Plant Species Identification using SIFT and SURF Technique

This research investigates the performance of plant species identification scheme using support vector regression (SVR) with different classification method. The classification involved combination of features derived from shape, shape+texture, shape+color+texture, correlation coefficient, color, and texture and cosine feature of plant species. The present paper introduces support vector regression (SVR) procedure to image based identification of species of plant. Visual contents of images are applied and three usual phases in computer vision are done (a) feature discovery, (b) feature explanation, (c) image depiction. Three dissimilar approaches are castoff on digital databases of plants. The proposed approach is done by scale invariant feature transform (SIFT) method and two combined method, SURF and features from accelerated segment test-SIFT and also clustering of data done by F-Dbscan. Vision comparison is investigated for four different species. Some quantitative results are measured and compared.