Identification Of Terahertz Spectroscopy Of Illicit Drugs Using Artificial Neural Networks

The terahertz (THz,1012 Hz) wave, categorized between millimeter radio waves and far infrared light wave, with a frequency lying in 0.1THz and 10THz, belongs to far-infrared band (wavelengths 30μmto 3mm). Since it possesses many attractive characteristic properties, the studies of its applications have caused more and more attentions. It is believed that THz technique has a good future in many different fields, such as physics, chemistry, biological medicine, communication, radar, security inspection, etc.Illicit drugs have been flooding everywhere at present. Due to the limitation of existing techniques, it is important for security departments to employ effective methods in the detection and identification of illicit drugs. It is very important to study the characters of materials in THz region since it contains rich physical and chemical information. What make THz waves interesting are that the collective nature modes of most bio-molecules are located in THz range and there are possible applications because of their interesting properties. Especially, it has attracted more and more attention in nondestructive inspection and detection in biomedicine and chemistry since THz radiation itself has low energy and strong penetrability. This thesis, based on THz spectrum technique, a frontier subject, mainly focused on spectrum applications and identification arithmetic.The main work of this thesis was proposed in chapter three and chapter four.In chapter three, Absorption spectra of six illicit drugs were measured by using the terahertz time-domain spectroscopy technique in the range 0.2–2.6 THz and then clustered with self-organization feature map (SOM) artificial neural network. After the network training process, the spectra collected at another time were identified successfully by the well-trained SOM network. The results indicated that SOM network is feasible in the classification and identification of illicit drugs.In chapter four, 11 illicit drugs were studied by the terahertz time-domain spectroscopy technique. The THz absorption spectra of 11 different illicit drugs were identified successfully by Radial basis function (RBF) neural networks. The results indicate that RBF neural network would be another powerful method in identification of illicit drugs, and the speed of training and identification is fast, almost reach real time. Therefore, RBF neural network provide a fast and effective method in the secure inspection and identification for illicit drugs using THz-TDS technique.In the other chapters, chapter 1 was the introduction of this thesis. Chapter2 introduced the arithmetic of SOM network and RBF network. Chapter5 was the summary of this thesis.