Farmers as Sensors: A Crowdsensing Platform to Generate Agricultural Pest Incidence Reports

Abstract

Over the years, the food produced for human consumption is lost or affected due to many factors. Among these, pest/disease incidence is one significant factor contributing to crop losses. Hence, early identification of the presence of pest/disease incidence is essential to manage crop losses. In Sri Lanka, farmers identify a pest/disease incidence by mainly relying on the input given by agricultural experts, and sometimes they rely on fellow farmers, pesticide dealers, and even on their own experience. The current approaches followed by the farmers to communicate the pest/disease symptoms to agricultural experts are not appropriate and resulted in many incorrect choices made by the farmers. In this paper, we discuss about an extension we proposed for a mobile application we developed for farmers in Sri Lanka called Govi Nena. This extension aimed to capture the conditions concerning pest/disease symptoms and climate present in the field to assist agricultural experts in the decisionmaking process. We consider each farmer as a sensor to capture information such as symptoms present in the crop, distribution of symptoms, affected parts of the plant, and growth stage of it. The enhanced version of the application was given to agricultural experts to understand their feedback in regards to this. The feedback was positive, and now, we have undertaken to deploy the application among several farmers based in Sri Lanka for testing purposes.