太赫兹科学与电子信息学报, 2016, 14 (1): 7, 网络出版: 2016-03-24
基于太赫兹光谱成像技术的智能处理方法分析
Intelligent processing methods based on the terahertz spectral imaging technique
太赫兹时域光谱成像 空间成分识别 化学模式识别 人工神经网络 THz time-domain spectroscopic imaging component spatial pattern analysis chemical pattern recognition artificial neural network
摘要
太赫兹图谱的智能处理技术可以将样本的特性提取出来，并有目的地加以分类和识别，能够提高太赫兹成像的准确性，是太赫兹成像技术发展的重要标志和方向。介绍了几种智能处理太赫兹成像技术的方法的基本概念和原理，分析了它们的优缺点，并从实际应用场景出发对空间成分识别、化学模式识别、人工神经网络做了具体的分析及分类。
Abstract
Intelligent processing techniques for terahertz spectral image can extract characteristics of the samples, classify and identify them for a certain purpose. These techniques, which can improve the accuracy of terahertz imaging, are an important symbol and direction for the development of terahertz imaging techniques. In this paper, basic concepts and principles of several intelligent processing methods for terahertz spectral imaging techniques are introduced, advantages and disadvantages of these methods are analyzed. Based on practical applications, specific analysis and classification are made for component spatial pattern analysis, chemical pattern recognition and artificial neural network.
陈韬, 孟坤, 朱礼国, 李泽仁, 刘乔. 基于太赫兹光谱成像技术的智能处理方法分析[J]. 太赫兹科学与电子信息学报, 2016, 14(1): 7. CHEN Tao, MENG Kun, ZHU Liguo, LI Zeren, LIU Qiao. Intelligent processing methods based on the terahertz spectral imaging technique[J]. Journal of terahertz science and electronic information technology, 2016, 14(1): 7.
PDF全文
