Hybrid Fourier Descriptor Naïve Bayes dan CNN pada Klasifikasi Daun Herbal

Plants are vital to human life on earth, and the leaves their whole parts have many benefits. These of plant can help distinguish between different species. The leaf identification be performed at any time, while other plants only identified a certain time. study aims classify two types herbs i.e. saur-opus androgynous moringa oleifera, implementing Fourier Descriptor method extract shape texture features. In process classification using Naïve Bayes with three nuclei (Gaussian, Bernoulli, Multinomial) Convolutional Neural Network. testing was carried out scenarios, dark light, where each scenario consisted 240 images for total 480 divided into 20% data 80% training data. Descriptor-Bernoulli Naive gives lowest accuracy in both light 46% 52%, respectively. As herbal combination Descriptor-Convolutional Network method, it is recommended used image scenarios Descriptor-Gaussian because able detect 100% accuracy.