Fig. 1
From: Early detection of dark-affected plant mechanical responses using enhanced electrical signals
Pipeline for the experimental setup and the data analysis. A Overview of the data acquisition system. To avoid external interference, 5-week-old Arabidopsis plants were treated under different light conditions before measurement in a Faraday cage. Light conditions: Normal: 10 h light/14 h night; SED: short extended darkness, 10 h light/18 h night; LED: long extended darkness, 10 h light/40 h night. Leaf 8 was wounded with forceps and the electrical signals were measured by two electrodes (e1, e2) that were placed on the petioles of leaf 8 and leaf 13, respectively. The signals were amplified through a signal amplifier and then converted for visualization on the laptop as shown. B Overall workflow for classifying the electrical signals. Following preprocessing, characteristics were extracted from the leaf 8 and leaf 13 electrical signals, respectively. Machine learning-based classification was performed using either the extracted characteristics from original samples or extended samples upon data augmentation. The colored bars next to “Data augmentation” represent the structural network for data augmentation