Title: Comparison of lossy image compression techniques
Authors: Sriram Kannan; Swetha Saseendran; Sowmya Ramesh; Sriram Rajasekar; Shriya Baskaran
Addresses: Sri Sivasubramaniya Nadar (SSN) College of Engineering – Campus, Rajiv Gandhi Salai (OMR), Kalavakkam – 603 110, Tamil Nadu, India ' Sri Sivasubramaniya Nadar (SSN) College of Engineering – Campus, Rajiv Gandhi Salai (OMR), Kalavakkam – 603 110, Tamil Nadu, India ' Sri Sivasubramaniya Nadar (SSN) College of Engineering – Campus, Rajiv Gandhi Salai (OMR), Kalavakkam – 603 110, Tamil Nadu, India ' Sri Sivasubramaniya Nadar (SSN) College of Engineering – Campus, Rajiv Gandhi Salai (OMR), Kalavakkam – 603 110, Tamil Nadu, India ' Sri Sivasubramaniya Nadar (SSN) College of Engineering – Campus, Rajiv Gandhi Salai (OMR), Kalavakkam – 603 110, Tamil Nadu, India
Abstract: Image compression is crucial task in the current era. This is mainly due to the ubiquitous nature of images in the applications we use today and with the need to transmit these images without spending fortunes while also maintaining the quality up to a certain mark. This paper will provide an insight into the problems of image compression, comparing different models. In this paper, autoencoders, SVDs, GANs, JPEG, K-means, and PCA models are used to perform image compression. It further aims to compare the performance of the aforementioned models to understand which model gives the most preferable output.
Keywords: lossy image compression; joint photographic experts' group; JPEG; principal component analysis; PCA; autoencoders; single value decomposition; SVD; generative adversarial network; GAN; K-means.
International Journal of Hybrid Intelligence, 2023 Vol.2 No.2, pp.151 - 174
Received: 04 Nov 2022
Accepted: 26 Nov 2022
Published online: 06 Mar 2023 *