PlantIntellect: Smart Crop Recommendation using CNN and Hybrid Weather Analysis

¹ Department of CSE, R.M.K College of Engineering and Technology Chennai, India
² Department of CSE, R.M.K College of Engineering and Technology Chennai, India
³ Department of CSE, R.M.K College of Engineering and Technology Chennai, India.
⁴ Department of CSE, R.M.K College of Engineering and Technology Chennai, India

This email address is being protected from spambots. You need JavaScript enabled to view it.
This email address is being protected from spambots. You need JavaScript enabled to view it.
This email address is being protected from spambots. You need JavaScript enabled to view it.
This email address is being protected from spambots. You need JavaScript enabled to view it.

Abstract

Runway incursions pose a serious threat to the safety of aviation and, as such, necessitate proactive and intelligent mitigation measures. Agriculture has ceased to be seasonally minded. Climatic conditions cannot be relied on as they change with a lot of velocity and costly losses may arise as a result of reliance on traditional knowledge alone. On-farm technologies require decision-support systems that are fully intelligent, open, and responsive to real-life conditions. The current paper describes Plant Intellect, a smart climate-sensitive crop suitability model based on AI, which can be used to guide effective agricultural planning. This system starts with the recognition of the species of plants in user-posted images through a convolution neural network-based recognition API. After the identification of the plant, its agronomic needs are then fetched out of a structured knowledge base and compared against the real time environmental conditions of temperature, humidity, wind speed and rainfall among others. Plant Intellect uses an explicit, rule-based Suitability Decision Diagram instead of being a black-box predictive model to consider every environmental factor individually and then make an overall judgement of suitability. In order to enhance reliability, mathematical extraction methods such as averaging of time and variance analysis are employed to stabilize varying environmental data. In case the growing conditions are not favourable then the system prescribes other crops on the basis of small environmental deviation. High performance with experimental results of 94 % accuracy with high recall and low false-positive rates. Plant Intellect will provide a viable and reliable answer to climate-informed precision agriculture through a combination of on-the-fly sensing, explainable logic, and adaptive recommendations.