Convolutional Autoencoder based Deep Learning Model for Identification of Red Palm Weevil Signals

This paper presents a Convolutional Autoencoder based Deep Learning model for identification of Red Palm Weevil acoustic emissions from other background noise. Mel spectrogram of acoustic samples was chosen as the extracted feature for the proposed model. The designed Convolutional Autoencoder was trained using Mel spectrogram images of Red Palm Weevil acoustic activities which are regarded as the normal instances. Unbiased evaluation of the model was done with a test dataset composed of normal RP'W acoustic emissions as well as anomalous acoustic samples. The model could achieve a very high classification accuracy of 95.85%. The results confirmed that the proposed method is highly efficient for the identification of Red Palm Weevil signals.