Auto-Aqua Classification and Production Enhancement using a Hybrid Deep Learning Approach

There were specific regions that were recognized by the abundance of paddy and aquaculture. Every season and throughout the year, aqua crops are increasing, and many farmers are converting their cultivated lands into aqua ponds to gain more income than from normal seasonal crops. In order to increase production in the aqua crop, a few significant enhancements and technology-oriented principles are being adopted. One of the factors considered a challenge is feeding the fish and related aqua stock. The other factor is an efficient approach that processes and classifies specific species. The competitive challenges influence the production enhancement in order to get the expected returns. To address and overcome both these challenges, a hybrid deep learning approach is used and implemented in the observation of aqua farming, and effective monitoring of the aqua ponds is provided using sensors. The proposed method, which consists of CNN, LSTM, and Bagging, is useful for classifying the species. The integration of IoT sensors recommends that the pond life always be healthy, and also alerts the farmer on specific species classification for easy identification and marketing. The samples from specific regions are taken as a study, and reported successful classification is reported using modern technologies like the integration of machine learning and the Internet of Things in the model. The accuracy of the proposed model in the context of aqua ponds and normal aqua ponds is analyzed. There is a significant improvement noticed between the existing and proposed environments.