Classification Method of Plants with Support Vector Machine (SVM) Using Local Binary Gaussian Model (LBGP) for Smart Agriculture

Abstract

In this study, plant species are classified for smart agriculture. 6170 images of 17 plant species were collected using the camera on the UAV. Histogram, Local Binary Pattern (LBP) and Local Binary Gauss Pattern (LBGP) filters were used to extract features from these images. The obtained features were classified with Support-Vector Machine (SVM). Each image is 1280x720 pixels in size. Gray histogram of these Red-Green-Blue (RGB) images are extracted. These features are classified using LBP and LBGP filters. An accuracy of 93.74% was achieved with our proposed method using the SVM classifier and LBGP filter. The proposed method is a lightweight method that gives much better results than histogram and LBP filters.