Implementation of PCA and KNN Algorithms in the Classification of Indonesian Medicinal Plants

Abstract

Maintaining and increasing body immunity in the midst of the Covid-19 pandemic needs to be done so that the risk of contracting this disease can be reduced. One way to increase immunity is to consume herbs or medicinal plants. Since ancient times, plants have been used as medicine and are still used today. Medicinal plants are a wide range of plants that are known to have great properties in assisting with keeping up with wellbeing and treat an infection. But many people do not know the characteristics and forms of these plants. This study performs image classification of Indonesian medicinal plants using a combination of PCA and CNN. PCA is used as a feature extraction based on the characteristics formed from each spatial property and is utilized for grouping of items dependent on learning information that has the nearest distance to the object. This investigation obtained the results that the image classification of the application of PCA and KNN on Indonesian medicinal plants with an accuracy of 88.67%.