Publication

Abstract : Plants experience a variety of abiotic stresses as a result of environmental changes. These stresses differ significantly in plant electrophysiology’s potential and frequency. The persistent physiological status of the plant is interpreted by electrical impulses in the plants. In this study, broccoli plants are cultivated in a soilless substrate and electrical signals are acquired from both healthy and waterlogging stress states. Using the BIOPAC®MP150 setup, the short-term electrophysiological signals were recorded for eight days in both the stress (4 days) and healthy (4 days) states. Using the fast Fourier transform, signals are converted into the frequency domain (FFT). Five window lengths are used to extract frequency domain characteristics. K-nearest neighbors (KNN) classifier is used for binary classification. Results were reported with a classification rate of up to 98.8%. The findings showed that plant electrical signals identified stress more accurately before its manifestation.