CAJ | 학술논문

运动估计是图像超分辨率复原重要的步骤，直接影响最终的复原结果。针对运动估计中特征点匹配的问题，提出运用改进加速鲁棒特征（SURF）算法对图像的特征点进行匹配。再使用凸集投影（POCS）算法重建图像序列，最终得到复原的高分辨率图像。所提出的基于改进SURF 算法的POCS 算法对比其他图像复原算法，得到了峰值信噪比值较高、均方误差较低的复原图像，说明该算法的有效性。
운동고계시도상초분변솔복원중요적보취，직접영향최종적복원결과。침대운동고계중특정점필배적문제，제출운용개진가속로봉특정（SURF）산법대도상적특정점진행필배。재사용철집투영（POCS）산법중건도상서렬，최종득도복원적고분변솔도상。소제출적기우개진SURF 산법적POCS 산법대비기타도상복원산법，득도료봉치신조비치교고、균방오차교저적복원도상，설명해산법적유효성。
Motion estimation is an important step of image restoration technique , which directly affects the result of the final restoration. To solve the problem of feature point matching in motion estimation , this paper used improved speeded up robust features (SURF) algorithm to match the feature points of the image. Using the projection onto convex set (POCS) reconstruction algorithm for image sequences, finally it got the high resolution image restoration. Compared with other image restored algorithm, the POCS algorithm based on improved SURF algorithm gets higher peak signal-to-noise ratio and lower mean square error restored image , and the experimental data has demonstrated the effectiveness of the proposed method.