Enhancement in Multi Spectral Camera System for Precision Agriculture using NDVI Calculation Algorithm

Abstract

Precision agriculture improves productivity, resources optimization, and sustainability in crops. However, despite having an elevated cost for commercial multispectral imaging systems, the current application of precision agriculture technologies in crop health monitoring has not been adopted by small and medium-sized farmers. Thus, this paper presents the design, development, and validation of an indigenously developed low- cost multispectral camera system that is affordable to use as an alternative in crop health monitoring. With the current system, use of readily available off-the-shelf components will be employed as a Raspberry Pi microcontroller and low-cost camera modules with an optical filter to capture in key bands of the spectrum such as Red, Green, Blue, and Near-Infrared. Calculations of vegetation indices, such as NDVI, will be made, offering real-time information relating to the health of the plants, water stress, and nutrient deficiencies. The system is highly accessible to users who are not nearly as technically oriented: farmers, for example. Field tests indicate that the system's spectral accuracy matches commercial ones, but at a fraction of the cost and therefore accessible to the smallholder. Consequently, it will improve crop yields, reduce wastage of resources, and promote sustainable farming practices through farmer decision-making based on data. This paper discusses the system architecture, its implementation, field validation, and future enhancements which will lead towards a scalable solution for democracies of precision agriculture technology.