Cap-DiBiL: an automated model for crop water requirement prediction and suitable crop recommendation in agriculture

Abstract

In this technological era, various agriculture techniques are used to predict crop growth and crop recommendations. Some of the approaches depend on the IoT smart agricultural-devices to gather information from surrounding. However, multiple information from the sensor and IoT smart device does not provide a clear vision of data based on the agricultural land. Therefore, the proposed model uses various techniques for crop recommendation and crop water forecast requirement. This proposed model effectively predicts crop water requirements and recommendations for suitable crops. The steps involved in crop water requirement prediction are data pre-processing, feature extraction, feature selection, water requirement prediction and recommendation. Initially, the collected IoT data are pre-processed using data normalization, missing value imputation and one-hot encoding. From the pre-processed data, the features are extracted using the Gated Residual autoencoder (GRA) model, whereas the optimal features are selected using the Chaotic Northern Goshawk Optimization (ChaNgo) algorithm. Based on the farmland details, the crop water requirement prediction and suitable crop recommendation based on the market price are carried out using a novel hybrid deep learning model called Channel capsule-assisted stacked dilated Bi-LSTM (Cap-DiBiL). The channel capsule network predicts the crop water requirement and Stacked dilated Bi-LSTM is used for suitable crop recommendations. Then the proposed model analyses the performance and compares it with several existing techniques to prove the proposed model’s enhancement. The proposed model improves the accuracy of predicting the required water and crop recommendations by 98.18%.