<![CDATA[This research aims to address two crucial issues in the world of shark conservation, namely the decline in the number of shark species facing the risk of extinction and the difficulty in distinguishing visually between several similar shark species. To achieve this goal, the research applies the Convolutional Neural Network (CNN) method with the ResNet-50 architecture. The decision to use the ResNet-50 architecture is based on its ability to handle complex training issues, computational efficiency, and high performance in image recognition. The dataset used consists of 1497 images of sharks, divided into training data, validation data, and test data with a ratio of 60:20:20. Each image is resized to 224x224 pixels to facilitate analysis. The results of the study indicate that the method used successfully classified 14 species of sharks effectively, with an average accuracy of 96.16%. This discovery provides new hope in shark conservation efforts and opens opportunities for the development of similar methods to safeguard the sustainability of shark populations in the future.]]>
Copyrights © 2024
