计算机应用研究
計算機應用研究
계산궤응용연구
APPLICATION RESEARCH OF COMPUTERS
2015年
5期
1532-1535
,共4页
图像复原%图像细节%非局部全变差%局部全变差
圖像複原%圖像細節%非跼部全變差%跼部全變差
도상복원%도상세절%비국부전변차%국부전변차
image restoration%image details%nonlocal total variation%local total variation
为了更好地复原图像的细节，提出了一种结合局部与非局部的图像复原方法。将图像中的细节准确地提取出来，对提取的细节进行非局部全变差约束，同时对剩下的图像成分进行局部全变差约束。提出的方法很好地结合了非局部全变差和局部全变差的优点，实现了图像细节更好的复原。实验结果表明，提出的方法与近几年的一些较好的图像复原方法相比，不仅主观的视觉效果得到了明显的改进，而且客观的峰值信噪比也增加了0．11～2．28 dB。
為瞭更好地複原圖像的細節，提齣瞭一種結閤跼部與非跼部的圖像複原方法。將圖像中的細節準確地提取齣來，對提取的細節進行非跼部全變差約束，同時對剩下的圖像成分進行跼部全變差約束。提齣的方法很好地結閤瞭非跼部全變差和跼部全變差的優點，實現瞭圖像細節更好的複原。實驗結果錶明，提齣的方法與近幾年的一些較好的圖像複原方法相比，不僅主觀的視覺效果得到瞭明顯的改進，而且客觀的峰值信譟比也增加瞭0．11～2．28 dB。
위료경호지복원도상적세절，제출료일충결합국부여비국부적도상복원방법。장도상중적세절준학지제취출래，대제취적세절진행비국부전변차약속，동시대잉하적도상성분진행국부전변차약속。제출적방법흔호지결합료비국부전변차화국부전변차적우점，실현료도상세절경호적복원。실험결과표명，제출적방법여근궤년적일사교호적도상복원방법상비，불부주관적시각효과득도료명현적개진，이차객관적봉치신조비야증가료0．11～2．28 dB。
In order to recover the details better, this paper proposed a new image restoration method, which combined the lo-cal total variation (LTV) and the nonlocal total variation (NLTV) models.First, it extracted the details accurately from the image.Then, the method applied the NLTV model only on the details, and applied the LTV model only on the residual image components.The proposed method, which used the advantages of both the LTV and NLTV models, could recover the image details better.A large number of experimental results indicate that the proposed method outperforms several recent image resto-ration methods, not only the subjective vision has the betterment obviously, but also the PSNR improves between 0.11 dB and 2.28 dB.
