This paper. details experiments undertaken in the UK Coastal Research Facility (CRF)at Hy draulies Research (HR), Wallingford, on transformation and run-up of wave trains. The purpose of these experiments is to provid...This paper. details experiments undertaken in the UK Coastal Research Facility (CRF)at Hy draulies Research (HR), Wallingford, on transformation and run-up of wave trains. The purpose of these experiments is to provide verification data for numerical models of wave transformation in shoaling. surf and swash zones. This is the kind of data ih:lt flume experiments are unable to provide, and is collected in the highly controlled environment of CRF where extrinsic factors present in the field are not an issue. The experiments concerning wave trains are undertaken by use of existing wave generation software, and the run-up measurements are made with large experimental run-up gauges.展开更多
针对传统的基于线性回归模型插值算法不能对变化剧烈的边缘进行有效插值的问题,该文提出一种基于正则化的边缘定向插值算法。算法主要分为两部分:参数估计部分与数据估计部分。在参数估计部分,为了更加准确地描述图像局部结构,把已估计...针对传统的基于线性回归模型插值算法不能对变化剧烈的边缘进行有效插值的问题,该文提出一种基于正则化的边缘定向插值算法。算法主要分为两部分:参数估计部分与数据估计部分。在参数估计部分,为了更加准确地描述图像局部结构,把已估计的高分辨率像素作为训练像素的一部分,用以进行回归模型参数的估计。在数据估计部分,引入像素平滑方向作为正则化项,以降低参数的误估计引起的数据估计偏差。实验结果表明,该算法能很好地保持图像的边缘特征,尤其在变化比较剧烈的边缘区域;与双三次插值算法及基于正则化的局部线性回归插值算法(Regularized Local Linear Regression,RLLR)相比,该算法能取得更好的视觉效果及较高的PSNR值。展开更多
基金This project was supported by the Flood and Coastal Defense Commission of UK(FD0204)the National Natural Science Foundation of China(59809001)
文摘This paper. details experiments undertaken in the UK Coastal Research Facility (CRF)at Hy draulies Research (HR), Wallingford, on transformation and run-up of wave trains. The purpose of these experiments is to provide verification data for numerical models of wave transformation in shoaling. surf and swash zones. This is the kind of data ih:lt flume experiments are unable to provide, and is collected in the highly controlled environment of CRF where extrinsic factors present in the field are not an issue. The experiments concerning wave trains are undertaken by use of existing wave generation software, and the run-up measurements are made with large experimental run-up gauges.
文摘针对传统的基于线性回归模型插值算法不能对变化剧烈的边缘进行有效插值的问题,该文提出一种基于正则化的边缘定向插值算法。算法主要分为两部分:参数估计部分与数据估计部分。在参数估计部分,为了更加准确地描述图像局部结构,把已估计的高分辨率像素作为训练像素的一部分,用以进行回归模型参数的估计。在数据估计部分,引入像素平滑方向作为正则化项,以降低参数的误估计引起的数据估计偏差。实验结果表明,该算法能很好地保持图像的边缘特征,尤其在变化比较剧烈的边缘区域;与双三次插值算法及基于正则化的局部线性回归插值算法(Regularized Local Linear Regression,RLLR)相比,该算法能取得更好的视觉效果及较高的PSNR值。