<![CDATA[Image processing has become an increasingly important technology in various fields, including botany, particularly to support the automatic identification of plants. One of the main challenges in identifying jasmine flowers lies in the manual process, which is time-consuming and heavily reliant on individual expertise. To address these limitations, this research aims to design a detection and classification system for jasmine flowers using Convolutional Neural Network (CNN), capable of identifying four jasmine flower types: Melati Putih, Melati Jepang, Melati Gambir, and Melati Kuning. The system employs a modified CNN architecture, ResNet50v2, incorporating a 50% dropout layer, Adam optimizer with a learning rate of 0.001, and data augmentation techniques to enhance model performance. The dataset used consists of 350 images for training and 88 images for validation. Additionally, the system is designed as a web-based application to provide real time detection features and classification history. Evaluation metrics include accuracy, precision, recall, f1 score, MSE, RMSE, and MAPE. Results indicate that the developed system achieves an accuracy of 97%, MSE 0,33, RMSE 0,18, dan MAPE 1,8%.. These findings demonstrate that the system can effectively detect and classify jasmine flowers with high accuracy, enabling fast and precise identification. Future research is recommended to expand the dataset to improve the model's generalization across broader variations and explore other model architectures for performance comparison. This system is expected to provide significant contributions to education, agriculture, and plant conservation, especially in facilitating the automatic identification of jasmine flowers.]]>
