<![CDATA[The use of artificial intelligence, especially Convolutional Neural Networks (CNN), has shown significant progress in image classification and object recognition. This research aims to develop an effective CNN model for automatically classifying gym equipment types, with the potential to improve the operational efficiency of fitness centers. The CNN model was trained using TensorFlow and Keras with the Adam optimizer and categorical cross-entropy loss function for 10 epochs, with data augmentation using ImageDataGenerator. The model evaluation shows satisfactory accuracy with a precision value of 0.9760, recall of 0.9772, and F1-score of 0.9766. The model successfully identified image samples from test data with a high level of confidence. The results of this study show that the use of CNNs in gym equipment classification has great potential to improve the efficiency of equipment recognition and contribute to the development of more advanced fitness technologies.]]>
Copyrights © 2025
