No Detail Found!

Extracting the most important discrete cosine transform (DCT) coefficients for image compression using deep learning

Author(s)
Zirak Khan Department of Computer Engineering, College of Electrical & Mechanical Engineering (CEME), National University of Sciences & Technology (NUST),Islamabad,Pakistan
Farrukh Arslan Department of Electrical Engineering, University of Engineering and Technology (UET),Lahore,Pakistan
Faseeha Munir Department of Computer Science, Muhammad Nawaz Shareef University of Agriculture (MNSUAM),Multan,Pakistan
Mubashir Ali Department of Software Engineering, Lahore Garrison University (LGU), Lahore, Pakistan
Shahrukh Department of Software Engineering, Lahore Garrison University (LGU), Lahore, Pakistan
Abstract
Image compression is all about reducing storage costs and making the transmission of huge image files feasible. This paper targets lossy image compression by estimating the most important Discrete Cosine Transform (DCT) coefficients through employing deep learning. DCT basically results in the transformation of an image to the frequency domain from the spatial domain. The first few coefficients, in the frequency domain of a transformed image, have great importance. They are the most informative, while the others are of the least importance. The abilities of Multi-Layer Perceptron (MLP) and Convolutional neural network (CNN) were exploited in order to find a reasonable estimate of important DCT coefficients. The target was to get a deep neural network (DNN) for the compression of digital images that has a reduced number of DCT coefficients that is; higher compression rate and better image quality upon reconstruction and improved generalization ability. To shorten the encoding-decoding time and to fasten the training of our deep neural networks, RELUs and Tangent Sigmoid were used. Experiments performed on a large set of grayscale images shows that only 15 out of 64 total available DCT coefficients result in more than 70% image quality and have a good compression ratio of 4:1. Moreover, the quality of images upon a subjective and objective evaluation of unseen data proves that our proposed MLP achieved better generalization as compared to CNN.
Publication Details
Page(s) 56-63
DOI DOI not available
Published Journal: VFAST Transactions on Software Engineering, Volume: 10, Issue: 4, Year: 2022

Request the full text from PASTIC

Keywords
Deep neural network, Deep learning, Image Compression
0

Citations

0

Downloads

84

Views