Research On The Influence Of Light Intensity On Plant Electric Signal Based On Deep Learning

In the process of growth,plants always maintain material and energy transmission with the environment.Plant electrical signal is a response of plants to changes in external environmental factors,which is ubiquitous in plants.As the main physiological signal,plant electrical signal can reflect the growth of plants and the influence of the surrounding environment.Different light intensity has different influence on plant electrical signal,so the light intensity in environment can be judged according to the size of real-time plant electrical signal.Based on deep learning and convolutional neural network theory,this paper builds a relationship model between light intensity and plant electrical signals according to the characteristics and solution methods of convolutional neural network,and studies the influence of different light intensity on plant electrical signals.First,a plant electrical signal acquisition system is to collect the electrical signals of aloe under different light levels.The acquisition experiment and electrodes were designed,and the BL-420 N signal acquisition and processing experimental equipment was used to collect the electrical signals of aloe under five different light intensities.Second,CNN neural network is used to train aloe signal.Combined with EMD and improved wavelet threshold method,the signal was de-noised smoothly,and signal data was transformed by STFT and input into CNN.Through deep learning training,the classification accuracy of aloe electrical signals to light intensity is maintained at 95%,and the loss function is 2%,to obtain the general law of light intensity and plant electrical signals,and solve problems of under-fitting and over-fitting in model training.Final,this paper optimizes the CNN classification model and builds a neural network combines convolution and long shortterm memory.Experiments on the classification model show the deep learning-based CNN-LSTM neural network training aloe electrical signal has a good classification effect,and can improve the training efficiency,computing power and model accuracy,and the classification accuracy rate is maintained at about 98%,the loss function approaches 0,and the model F1-score value approaches1.The law between the light intensity and the electrical signal of the aloe can be obtained,and the light intensity of the environment can be judged according to the electrical signal of the aloe.The research results in this paper are helpful to understand the influence of light intensity on plant electrical signals and the CNN-LSTM neural network classification model,provide theoretical guidance for the practical application of deep learning in the field of plant electrical signals,and provide theoretical guidance for the later exploration of various environmental factors and plant electrical signals.The relationship between electrical signals lays the foundation.