Effectiveness of Deep Learning and IoT for Disease Classification and Characterizing Tissue Culture Calli

This research employs cutting-edge deep learning and image processing techniques to revolutionize tissue culture monitoring, addressing key research gaps and bridging the scope of research in coconut tissue culture productivity in Sri Lanka. Through advanced technologies like wise deep learning, convolutional neural networks and image processing the study brings up solutions for monitoring coconut in vitro cultures. The methodology involves capturing high-resolution images using an IoT device and a digital microscope to identify cultures and initial callus tissues respectively. Firstly, through deep learning identifies the characteristics (shape, colour and size) of the calluses, thereby classify as Embryogenic calluses and Non Embryogenic calluses. Secondly the IoT device monitor cultures in regular periods. Significantly improving disease identification and management and ensuring the culture protection. This approach integrated will demonstrate the feasibility of using deep learning and IoT in agricultural biotechnology, particularly in large-scale applications. The system not only improves the speed and accuracy of tissue culture processes however also reduce the need for manual intervention, offering a scalable solution for the Coconut Research Institute (CRI). These will show the feasibility and possible scaling up of deep learning and IoT in agricultural biotechnology for the strong solution of part of the challenges faced in tissue culture in vitro propagation.