Implementation of Self-Organizing Map (SOM) Algorithm for Image Classification of Medicinal Weeds

Wild plants or weeds often become enemies disturb the main cultivated plants. In its development, wild actually have ingredients that are beneficial to body and can be used as medicine. However, many people still need knowledge about types of weed medicinal properties, especially leaves. The purpose this research is classify image leaves with properties based on color texture characteristics an artificial neural network using a Self-Organizing Map (SOM). To improve information in feature extraction, RGB HSV features well Gray Level Co-occurrence Matrix (GLCM). Furthermore, results extraction will identified groups classes (SOM) algorithm which divides input pattern into several so output form group most similar provided. test produces precision value 91.11%, recall 88.17% accuracy 89.44%. SOM model for classification good category.