作者: Yang, J (Yang, Jian); Gong, W (Gong, Wei); Shi, S (Shi, Shuo); Du, L (Du, Lin); Zhu, B (Zhu, Bo); Sun, J (Sun, Jia); Song, SL (Song, Shalei)

来源出版物: IEEE GEOSCIENCE AND REMOTE SENSING LETTERS 卷: 13 期: 7 页: 977-981 DOI: 10.1109/LGRS.2016.2557484 出版年: JUL 2016

摘要: Laser-induced fluorescence (LIF) is an active technology that is closely related to excitation wavelength (EW). This study has mainly analyzed the performance of LIF LiDAR with different EWs in distinguishing plant species. The 355-, 460-, and 556-nm lasers were utilized to excite leaf fluorescence. The fluorescence signals were measured by the LIF system built in the laboratory. Subsequently, principal component analysis combined with back-propagation neural network was used to analyze fluorescence spectra. For the three EWs, the overall identification rates of the eight plant species were 75%, 80%, and 87.5%. However, when the plant species of the same genus were taken as a category, the overall classification rates were 92.5%, 81.3%, and 86.3%. Experimental results demonstrated that, when the plant species of the same genus were regarded as a category, 355 nm was the optimal EW. However, 556 nm was superior to 355 and 460 nm in the identification of plant species of the same genus.