This paper presents an efficient compression method for the integral images where an adaptive 3D block construction is employed for three dimensional discrete cosine transform (3D-DCT). The existing 3D-DCT based techniques take disadvantages in 3D block construction as applying a fixed block size and a fixed scanning regardless of the image characteristics. In this paper, therefore, we propose more flexible construction, i.e. four different block size modes and two scanning modes which are adaptively selected for coding of elemental images in an integral image. Experimental results show that the proposed method gives significant improvement in coding efficiency. In particular, at the high bit-rates, the proposed method is more efficient since the overhead bits for signalling of a block size and a scanning mode take less part of the total bits. © 2010 Springer-Verlag.
CITATION STYLE
Jeon, J. I., & Kang, H. S. (2010). 3D DCT based compression method for integral images. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6454 LNCS, pp. 659–668). https://doi.org/10.1007/978-3-642-17274-8_64
Mendeley helps you to discover research relevant for your work.