基于太赫兹时域光谱技术的伪色彩太赫兹成像的实验研究

False-color terahertz imaging system based on terahertz time domain spectrocsopy

提出了一种伪色彩太赫兹成像技术.通过引入频域色彩区间积分,建立了一套基于太赫兹时域光谱技术的伪色彩太赫兹成像系统,实验分别研究了乳糖和对氨基苯甲酸两种不同白色化学粉末的伪色彩成像和灰度成像,研究了不同颜色区间定义对伪色彩图像的影响,讨论了利用不同频率信息成像系统所能达到的空间分辨率.研究结果表明,伪色彩成像技术可以将不同的物质信息同时成像在一张太赫兹图像中,通过不同物质在太赫兹图像中呈现出的颜色差别来区分不同的物质及其分布.克服了传统的太赫兹灰度成像技术中,需要多张图像来区分不同的物质的问题,提高了成像速度,降低了筛选难度.利用高频信息进行伪色彩成像,可以将系统成像的空间分辨率提高到0.4 mm.伪色彩成像方式可以更直观快捷地显示样品的基本属性,对于实现太赫兹安检的初检和快速筛选具有重大的现实意义.

Based on terahertz time domain spectroscopy, a false-color imaging system is demonstrated by experiments. Three frequency ranges are defined as color ranges for three primary colors （red, green and blue）. The mixture of the spectral integral values in each color range presents the final color of each pixel on the false-color THz image. Since the absorption frequencies of different materials are different, the spectral integral values in defined ranges are different, leading to different color on the false-color THz image. The false-color THz images of two kinds of white powder which are lactose and 4-aminobenzonic acid are obtained from the imaging system with two different definitions of color ranges. From the first color range definition, the absorption frequency of lactose lies in the green range, so only the green light is absorbed and the color of lactose is magenta. In the meanwhile, there are two absorption frequencies for 4-aminobenzonic acid lying in the green and blue ranges, so both green and blue light are absorbed, and the color of 4-aminobenzonic acid is red. They can be told easily by different, colors on the false-color THz image. From the second color range definition, the dolors of two kinds of powder are more different. Both false-color THz images can present the cuvette and two kinds of powder clearly. By comparing the THz imaging with grayscale images, false-color THz imaging can display different materials by different colors in one image, instead of the requirement of many grayscale images. It is no need to generate grayscale images at each frequency, making false-color THz imaging consume less time. The false-color imaging is clearer and more efficient, which is more suitable for recognition in a rapid security check. In the situation of complex materials, more false-color THz images can be generated by different color range definitions to assist the detectipn. The spatial resolution of the imaging system is also investigated. The resolution of imaging system is investigated by imaging home-made standard sample plate. For the frequency range that is higher than 0.3 THz, the resolution can reach 0.4 mm, which is larger than enough for most practical applications.