ZHAO Yue, HAN Qiaoling, ZHAO Yandong, LIU Jinhao
Computed tomography (CT) is a wide-used technology of nondestructive dynamic scanning in medical science, which can scan the samples without disturbing the internal structures, and has been successfully applied in soil science as an effective mean for quantifying the internal structures and change of frozen soil. In this article, the achievements about frozen soil research are summarized and four key links involved in frozen soil research are introduced based on computed tomography scanning technology. They were preparation of frozen soil specimen, computed tomography image acquisition, computed tomography image processing and comprehensive analysis of frozen soil computed tomography image. In the first key kink, it is illustrated how to obtain the frozen soil specimen and has listed all the equipment used for freezing soil involved in the experiment. The aim of this link is to offer the scan specimen for the subsequent computed tomography image acquisition. The second key link is to state the principle of the computed tomography images and the development of the computed tomography machine, especially accounting for the advantages about the new generation of the computed tomography machine. In the third key link, three procedures of computed tomography image processing are mainly stated, which are computed tomography image preprocessing, computed tomography image enhancement and computed tomography image segmentation, respectively. The step of computed tomography image preprocessing offers available and manageable image for the experiment. Then, the step of computed tomography image enhancement improves image quality and supplies abundant information. At last, the purpose of computed tomography image segmentation is to divide the image into several unique area and extract the target of interest. In a word, the technology of computed tomography image processing aims at providing ideal image for the next technology, and the effect of it has a great influence on the experimental results. The last key link is the analysis about the computed tomography image of frozen soil, which offers effective internal information. Finally, on the basis of the above summary, three problems are found in the computed tomography technology in current research condition of frozen soils. The first one is the limit of computed tomography machine. There are five disadvantages about the computed tomography machine, they are cost, calibration, hardware and machine performances about gray resolution and density resolution. The development of computed tomography would be a great challenge, so it is necessary especially to illustrate the technical problems in the computed tomography image processing and combine with frozen soil science. The possible methods are put forward in order to solve the key issues above, which require the integration of advanced image processing technology and efficient image analysis into the frozen soil research. Images are presented through the method of image enhancement and image segmentation to prove the feasibility of this method. Theoretical and technical guidance are provided for the recognition and quantification of frozen soil and a foundation has been laid for wide application of computed tomography technology in the further research of frozen soil.