Terahertz time-domain spectroscopy combined with PCA-CFSFDP applied for pesticide detection

The purpose of this paper was to apply terahertz time-domain spectroscopy (THz-TDS) combined with clustering by fast search and find of density peaks (CFSFDP) for pesticide detection. To solve similar distance problem in THz spectra data, we proposed a method PCA-CFSFDP. Principal component analysis (PCA) was used to reduce dimensions of THz spectrum data, and then used the result of PCA as the input of CFSFDP. The performance of PCA-CFSFDP was investigated, and the experimental results showed that PCA-CFSFDP approach can get a satisfactory detection result. In addition, the proposed method was compared with K-means. The comparison results demonstrated that PCA-CFSFD outperforms K-means. Therefore, THz-TDS combined with PCA-CFSFDP is a potential identification tool for pesticide detection.