Research on Plant Growth State Classification Based on CNN- LSTM

Abstract

The plant electrical signal is a physiological signal that reflects the growth state of plants affected by the external environment. Online monitoring of plant growth states is realized by studying the electrical signal changes of plants in different growth states. In this paper, a Convolutional Neural Network(CNN) based and Convolutional Neural Network and Long Short-Term Memory Neural Network(CNN-LSTM) based classification model of plant growth state is built to realize feature extraction and training and classification studies of Aloe Vera electrical signals in different growth states. The short-time Fourier transform (STFT) is used to convert the de-noised aloe electrical signal into a signal energy map, which is used as the input of the classification model, and the different growth states of the aloe are used as the output of the classifier. It is concluded that the CNN-LSTM neural network model has high accuracy in the classification of aloe electrical signals in different growth states when training, and the plant electrical signals can be used as an effective evaluation index for plant growth state detection.