摘要
以纸张为研究对像,利用计算机视觉方法,在分析了标准匀度检测方法的基础上,采用类金字塔结构方差给出了改进的梯度指数纹理分析算法,它既体现了微观匀度特性,又体现了宏观匀度特性.实验结果表明,该方法对于分析纸张匀度是可行的,并具有较高的鲁棒性能.
An advanced gradient index texture analysis algorithm based on computer vision methods is proposed with paper as the research object, which uses pyramid-like structure squared variance based on the standard evenness detecting method. This method has both micro evenness characteristics and the macro ones. Experimental results indicate that the method is feasible for analyzing paper evenness and is of higher robustness.
出处
《控制与决策》
EI
CSCD
北大核心
2004年第10期1125-1128,1132,共5页
Control and Decision
基金
国家自然科学基金资助项目(60104009)
山东省自然科学基金资助项目(Y2001G06).
关键词
计算机视觉
匀度分析
纹理分析
梯度指数
Computer graphics
Computer vision
Feature extraction
Paper
Robustness (control systems)
Textures
