CAJ | 학술논문

常规的超分辨复原方法需要预知退化图像的点扩展函数,但实际应用中许多退化图像的点扩展函数是未知的,因此提出一种新的超分辨率图像盲复原算法,在点扩展函数未知或不确知的情况下对图像进行恢复.该联合插值-恢复的超分辨率图像盲复原方法,利用多信道盲复原估计未知的点扩展函数,迭代运用帧间相似性确定模糊特性,同时结合超分辨率方法得到高分辨率图像.实验结果表明,该算法能有效地实现超分辨率图像的盲复原.
상규적초분변복원방법수요예지퇴화도상적점확전함수,단실제응용중허다퇴화도상적점확전함수시미지적,인차제출일충신적초분변솔도상맹복원산법,재점확전함수미지혹불학지적정황하대도상진행회복.해연합삽치-회복적초분변솔도상맹복원방법,이용다신도맹복원고계미지적점확전함수,질대운용정간상사성학정모호특성,동시결합초분변솔방법득도고분변솔도상.실험결과표명,해산법능유효지실현초분변솔도상적맹복원.
The primary motivation for developing the blind algorithm is due to the observation that most existing Super-Resolution (SR) methods assume the point spread function are known in advance, which may not be true in many real-world applications. In view of this, this paper presented a novel blind SR algorithm to enhance image resolution without knowledge of the point spread function. The authors formulated the SR problem into a new framework of joint interpolation-restoration scheme. The main contribution of this method was the development of multi-channel blind restoration to estimate the unknown point spread function, and its integration into the SR scheme to render High-Resolution (HR) images. The blurs were identified iteratively with the utilization of inter-frame blur similarity. Experimental results show that the new algorithm is effective in performing blind SR image reconstruction.