Publication

Abstract : Image classification and interpretation pose significant challenges in the field of artificial intelligence (AI), with the rapid growth of technology and the availability of vast image datasets offering numerous opportunities for advancement. Transfer learning, a subset of machine learning, is widely applied in addressing these challenges. While existing machine learning models have shown impressive performance in image interpretation and scene classification, there are still hurdles to overcome. In many cases, relying solely on the weights of data-dependent models is insufficient. used model for image classification demonstrates rapid convergence. This study employs two distinct deep learning models for prediction: Inception V3, which achieves a Training Accuracy of 86% and Validation Accuracy of 88%, and ResNet50, trained for 19 epochs, reaching a maximum accuracy of 94.26%. This research addresses the strengths and weakness of current models and uses the potential of transfer learning to further advance the field of image classification and interpretation.