Skip to content

This repository contains the keras implementation of an algorithm to classify different species of butterflies

Notifications You must be signed in to change notification settings

848475885/Butterfly_species_detector_keras

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Butterfly_species_detector_keras

This repository contains the keras implementation of an algorithm to classify different species of butterflies.

Dataset Download the dataset from http://www.josiahwang.com/dataset/leedsbutterfly/. Josiah Wang, Katja Markert, and Mark Everingham. "Learning Models for Object Recognition from Natural Language Descriptions". In Proceedings of the 20th British Machine Vision Conference (BMVC2009), September 2009.

Please organize the images into 10 different folders named from 1 to 10 containing images of each class.

Requirements: 1)Keras 2)matplotlib 3)PIL 4)numpy 5)h5py

We choose 782 training samples and 50 validation samples and manage to achieve a training accuracy of 90.2% and a validation accuracy of 89.5%.

To download the trained weight file for just testing. Download trained weights at https://drive.google.com/open?id=1t5cErFikxYl_ZtvSEr9IXSDpF5opadE1

About

This repository contains the keras implementation of an algorithm to classify different species of butterflies

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%