- Fashion-MNIST is a dataset of Zalando's article images—consisting of a training set of 60,000 examples and a test set of 10,000 samples.
- Tensorflow
- Matplotlib
- Numpy
- Random
-
Used CNN (Convolutional Neural networks) and RelU Activation function on Fashion-MNIST Dataset to denoise the images.
-
Our project outputs de-noised images directly, we can visually compare the difference between Noised and De-noised images. We can evaluate the difference between images through image properties like white balance, reduced noise spots, and sharper edges of the image subjects.
- Our Image denoiser is a prototype software that can denoise the images of resolution 28X28 in grayscale.