Global Journal of Science Frontier Research, A: Physics and Space Science, Volume 23 Issue 1

Figure 4: System architecture for farm data analysis and prediction based on Google cloud (Note: The GCP icons are available in the open domain) i. Use case-Taranis (Google cloud along with Tensor Flow) As much as forty percent of crops in agriculture are regularly lost due to pests, crop illnesses, weeds, and lack of nutrients. Using drone technology and artificial intelligence, Taranis (Taranis, n.d.) , working in several countries worldwide, with more than 20 million acres of land in Russia, Eastern Europe, and South America, is providing farmers with modern tools on the cloud along with Tensor Flow to reduce crop loss, increase yields, and lower costs (Google Cloud, n.d.) .To develop BI (Business Intelligence) models, drones were deployed to capture tens of millions of photographs over the past year and a half, which have been analyzed and tagged. Each photo has up to a thousand items of interest, such as insect damage or leaf discoloration. The company claims to have processed around 100 million distinct features in approximately 700,000 images. The insights provided through visualization provide farmers with the information that enables them to intervene early and prevent crop loss. Farmers can target problems with concrete solutions, like adding fertilizer in a specific area with low in nutrients. The scalable Compute Engine dedicates information from the images when new images arrive. Data is uploaded to a Cloud SQL database for further analysis. Tensor Flow ( TensorFlow , n. a.) is an open- sourced machine learning platform that delivers a complete functional solution with a focus on deep neural networks to analyze massive amounts of unstructured data. It contains a diverse and complete set of libraries, tools, and community resources. Tensor Flow is used for building the machine learning models by using tens of millions of photographs of agriculture fields for training. Feeding the dataset for the new photographs and comparing them with models is used to discover items of interest, such as insect damage or leaf discoloration. b) Amazon Web Services (AWS) based architecture © 2023 Global Journals 1 Year 2023 22 Global Journal of Science Frontier Research Volume XXIII Issue ersion I VI ( A ) Data-Driven Knowledge Agriculture: A Paradigm Shift for Enhancing Farm Productivity & Global Food Security A typical broad system architecture using AWS is shown in Figure 5. It would be observed from the Figure that broad modalities are similar. The IOT data is streamed using a tool known as Kinesis which allows real-time streaming of IOT data from the devices deployed on the farm. Another tool, IoT Core connects billions of IoT devices such as sensors, actuators, embedded devices, wireless devices, and smart appliances over MQTT, HTTPS, and LoRa WAN protocols and routes trillions of messages to users without managing any infrastructure.