Indian Journal of Science and Technology

Abstract

In an agricultural field, farmers have a wide range of diversity to select suitable fruit and vegetable crops. However, the cultivation of these crops with quality produce is highly technical. Most challenging for farmers is to differentiate between crops and weeds. The proposed method identifies the weeds by using leaf parameters such as shape, color, and texture. Pest/disease detection is also possible by detection of varied shape and color of disease affected crop leaves. In order to improve the accuracy of weed detection, need to develop a weed detection algorithm which could be supported in all the cases accurately. Adding on, crops face many diseases. Pest/diseases are seen on the leaves of the plant. To avoid misclassification of such disease affected plant as weed, a new algorithm is developed for disease identification and included in weed detection algorithm. Recognition of normal leaf, diseased leaf and weed is based on similarity measure.
Keywords: Euclidean Distance, Leaf Parameters, Image-Processing, Plant Disease Detection, Weed Detection