Image haze removal of optimized contrast enhancement based on GPU

Che Lun Hung*, Zhaohui Ma, Chun Yuan Lin, Hsiao Hsi Wang

*此作品的通信作者

研究成果: Chapter同行評審

5 引文 斯高帕斯(Scopus)

摘要

In the domains of computer vision and graphical computation, image haze removal has been a significant issue. By the use of haze removal process, it can significantly improve the visibility of the scene in the image. However, most of the haze removal algorithms bring high computational cost and make algorithms failed in processing huge amount of images. In this paper, we propose a parallel image haze remove algorithm, adopting optimized contrast enhancement approach, to optimize the performance based on GPU platform. The optimization from the proposed algorithm obtains performance acceleration with about 5 times as compared the original version while the haze removal effect is the same. Some haze free images and its original hazy images are shown in the later chapter during this paper. Our work after improvement can process a single picture in a much higher speed after optimization and make it more sufficiently fast for large-scale application which needs image haze removal in computer vision area.

原文English
主出版物標題Lecture Notes in Electrical Engineering
發行者Springer Verlag
頁面53-63
頁數11
DOIs
出版狀態Published - 2016

出版系列

名字Lecture Notes in Electrical Engineering
375
ISSN(列印)1876-1100
ISSN(電子)1876-1119

指紋

深入研究「Image haze removal of optimized contrast enhancement based on GPU」主題。共同形成了獨特的指紋。

引用此