Image Compression Using Deep
Convolutional Adversarial Networks
Shiv Ashish Dhondiyal, Manisha Aeri, Manika Manwal, Sugandha Sharma,
and Sumeshwar Singh
Abstract Image compression is a kind of compression of data, which is used to
images for minimizing its cost in terms of storage and transmission. Neural networks
are supposed to be good at this task. One of the major problem in image compression
is long-range dependencies between image patches. There are mainly two approaches
to solve this problem; one is to develop a better residual patch-based encoder, and the
second one is to create an entropy coder capable of collecting long-term dependencies
inside the picture between patches. We address both the problems in this paper and
fuse the two possible solutions to improve compression levels for a given material.
Results of the simulation reveal that the new algorithm works much better in all
parameters than in the current model.
Keywords Image compression · Deep CNN · DST · Neural network · Deep
learning
1 Introduction
Image and video compression play a significant role in delivering high-quality
image/video content under the finite bandwidth of delivery and storage networks
[1]. Video and image redundancies are of fundamental importance for image and
video compression, which includes visual redundancy, spatial redundancy, and statis-
tical redundancy. In addition, temporal consistency in video sequences allows video
compression to achieve a higher compression ratio compared to image compres-
sion [2]. The early methods mainly perform compression for image compression
S. A. Dhondiyal (B )
Graphic Era Deemed to be University, Dehradun, Uttarakhand, India
e-mail: shivashish1234@gmail.com
M. Aeri · M. Manwal · S. Singh
Graphic Era Hill University, Dehradun, Uttarakhand, India
S. Sharma
UPES, Dehradun, Uttarakhand, India
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021
T. P. Singh et al. (eds.), Data Driven Approach Towards Disruptive Technologies,
Studies in Autonomic, Data-driven and Industrial Computing,
https://doi.org/10.1007/978-981-15-9873-9_31
389