Ripeness Detection of Dragon Fruit with Temperature Factors using IoT and LSTM

The project focuses on the monitoring and prediction of dragon fruit growth stages using machine learning techniques and environmental data. The four key growth stages of dragon fruit germination, vegetative growth, flowering, and fruiting are analyzed in terms of their optimal temperature requirements. The VGG16, a pre-trained Convolutional Neural Network (CNN) is used to extract relevant features from images captured at each growth stage. The extracted features are then passed to a Long Short-Term Memory(LSTM) network, which processes the sequential data to learn temporal dependencies between growth stages. The LSTM network is designed to predict the future growth stage of the dragon fruit based on the temporal sequence of features extracted from the images. This approach allows for more accurate predictions of the plant’s development, which can aid in better understanding and optimizing cultivation conditions. Alongside environmental data such as temperature and humidity of the field are continuously monitored to ensure optimal growth conditions. These conditions are displayed in real-time, facilitating timely adjustments to the cultivation environment. By combining the power of deep learning and environmental monitoring, this system enables more efficient and precise management of dragon fruit cultivation.