Loading…
Bird Species Recognition using Deep Learning
Birdwatching or Birding is observing or watching the birds which is found to be one of the serene activities to do in daily life. But recognizing the bird species is hard for humans as it requires a bit of support from the bird book. Also, in the olden days, many ornithologists and researchers faced...
Saved in:
Main Authors: | , , , , , |
---|---|
Format: | Conference Proceeding |
Language: | English |
Subjects: | |
Online Access: | Request full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | Birdwatching or Birding is observing or watching the birds which is found to be one of the serene activities to do in daily life. But recognizing the bird species is hard for humans as it requires a bit of support from the bird book. Also, in the olden days, many ornithologists and researchers faced hardships in detecting bird species and learning about the different patterns of bird species. We have developed an Arduino-based system that performs automatic bird species recognition. This system will be helpful for ornithologists, researchers, and other enthusiasts to learn about the existence of different bird species in a given geographical region. This system is developed using Arduino Uno, PIR Motion Sensor, and an ESP-32 camera. When the motion is detected, the ESP-32 camera captures the image and uploads the image to Google Drive. The images in Google Drive can be given to the trained deep learning models to predict the name of the bird species. To develop the deep learning model, we have used a Kaggle data set that consists of 450 bird species images of which 70,626 training images, 22500 test images, and 2250 validation images. |
---|---|
ISSN: | 2640-5768 |
DOI: | 10.1109/AISP57993.2023.10134804 |