Embedded Compression Algorithm Study for alpha RGB Graphics Content
نویسندگان
چکیده
Preface In the past nine months, I carried out my the final year project of KTH international master's in SoC (System-on-Chip) at NXP Semiconductors Research (NXP), the Netherlands. Thanks to my supervisors, Prof. Axel Jantsch in KTH and Senior Scientist Tomas Henriksson in NXP. Also thanks to Pieter van der Wolf and Bram Riemens in NXP. During my internship they gave me a lot of help on the technical issues. Thanks to my family. Without their support, I would not have had the chance to carry out my internship in the Netherlands. Also thanks to my friends, Abstract The bandwidth requirement between SoC and external DRAM is increasing rapidly. Since the external DRAM cannot always meet the requirements, embedded compression is introduced to decrease the bandwidth requirement for video and image data. The existing embedded compression algorithm (EA) in NXP is designed for the YUV or RGB format video and image data with natural content. It is able to compress the input data visually loss-lessly at compression ratio of 1.6 to 2.0. Our work shows that the EA is not visually lossless when compressing the alpha RGB format video and image data with graphics content, e.g. overlays in mobile phones, at a high compression ratio. The goal of this master thesis project is to address a better embedded compression algorithm for the alpha RGB format video and image data with graphics content than the EA. The new algorithm must be able to compress the alpha RGB video and image data visually lossless at a higher compression ratio. Deriving the new embedded compression algorithm from the EA is my main contribution. Several modifications are introduced and evaluated. New methods are also introduced for the alpha RGB format video and image data with graphics content. In the end of this thesis project, a new embedded compression algorithm optimized for the alpha RGB format video and image data with graphics content is developed. This algorithm includes several modifications from the EA and a new method. This algorithm has a PSNR improvement of around 2 dB for all four components (Red, Green, Blue, and Alpha) for a test image at compression ratio of 2.667, which is the typical visually lossless compression ratio for the alpha RGB format video and image data with graphics content. The implementation costs of embedded compression algorithms that are involved during the development of the final algorithm are estimated and …
منابع مشابه
YUV Conversion Subsampling YUV - > RGB Conversion Upsampling
In this paper, we present a new wavelet-based compression algorithm for still color images especially for low bitrates. The coding algorithm distinguishes itself by a high adaptivity to the image content. It yields an embedded bitstream with the advantage of spatial and quality scalability. The performance of the wavelet codec is demonstrated in comparison with the JPEG standard for diierent co...
متن کاملContent Aware Texture Compression
Efficient and high quality texture compression is important because computational powers are always limited, especially for embedded systems such as mobile phones. To access random texels instantly, down sampling and S3TC are the most commonly used methods due to the fixed compression ratio at every local region. However, the methods are content oblivious, which uniformly discard texel informat...
متن کاملCompression of Compound Images and Video for Enabling Rich Media in Embedded Systems
It is possible to improve the features supported by devices with embedded systems by increasing the processor computing power, but this always results in higher costs, complexity, and power consumption. An interesting alternative is to use the growing networking infrastructures to do remote processing and visualization, with the embedded system mainly responsible for communications and user int...
متن کامل2 Megapixel single-chip camera module
■ 1600H x 1200V active pixels ■ Class leading 30 fps UXGA progressive scan ■ 2.2 μm pixel size, 1/3.8 inch optical format ■ Quad-element plastic lens, F# 3.2, 52° horizontal field of view ■ 8.0 x 8.0 x 5.55 mm ultra low profile fixed focus camera module with embedded passives ■ RGB Bayer color filter array ■ Integrated 10-bit ADC ■ Integrated digital image processing functions, including defect...
متن کاملRandom Walks for Interactive Alpha-matting
Interactive, efficient, methods of foreground extraction and alpha-matting are of increasing practical importance for digital image editing. Although several new approaches to this problem have recently been developed, many challenges remain. We propose a new technique based on random walks that has the following advantages: First, by leveraging a recent technique from manifold learning theory,...
متن کامل