Paddy Weed Leaf Classification Using Neuro-fuzzysystem

The productivity of paddy rice in Malaysia has improved as the technology advances. However, several concerns need to be considered to maintain the productivity. The growth of paddy weeds in the fields is one of the concerns that need to be taken care of as it resulted to the decrease of the paddy yield. This paper presents an integrated method for classifying paddy weeds through the shape of their leaves by applying neuro-fuzzy techniques. The focus weed types are Sphenoclea zeylanica, Ludwigia hyssopifolia and Echinochloa crus-galli. The developed e-prototype is able to classify the paddy weeds with 83.78% accuracy. Hopefully the findings in this study may assist the farmers and researchers in increasing their paddy yield.