Year: 2025
Type: conference-paper
Abstract: The detection and classification of invasive plant species are crucial for biodiversity conservation and ecosystem management. This research work explores the use of Convolutional Neural Networks (CNNs) to identify invasive species from images. By leveraging the powerful feature extraction capabilities of CNNs, we developed a model that accurately classifies the non-invasive and invasive plant species. The model was trained on a diverse dataset, incorporating various environmental conditions and geographical locations to enhance its robustness. Evaluation metrics indicate high precision and recall, demonstrating the model's effectiveness in real-world applications. Our findings suggest that CNN-based approaches can significantly enhance monitoring and management efforts for invasive species, offering a scalable and automated solution for environmental scientists and policymakers. Our proposed model achieved 92.5% accuracy by utilizing Resnet50 architecture which is superior than existing state of the art models.
Source: EPJ Web of Conferences
Authors: Mukunda Sri Konatham, Jayasree marrapu, M. Gopinath
Institution: Siddhartha Academy of Higher Education
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
SDG: Life on land
Open Access status: diamond
OpenAlex ID: W4411430388