DocumentCode :
2451488
Title :
A DCT-domain compressed images registration approach
Author :
Xiangyu Yu ; Hengyu Zhao ; Tao Zhang
Author_Institution :
Sch. of Electron. & Inf. Eng. Sch. of Journalism & Commun., South China Univ. of Technol., Guangzhou, China
fYear :
2012
fDate :
16-18 July 2012
Firstpage :
909
Lastpage :
913
Abstract :
Image registration is an important task in image processing and there are large number of methods have been proposed. However, most current approaches are processed in the spatial domain, when being applied on some images in the format of compressed such JPEG or others, de-compress procedure is necessary. In this paper, a compressed domain image registration approach is proposed, in the proposed method, edge detection is processed on DCT domain directly and cross-correlation is applied on the binary edge map. Experimental results showed that the proposed approach is faster than spatial domain method while maintain good registration precision.
Keywords :
data compression; discrete cosine transforms; edge detection; image coding; image registration; DCT-domain compressed images registration approach; JPEG; binary edge map; cross-correlation; decompress procedure; edge detection; image processing; spatial domain; Computational complexity; Discrete cosine transforms; Educational institutions; Image coding; Image edge detection; Image registration; Transform coding;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Audio, Language and Image Processing (ICALIP), 2012 International Conference on
Conference_Location :
Shanghai
Print_ISBN :
978-1-4673-0173-2
Type :
conf
DOI :
10.1109/ICALIP.2012.6376743
Filename :
6376743
Link To Document :
بازگشت