Skip to content

Using a data set including five types of grape leaves (100 unique photos of each type) and deep learning, we are trying to build a model to perform this classification well.

Notifications You must be signed in to change notification settings

SamanArzaghi/classification-grape-leaves

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

15 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

classification-grape-leaves

Using a data set including five types of grape leaves (100 unique photos of each type) and deep learning, we are trying to build a model to perform this classification well.

With simple convolutional neural network we would not achive high accuracy, so we done below steps to improve our accuracy(you can check the complete details in report.pdf).

  • Used transfer learning(use pretrained MobileNetV2)
  • Used augmentation to increase data size
  • Used fooly connected net as classification part
  • Used SVM as classification part(for better result)

Data set

You can get the dataset with the following link: https://www.muratkoklu.com/datasets/Grapevine_Leaves_Image_Dataset.zip

Libraries

The main librarie that i use for neural networks is TensorFlow,

also i tried the transfer learning with PyTorch to try somthing different(why not?),

and for the SVM part i used ScikitLearn.

Report

Also there is a report as PDF that you can follow the code with more details.

Contact

Created by [email protected] / [email protected] - feel free to contact me!

About

Using a data set including five types of grape leaves (100 unique photos of each type) and deep learning, we are trying to build a model to perform this classification well.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published