Terahertz Spectroscopy Of Chinese Medicine Based On Support Vector Machine

Terahertz radiation is sandwiched in between microwave and infrared,so using these new THz technique,We can get more characteristic of substance.As well known,the quality estimate and control of Chinese traditional medicine is a difficult problem for ages.The spectrum technology is one of important method of quality estimate of Chinese traditional medicine,however these study is concentrating on infrared spectroscopy(FTIR).THz technique is an availability supplement.There will have different result if we use different distinguish method.So the method is very important.Support Vector Machine(SVM) is a new machine learning algorithm proposed by Vapnik and his group for classification and regression question.It base on structural risk minimization can effectively solve the over study problem and has the good extension and the better classified accuracy.It has become the new research hotspot after the research of nerve net.Applied THz and SVM technique in distinguish Chinese traditional medicine,we get successful result.First,we study the THz spectroscopy of three group Chinese Traditional Medicine.Using SVM technique,we can successfully distinguish them although they have same shape.At the same time,we use nerve net to distinguish.Through comparing the result,we know that SVM is better than nerve net in Chinese Traditional Medicine distinguish.Second,we use THz spectroscopy technique to study five kinds of Oral liquid.The result is contentment.So,we can know that THz and SVM technique in distinguish Chinese traditional medicine is very effective.Chapter one introduces the basic theories of THz spectroscopy technique, including the terahertz pulses generation,the terahertz pulses detection and its characteristics.Traditional methods of recognition of Chinese traditional medicines were summarized.Our major work and signification for this thesis were presented.Chapter two introduces the principle of optical path which was built by our laboratory,experimental measurement procedure and sample preparation.This chapter especially emphasized on the principle part which were used in the article, including how to calculate optical constant of sample.Chapter three introduces the principle of SVM,which is the basic of our work.Chapter four particularly intruduces the experiment which distinguish three group of Chinese traditional medicines.Through analyse the result of SVM and artificial nerve net,we can conclude that the method of SVM is better than artificial nerve net.Chapter five particularly intruduces the experiment which distinguish five Oral Liquid.The result indicate that the method of SVM with THz time-domain Spectroscopy Technique is a efficient paths.Chapter six is the summary of my work including experimental results and pattern recognition.The future and hurdles of terahertz spectroscopy technique in the idention of Chinese traditional medicine were discussed.