| Terahertz wave is electromagnetic wave whose frequency is ranged between0.1THz-10THz (1THz=1012Hz). Due to the special properties of penetrability, security, sensitivity of the biochemical composition, THz imaging is more and more widespread concerned. It has broad application prospects in many fields, such as national defense, security, biomedical engineering, product quality control.In this thesis, we analysis the key problems of the technology both theoretically and experimentally and demonstrate the applicability of THz continues wave imaging system and image clustering segmentation algorithm for THz images. The main results and conclusions are listed as follows:1. Design and build terahertz imaging systems. Test and study BWO source’s performance, complete building and debugging work on the optical platform, realize the transmission optical imaging experiments; investigate parameters of the experiment hardware such as detector and lock-in amplifier; realize the signal acquisition system functions. Discuss the optimization of THz imaging system, including the parameters testing, calibration, the optimal allocation. By optimizing sample scanning, data acquisition, improve the collection rate, the quality of the signal so as to obtain satisfactory imaging system and the rate of image quality.2. The images captured by THz-CW imaging system based on BWO have the defects such as low contrast and strong interference fringes. This thesis proposes a cluster method to detect target objects from terahertz images which are disturbed by irregular interference fringes. The proposed method revises the membership function of the fuzzy local information C-means (FLICM) technique so that it is suitable for terahertz images. Experiment results verify that the proposed clustering algorithm can detect target objects from terahertz images corrupted with irregular interference fringes, and performs better than existing methods such as Otsu, Kmeans, FLICM. |
PDF Full Text Request