Support Vector Machine-Based Classification Scheme of Maize Crop

This paper presents the information pertaining to the damage of crops in recent years due to the growth of weeds. Weeds are one of the considerable menaces to the real home and humankind. They are destroying native habitats, forbidding native plants. Weeds cause decrease in farm and forest yield, invade crops and they can also harm livestock. The only solution to this problem is to avoid the growth of weeds either by removing them or using herbicides. In this proposed automated idea, Support Vector Machine (SVM) Classifier is utilized to make out whether plant is crop or weed. The maize crops are continuously monitored by capturing images using camera. In order to classify a plant as a crop or weed, various features are extricated which among them are shape, texture, color. We have made use of statistical textural features such as Intensity, Mean, Energy, Entropy, Standard Deviation, Smoothness and Third Moment. Machine learning algorithm such as Support Vector Machine (SVM) Classifier is used for accurate prediction to find out whether the plant is crop or weed. The data-set consists of images of maize crop which are 500 in number and equal number of weed images. The accuracy of prediction made by Support Vector Machine (SVM) Classifier is determined by using K-fold cross validation, according to which the accuracy is 82%.