A Deep Learning Approach to pet Classification: A Study Using Image-Based Data
DOI:
https://doi.org/10.56536/jicet.v4i2.179Abstract
Recent advancements in deep learning, particularly in image recognition, have revolutionized robotics, computer vision, and natural language processing. These innovations hold significant promise for advancing ecological research and improving animal welfare. By leveraging the power of deep learning algorithms, this dissertation investigates the application of image recognition techniques for accurate animal species identification. The research will encompass several key aspects, including curating and managing a comprehensive image dataset and developing and evaluating state-of-the-art deep learning models, such as ImageNet, ResNet, and DenseNet. Preliminary investigations have demonstrated promising results, with DenseNet achieving an accuracy of 0.97, highlighting the potential of these models for effective animal species classification.