China
Africa
Explore chapters and articles related to this topic
HDR Images Compression
Published in Francesco Banterle, Alessandro Artusi, Kurt Debattista, Alan Chalmers, Advanced High Dynamic Range Imaging, 2017
Francesco Banterle, Alessandro Artusi, Kurt Debattista, Alan Chalmers
To solve this issue, in 2012, the Joint Photographic Experts Group (JPEG), formally known as ISO/IEC JTC1/SC29/WG1, began the development of a new standard technology called JPEG XT (ISO/IEC 18477) [327]. The JPEG XT image coding system is organized into nine parts that define the baseline coding architecture (the legacy JPEG code stream 8-bit mode). This is an extensible file format specifying a common syntax for extending the legacy JPEG, and application of this syntax for coding integer or floating point samples within 8–16 bits precision [24]. This coding architecture is further refined to enable lossless and near-lossless coding, and it is complemented by an extension for representing alpha-channels. Thanks to its flexible layered structure, the JPEG XT capabilities can be extended into novel applications such as omni-directional photography, animated images, structural editing as well as privacy, and security. These are currently under examination and development [326].
Lossless HDR Image Compression by Modulo Encoding
Published in IETE Journal of Research, 2023
A. S. Anand Swamy, N Shylashree
Several techniques are available in the literature for fully lossless compression of HDR images. JPEG-LS (ISO/IEC 14495-1) [2] was the lossless compression method published by the JPEG group in 1999 for 16-bit images. The improved version, JPEG-XT (ISO/IEC IS 18477-8) [3] has been published in 2016. Updated JPEG-2000 was published in 2019 (ISO/IEC 15444-1:2019) [4]. All the above JPEG versions can handle images of bit-depth up to 16 bits [5,6]. To accommodate 32-bit images, in [5,6], the logarithm transforms and its inverse are used for additional pre-processing and post-processing whereas in [7], a two-layer approach is adopted. In [8], the floating point intensity values of HDR images are mapped into integers using raw binary formats. Then JPEG-2000 is used for compression and during decompression, integer-to-float reconversion is applied to get back the original HDR image. In [9], the integer formats are obtained by splitting the 32-bit pixel intensity values. Histogram packing techniques are used in [10–12] to generate index images and integer indices which, in turn, utilize lossless JPEG encoding.