Flowers Identification using First-order Feature Extraction and Multi-SVM Classifier

This research proposes a technique to identify flower images based on first order feature extraction and with Multi-Support Vector Machine (Multi-SVM). First-order feature extraction was chosen because it is the extraction of texture features in the macrostructure, which is considered suitable for identifying types of flowers. To perform feature extraction, color space conversion is done from RGB to Grayscale. After all features are extracted, the classification is done by the Multi-SVM classifier. Multi-SVM has the advantage of classifying more than two classes. In this study, five types of flowers were used, namely Calendula, Iris, Leucanthemum maximum, Peony, and Rose. Based on identification testing, the accuracy is 80%.