期刊文献+
共找到2篇文章
< 1 >
每页显示 20 50 100
自适应的Haar型LBP纹理特征提取算法研究 被引量:2
1
作者 刘天时 肖敏敏 李湘眷 《计算机工程与科学》 CSCD 北大核心 2015年第7期1381-1386,共6页
在提取纹理图像的Haar型LBP特征中,人为设定的判断阈值主观性强、局部性差,导致提取的纹理细节和边缘模糊、纹理图像的局部性易被忽略。为此,提出了一种自适应的Haar型LBP纹理特征提取算法。该算法在二值化Haar型特征时引入高斯加权矩阵... 在提取纹理图像的Haar型LBP特征中,人为设定的判断阈值主观性强、局部性差,导致提取的纹理细节和边缘模糊、纹理图像的局部性易被忽略。为此,提出了一种自适应的Haar型LBP纹理特征提取算法。该算法在二值化Haar型特征时引入高斯加权矩阵,以此获得客观、符合纹理图像局部特征的自适应判断阈值和Haar型LBP特征。实验结果表明,该算法能够有效地避免人为设定阈值对纹理特征的影响,可以准确地描述图像的纹理特征,Brodatz标准纹理库分类的正确率也得到了进一步的提高。 展开更多
关键词 纹理特征 HAAR特征 LBP 高斯加权矩阵
下载PDF
Gaussian Weighted Trajectory Method Ⅳ: No Rainbow Effect in Practice
2
作者 L. Bonnet 《Chinese Journal of Chemical Physics》 SCIE CAS CSCD 2009年第2期210-214,共5页
The Gaussian weighted trajectory method (GWTM) is a practical implementation of classical S matrix theory (CSMT) in the random phase approximation, CSMT being the first and simplest semi-classical approach of mole... The Gaussian weighted trajectory method (GWTM) is a practical implementation of classical S matrix theory (CSMT) in the random phase approximation, CSMT being the first and simplest semi-classical approach of molecular collisions, developped in the early seventies. Though very close in spirit to the purely classical description, GWTM accounts to some extent for the quantization of the different degrees-of-freedom involved in the processes. While CSMT may give diverging final state distributions, in relation to the rainbow effect of elastic scattering theory, GWTM has never led to such a mathematical catastrophe. The goal of the present note is to explain this finding. 展开更多
关键词 Gaussian weighted trajectory method Classical S matrix theory Rainbow effect
下载PDF
上一页 1 下一页 到第
使用帮助 返回顶部