Image Noise Reduction using Auto Encoder

Discription

In this project, a convolutional autoencoder is presented which is trained to denosify images. Model is trained on the actual images and nosiy images and tested with noisy images.

Background Detailes

What is Auto Encoder?

An autoencoder is a type of artificial neural network used to learn efficient data codings in an unsupervised manner. The aim of an autoencoder is to learn a representation for a set of data, typically for dimensionality reduction, by training the network to ignore signal “noise”

Data used to train the model

MNIST data is used to train and test the model

Architecture

Link

Results

50% noise in training data and 10% noise in testing data

50% noise in training data and 50% noise in testing data

70% noise in training data and 50% noise in testing data

70% noise in training data and 70% noise in testing data