<![CDATA[Plants are one of the components needed by humans. The science that studies plants has also made rapid progress, as well as plant recognition and identification systems that are useful in providing various information. The recognition process can be applied to various parts of the plant, one of which is the recognition of leaf images. The leaf image recognition process must go through a long learning process, so an image processing technique is used, namely Artificial Neural Networks (ANN). One of the artificial neural network training methods that is often used is Backpropagation. Backpropagation trains the network to obtain a balance between the network's ability to recognize patterns used during training and the network's ability to provide the correct response to input patterns that are similar (but not the same) to the patterns used during training [1]. Identification of leaf types using ANN in this experiment uses 3 types of leaf names such as kopasanda leaves, wild plant leaves sample A, wild plant leaves sample B with 20 leaf image samples with different leaf shapes for each type. Epoch in this Artificial Neural Network reaches a maximum value of 1000 iterations. Before conducting image testing, the image training process is carried out first. After testing on 20 leaf image samples, 19 leaf image samples were found to have correctly detected results and 1 leaf image sample had undetected results. From the results of this study, the success rate was 95% successfully detected and 5% were not successfully detected. The purpose of this study is to create a system that can recognize wild kopasanda plants based on texture and leaf shape features by implementing the Backpropagation Artificial Neural Network method.]]>
Copyrights © 2025
