Despite their detrimental nature, it turns out that wild plants have many benefits for human health. Wild plants with a form of herbaceous vegetation contain ingredients that can be used as medicine, especially in their leaves. However, because the information is very similar and the form is similar, people don't know about it. For this reason, the aim of this research is to implement an artificial neural network algorithm using Extreme Learning Machine (ELM) and the Principal Component Analysis (PCA) algorithm to classify images of wild plant leaves with medicinal properties, especially in herbaceous vegetation. The feature extraction used in this research involves morphological features by considering the shape of the object. The PCA algorithm will reduce data complexity and identify hidden patterns in the data by changing the original feature space to a new and more concise feature space. Next, the ELM algorithm is used to recognize class grouping patterns when solving classification problems. Accuracy test results show a value of 90.667%.