classification of endangered bird species using deep learning

This project aims to classify endangered bird species using deep learning techniques to enhance conservation efforts. The system employs several state-of-the-art models, including EfficientNetB3, DenseNet121, a custom CNN, and MobileNetV2, for accurate bird species classification. These models are trained to recognize bird species from images, particularly focusing on endangered species. The backend of the system uses TensorFlow and Keras to implement the classification task, while the frontend, developed with HTML, CSS, and JavaScript, provides an interactive user interface for uploading bird images. The Flask-based web application allows users to register, log in, and receive predictions for bird species from uploaded images. The system is connected to a MySQL database for managing user data. Furthermore, the project integrates Gemini AI to provide an additional layer of functionality, allowing users to ask bird-related questions and receive detailed, accurate answers. The deep learning models are optimized for high accuracy and performance, offering predictions with high confidence. This platform serves as an automated tool for bird identification and aims to raise awareness about endangered species. By utilizing multiple deep learning models, the system ensures robust and reliable predictions, contributing to the conservation of wildlife.

Keywords: Endangered Bird Species, Deep Learning, EfficientNetB3, DenseNet121, Custom CNN, MobileNetV2, TensorFlow, Keras, Flask, Bird Classification, Web Application, MySQL, Gemini AI, Conservation, Image Recognition.