An IoT-enabled energy-efficient approach for the detection of leaf curl disease in tomato crops

In tomato farms, the variation in ecological circumstances causes a disease known as leaf curl. in this paper, we have used a combination of deep learning (DL) and machine learning (ML) classifiers for the accurate detection and prevention of leaf curl disease. Initially, we designed a Smart Farm Monitoring System (SFMS), an energy-efficient prototype for monitoring and collection of real-time data from the tomato farms. Each sensing nodes senses temperature, humidity and soil moisture data from the tomato farms, aggregate them, and forward only a single packet towards the edge server. The energy-efficient data aggregation at the node level ensures that the highly-refined and fused data is available at the edge server. The use of ML and DL allows various computational models at the edge server to represent data with multiple levels of abstraction for the classification of tomato farm's data. The use of energy-efficient data aggregation at the node level and computational models at the edge server ensures that highly refined and redundant free data facilitate accurate disease prediction. Our proposed SFMS prototype shows better results as compared to the traditional approaches. For disease classification, the results obtained with different ML/DL algorithms were: ANN accuracy 92%, PNN 42% , logistic regression 97.46%, and GRNN 60%, respectively.