Zhang Zeng-yan, Ji Te, Xiao Ti-qiao, Zhao Hong-wei, Chen Min, Yu Xiao-han, Tong Ya-jun, Zhu Hua-chun, Peng Wei-wei
Guang Pu Xue Yu Guang Pu Fen Xi. 2015 Dec;35(12):3315-8.
A new method of terahertz (THz) imaging based on the mean absorption is proposed. Terahertz radiation is an electromagnetic radiation in the range between millimeter waves and far infrared. THz pulse imaging emerges as a novel tool in many fields because of its low energy and non-ionizing character, such as material, chemical, biological medicine and food safety. A character of THz imaging technique is it can get large amount of information. How to extract the useful parameter from the large amount of information and reconstruct sample's image is a key technology in THz imaging. Some efforts have been done for advanced visualization methods to extract the information of interest from the raw data. Both time domain and frequency domain visualization methods can be applied to extract information on the physical properties of samples from THz imaging raw data. The process of extracting useful parameter from raw data of the new method based on the mean absorption was given in this article. This method relates to the sample absorption and thickness, it delivers good signal to noise ratio in the images, and the dispersion effects are cancelled. A paper with a "THz" shape hole was taken as the sample to do the experiment. Traditional THz amplitude imaging methods in time domain and frequency domain are used to achieve the sample's image, such as relative reduction of pulse maximum imaging method, relative power loss imaging method, and relative power loss at specific frequency imaging method. The sample's information that reflected by these methods and the characteristics of these methods are discussed. The method base on the mean absorption within a certain frequency is also used to reconstruct sample's image. The experimental results show that this new method can well reflect the true information of the sample. And it can achieve a clearer image than the other traditional THz amplitude imaging methods. All the experimental results and theoretical analyses indicate that the method base on the mean absorption within a certain frequency can reflects sample absorb and thickness information, it can achieve good signal to noise ratio in the images. Because the absorption is mean absorption within in a certain frequency, so the method proposed in this article is especially suitable for samples with simple structure. And this new method can be a useful added tool for the other traditional THz amplitude imaging methods.
提出了一种基于平均吸收的太赫兹（THz）成像新方法。太赫兹辐射是毫米波与远红外之间的电磁辐射。太赫兹脉冲成像因其低能量和非电离特性，在材料、化学、生物医药和食品安全等诸多领域成为一种新型工具。太赫兹成像技术的一个特点是它能获取大量信息。如何从大量信息中提取有用参数并重建样品图像是太赫兹成像中的一项关键技术。人们已经在先进的可视化方法上做了一些努力，以便从原始数据中提取感兴趣的信息。时域和频域可视化方法都可用于从太赫兹成像原始数据中提取关于样品物理特性的信息。本文给出了基于平均吸收的新方法从原始数据中提取有用参数的过程。该方法涉及样品的吸收和厚度，在图像中能提供良好的信噪比，并且消除了色散效应。以一张带有“THz”形状孔洞的纸张作为样品进行实验。采用传统的时域和频域太赫兹幅度成像方法来获取样品图像，如脉冲最大值相对减小成像法、相对功率损失成像法以及特定频率下的相对功率损失成像法。讨论了这些方法所反映的样品信息以及这些方法的特点。基于特定频率内平均吸收的方法也用于重建样品图像。实验结果表明，这种新方法能够很好地反映样品的真实信息。并且它能比其他传统太赫兹幅度成像方法获得更清晰的图像。所有实验结果和理论分析表明，基于特定频率内平均吸收的方法能够反映样品的吸收和厚度信息，在图像中能获得良好的信噪比。由于吸收是特定频率内的平均吸收，所以本文提出的方法特别适用于结构简单的样品。并且这种新方法可以成为其他传统太赫兹幅度成像方法的一个有用补充工具。
