Wavelet-domain least squares based image superresolution | |
Zhao Shubin; Zhang Peng; Peng Silong | |
2003 | |
会议名称 | Proceedings of the Third International Conference on Wavelet Analysis and Its Applications (WAA) |
页码 | pp 269-274 |
会议日期 | 2003/5/29-2003/5/31 |
会议地点 | 中国Chongqing China |
摘要 | This paper proposes a wavelet-domain least-squares based algorithm for image super-resolution. Based on the fact of the self-similarity of multscale edges it is possible to predict the three subband wavelet coefficients. Once the wavelet coefficients are obtained high-resolution images can be reconstructed using inverse DWT. Because the algorithm properly preserves the local regularity around edges the induced images are of high visual quality. Besides since we only need to predict the wavelet coefficients near edges the algorithm is computationally efficient. Finally quantitative error analyses are provided and several images are shown for subjective assessment. |
关键词 | 无 |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/12900 |
专题 | 智能制造技术与系统研究中心_多维数据分析 |
通讯作者 | Zhao Shubin |
推荐引用方式 GB/T 7714 | Zhao Shubin,Zhang Peng,Peng Silong. Wavelet-domain least squares based image superresolution[C],2003:pp 269-274. |
条目包含的文件 | 条目无相关文件。 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论