Classification of Artocarpus species based on leaf recognition using multiclass support vector machine

Abstract

Abstract The demand of automated tools has been increasing regarding to the lack of people that expert in taxonomist. The aim of this research is to identify the classification of Artocarpus species based on leaf recognition using multiclass Support Vector Machine. This study focusses on identification and classification of selected Artocarpus species which are A. heterophyllus, A. altilis, A. integer and A. odoratissimus that belong to genus Artocarpus and family Moraceae through their morphological and features extraction by using image processing method. Multiclass Support Vector Machine (SVM) will be used to get the highest accuracy for the classification of Artocarpus species. The combination of Prewitt algorithm, Canny algorithm and gray level co-occurrence matrix will be used in SVM. This study capable to provide the results for current accuracy data representation of the selected Artocarpus species. The development of Graphical User Interface (GUI) for classification of Artocarpus species help user to identify and differentiate the species in faster and easier way especially botanist, taxonomist, and researcher. This system can increase the accuracy and speed of the processing and extraction of features from digital images of leaves samples. A Graphical User Interface utilizes a combination of devices and technologies to give a platform where users can interact with and producing information.