Low light image enhancement with ResNet and Keras

1. Introduction

In this project we are going to build a deep convolutional neural network model to improve low ligth images. we will compare our model to classic image proccessing method (median filter).

Download dataset About dataset:

So I have a knack of photography and travelling. I wanted to create a model for myself which can classify my own pictures. But to be honest, a Data Scientist should always know how to collect data. So I scraped data from google images using a Python Script and using other en-source data sources from MIT, Kaggle itself, etc. Request everyone to give a try. I'll update the no. of images in validation set as time goes on.

The link to the scripting file is here:

https://github.com/debadridtt/Scraping-Google-Images-using-Python

The images belong typically to 4 classes:

• Art & Culture
• Architecture
• Food and Drinks
• Travel and Adventure