Automatic classification of monocots and dicots using SVM classifier

In this work, the core objective is to implement an automatic and reliable system for the classification of plants into two plant categories named as monocotyledonous and dicotyledonous using the microscopic images of plant stem cross sections. The system can be used for the classification of plants when a large number of new plant species are discovered and it can be applied in plant disease detection project. A digital microscope is used for gathering the cross sectional images. Here, the proposed system mainly consists of three sections. The first section is image gathering and pre-processing and the next section is the feature extraction by spatial methods. The texture features are extracted using GLCM method in this project. The last section is the classification of the feature sets corresponding to the images into two categories such as monocots and dicots. Support vector machine is used as the classifier.