Rapid and Nondestructive Identification of Tomato Plants in Different Periods Based on Hyperspectral Imaging Technique

KeyWord:hyperspectral imaging technique  tomato leaves  salt stress  nondestructive testing

Abstract:

      A near infrared hyperspectral imaging£¨NIR£© technique was adopted for qualitative identification on the tomato leaves under different salt stress in order to quickly monitor the growth of tomato plants£®The average spectral reflectance data of 192 leaf samples were collected£®The original spectral data were preprocessed by multiple scattering correction£¨MSC£©£¬standardized normal variate£¨SNV£©£¬orthogonal signal correction£¨OSC£© and correlation optimized warping£¨COW£© to establish a partial least squares regression£¨PLSR£© model£®The modeling results showed that the modeling effect of OSC preprocessed spectrum was the best£®Interval variable iterative space shrinking analysis£¨iVISSA£©£¬interval random frog£¨IRF£©£¬genetic algorithm and partial least squares£¨GAPLS£©£¬competitive adaptive weighted sampling£¨CARS£© and variable combination population analysis£¨VCPA£© were used to extract the feature wavelengths£¬thus the PLSR model was established£®The results showed that the model established by VCPA extracting characteristic wavelengths was optimal£®The VCPA method was used to extract 11 characteristic wavelengths£¨945£¬975£¬990£¬1 002£¬1 005£¬1 067£¬1 204£¬1 326£¬1 595£¬1 642£¬1 660 nm£©£¬which were used to establish the qualitative discriminant prediction model for tomato leaves£¬while the determination coefficient£¨R2P£© of the optimal prediction model and the root mean square error£¨RMSEP£© were 0.917 and 0.456£¬respectively£®This study provided a technical support for online monitoring of plant growth in the future.