AUTOMATED BIRD SPECIES IDENTIFICATION USING DEEP LEARNING WITH IMAGE AND AUDIO

AUTOMATED BIRD SPECIES IDENTIFICATION USING DEEP LEARNING WITH IMAGE AND AUDIO

Tanay Sonavane¹, Kireet Verma², Hrithik Wadile³ , Prof. Satish Kuchiwale⁴

^1,2,3,4Dept. of Computer Engineering, Smt. Indira Gandhi College of Engineering, Navi Mumbai, Maharashtra, India

---------------------------------------------------------------------***---------------------------------------------------------------------

Abstract - With over 9000 species of birds in the world, some of which are rarely discovered, it can be difficult to predict their classification. Visual recognition of birds is also more easily understood by humans compared to audible recognition. To aid birdwatchers in appreciating the beauty of birds, a system utilizing Convolutional Neural Networks (CNN) has been developed to classify bird species. CNNs are a powerful set of machine learning techniques known for their effectiveness in image and sound processing. This system uses the Caltech-UCSD Birds 200 (CUB-200-2011) and Kaggle datasets for training and evaluating the CNN system based on image recognition. Additionally, several different sound sources were used to train the sound recognition model.

Key Words:  birds, deep learning, image identification,

ound identification