithm is simulated. The results show that the edges of the reconstructed image have been improved greatly.
参考文献
相似文献
引证文献
引用本文
谢寒,赵勋杰,李成金,张雪松.一种改善图像边缘效果的POCS超分辨率重建方法[J].激光与红外,2014,44(6):697~701 XIE Han, ZHAO Xun-jie, LI Cheng-jin, ZHANG Xue-song. POCS super-resolution reconstruction method to improve the quality of image edge[J]. LASER & INFRARED,2014,44(6):697~701