Lossless JPEG

About: Lossless JPEG is a research topic. Over the lifetime, 2415 publications have been published within this topic receiving 51110 citations. The topic is also known as: Lossless JPEG & .jls.

CUDA lossless data compression algorithms: A comparative study

Abstract: As data is being produced in an unprecedented rate, lossless data compression has become an important step in data storage and transmission processing as it helps to reduce the resource usage in these fields. However, the current bottlenecks of existing lossless data compression tools causes the compression and decompression process to be very time consuming for large-scale data processing. General purpose computing on graphic processing units (GPUs) introduces new opportunities where parallelism is available and this could be the solution to address the bottlenecks of the data compression. Several parallel lossless data compression algorithms on GPU have been proposed but there isn't much comparative study conducted on the performance among them. This paper examines the existing CUDA lossless data compression algorithms and compares their performance. These CUDA data compression algorithms are evaluated and tested on different datasets of different sizes. The article is concluded by a comparison of these CUDA lossless data compression algorithms from different aspects.

Lossless volume compression using a variation of Lempel-Ziv '77

Abstract: Volume data sets are becoming increasingly large and numerous with advances in acquisition methods and in computers. To date, the most commonly used lossless compression software for volume data sets are Lempel-Ziv-based applications designed for text compression. We present an improved lossless volume compression algorithm based on Lempel-Ziv '77.

Fast JPEG Color Space Conversion on Shared Memory

Abstract: With the increasing popularity of multimedia devices and applications, JPEG (Joint Photographic Experts Group) images have been being used in almost all kinds of digital devices. In JPEG compression, color space conversion has become a very important role in the image acquisition, display and the transmission of the color information. This paper experiments the performance of JPEG color space conversion on shared memory; for this purpose, we utilize OpenMP shared programming language and a RGB image is transformed into a luminance-chrominance color space such as YCbCr. The implementation results show 2 times of performance improvements on the processing speed compared with the serial implementation.

Improving JPEG Compression with Regression Tree Fields

Abstract: Storing digital images is a common task in a variety of domains, ranging from family photo albums to medical imaging. Where many images need to be saved, efficient coding is essential. JPEG, as a wellestablished codec for still-image compression, is used in a variety of use cases. This thesis analyzes improvements on compression performance by introducing a prediction step through regression tree fields. RTFs are used to predict coefficients of JPEG’s discrete cosine transform, which allows to encode only differences of small magnitude instead of the original coefficients. The adapted compression scheme’s degrees of freedom (e.g. prediction order, RTF loss function, and encoding procedure) are analyzed with respect to their influence on compression performance. This thesis’ results show that the general idea of predicting coefficient images can improve the compression performance significantly. Especially loss-specific optimization poses a great gain, which induced the development of an entropybased loss function. Furthermore, the results are evident that RTF models are not suited for predictions in the frequency domain with respect to compression, which suggests further work with different models and image representations based on the developed scheme.