Providing a Desired Compression Ratio for Better Portable Graphics Encoder of Color Images

Lukin, Vladimir and Li, Fangfang (2023) Providing a Desired Compression Ratio for Better Portable Graphics Encoder of Color Images. In: Advances and Challenges in Science and Technology Vol. 1. B P International, pp. 98-111. ISBN 978-81-19491-91-9

Full text not available from this repository.

Abstract

In this chapter, the task of providing a desired compression ratio (CR) is considered for the lossy compression of color images by better portable graphics (BPG) encoder. This encoder is characterized by many benefits compared to other modern encoders according to rate-distortion performance and some other important facilities. However, the produced CR for a given value of a parameter that controls compression (PCC) for the considered encoder varies in very wide limits depending on image content and the format the original image is presented. In practice, it can be desired to provide a desired CR for color images compressed by BPG. We show that this can be done using a recently proposed two-step method based on average rate-distortion curves and PCC refining at the second step. The applicability of the two-step approach to providing a desired CR is analyzed and proven for the formats 4:2:0, 4:2:2, and 4:4:4. It is demonstrated that, due to the second step, the accuracy of providing a desired CR radically improves. The variance of the residual errors of the provided CR is decreased tens of times. It is also shown that accuracy depends on a desired CR and the variance of residual errors is larger for the larger desired CR. Some other properties of the residual errors are studied. Two sets of test color images are exploited in our studies. The first set is used to obtain the average dependence of CR on PCC and get an idea of the accuracy of the proposed method. The second set is employed to verify the method's performance. It is shown that the results for both sets are in good agreement.

Item Type: Book Section
Subjects: STM Repository > Multidisciplinary
Depositing User: Managing Editor
Date Deposited: 07 Oct 2023 09:41
Last Modified: 12 Dec 2023 12:42
URI: http://classical.goforpromo.com/id/eprint/3841

Actions (login required)

View Item
View Item