Multi-band blending of aerial images using GPU acceleration

Aerial images are densely captured with high resolution. It is an important research topic to efficiently process the aerial images. Benefited from its powerful computation power, GPU acceleration techniques have been widely used in image processing, and many traditional techniques achieved real-tim...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP BMEI) s. 1 - 5
Hlavní autoři:	Zhao, Nan, Zheng, Xinqi
Médium:	Konferenční příspěvek
Jazyk:	angličtina
Vydáno:	IEEE 01.10.2017
Témata:	Acceleration aerial image algorithm parallelization GPU acceleration Graphics processing units image blending image decomposition Image resolution Laplace equations Mathematical model Memory management Random access memory
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Popis
Shrnutí:	Aerial images are densely captured with high resolution. It is an important research topic to efficiently process the aerial images. Benefited from its powerful computation power, GPU acceleration techniques have been widely used in image processing, and many traditional techniques achieved real-time processing through using the GPU acceleration. In this research, we study the aerial image blending problems using GPU acceleration. We parallelized the Multi-band blending algorithm and achieved 30 times acceleration compared with CPU implementation.
DOI:	10.1109/CISP-BMEI.2017.8302068