Fertilizer Classification System

Abstract

Land has been one of the scarcest resources in the present day, mainly due to the population explosion (approximately 6.9 billion in 2009). The land available for cultivation is even scarcer as only a miniature percentage of land is suitable for cultivation. In order to feed the whole population either the more land should be cultivated, food losses and wastages should be minimized or the crop yield should be increased. The latter option is more feasible as the increase in cropland may severely strain natural eco systems and may be quite difficult pragmatically and politically. Therefore, many researches have been conducted on achieving a Maximum Crop Yield (MEY). The researches have proved that the optimal growth and reproduction of plants could be achieved by providing the exact condition levels (Sunlight, water, fertilizer, temperature). Deviating from the optimal condition level will only hinder the MEY. This means that the plants also need a balanced nutrient level, just like humans need a balanced diet to maintain good health. The objective of this project is to develop fertilizer classification system which helps to automate the classification process of existing nutrient amount and fertilizer suggestions, which will be invaluable for farmers and agricultural consultants. The research will provide a detailed description about the Fertilizer Classification System that will be developed using Fuzzy Logic and Artificial Neural Network (ANN). One of the main purposes of this dissertation is to provide a better knowledge on the methodologies that will be used for developing the system. Furthermore, the research paper will provide an insight to the past literature on the topic, data collection methods, and the analyzing and designing techniques. The proposed system will classify the existing nutrition amount using crisp sets and also classify the existing nutrient amount depending on the nutrient category, using the ANN. The proposed system then computes the fertilizer suggestion based on the crop type using crisp sets and also computes the fertilizer suggestion using fuzzy theory which is more cost effective. The system will then derive the best fertilizer mixture especially for Papaya, Coconut and Banana plants. We can extend the system further for every plant that we need to get maximum yield. We can also extend this system to decide the crop type for a particular area/ soil sample based on the environmental condition and then decide the best fertilizer mixture.