Skip to content

A Convolutional Neural Netwok app made using Tensorflow 2.x, Keras & Streamlit.

Notifications You must be signed in to change notification settings

Vibhugupta10616/Image_Scene_Classification_with_CNN

Repository files navigation

Image Scene Classification using CNN

A Convolutional Neural Netwok app made using Tensorflow 2.x, Keras & Streamlit.

This app classifies the images and predicts whether the image has Buildings, Mountains, Sea, Street, Glacier in it.

Here is a Link to a Short Demo of the Complete App :- https://www.dropbox.com/s/vnsojin23kauzg8/Demo%20Video%20of%20the%20complete%20app.mp4?dl=0

App

Short Description

In this projects I have used two Deep Learning CNN Models:-

1. VGG 16 (Transfer Learning Model):- This model gives 93.6% accuracy on training data & 86.6% accuracy on validation data

2. Simple CNN Model:- This model gives 89.9.6% accuracy on training data & 80.46% accuracy on validation data

Thanks for Reading !!

About

A Convolutional Neural Netwok app made using Tensorflow 2.x, Keras & Streamlit.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published