Recognition of Drought Stress State of Tomato Seedling Based on Chlorophyll Fluorescence Imaging

Drought has become one of the main challenges facing global agricultural production and crop safety. stress will lead to termination photosynthesis, which seriously affect growth development crops. We aimed study a method for identificaton drought in tomato seedlings using chlorophyll fluorescence imaging. In this study, parameters there corresponding images 4 different levels were collected. Then three feature optimization algorithms Successive Projections Algorithm, Iteratively Retains Informative Variables Variable Iterative Space Shrinkage Approac used choose important parameters. A total five common obtained, selected. And two types image features analyze classes: histogram texture features. The Pearson correlations calculated high correlated input into models, Linear Discriminant Analysis (LDA), Support Vector Machines (SVM) k-Nearest Neighbor (KNN), identify classes. recognition accuracy rate LDA, SVM KNN 86.8%, 87.1% 76.5% respectively. Our experiment results showed that could be evaluate classes seedlings, had good evaluation effect. This research provideed new monitoring considerable prospects non-destructive diagnosis plant stress.