Image Compression Device Chroma Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing 3D graphic systems using floating-point number representations for image data processing and storage face challenges with high memory occupancy and processing speed due to the large data amount handled, despite achieving a high dynamic range of luminance levels.
Innovation Solution
The implementation of an image compression and decompression device that utilizes a floating-point texture-loading unit, a shader, an RGB to YCrCb compressor, and a parameter adjusting unit to discard non-sensitive chroma information and truncate the mantissa of the common chroma value, reducing data amount and memory occupancy while maintaining high dynamic range.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If floating-point number representation is used to process and store image data, then the dynamic range of luminance levels is expanded to 100,000:1, but the occupied memory space increases and processing speed decreases
Solution Approach 1:
The patent segments the chroma information processing by separating the common chroma value (applied to all pixels) from individual chroma values. This segmentation allows the system to process and store only one common chroma value for multiple pixels instead of individual chroma values for each pixel, significantly reducing data amount while maintaining HDR capability
Solution Approach 2:
The patent changes the precision parameter of chroma information by truncating the mantissa of the common chroma value. This parameter change reduces the data size of chroma information while the human eye's insensitivity to chroma variations ensures that image quality remains acceptable, thus improving processing speed and reducing memory usage
2Illumination intensity
If floating-point number representation is used to process and store image data, then the dynamic range of luminance levels is expanded to 100,000:1, but the occupied memory space increases
Solution Approach 1:
The patent extracts and discards non-sensitive chroma information from each pixel, keeping only the common chroma value that applies to all pixels. This extraction process removes redundant data that contributes to high memory occupancy while preserving the essential HDR luminance information
Solution Approach 2:
The patent reduces the precision parameter of chroma information by truncating the mantissa of the common chroma value. This parameter reduction decreases the memory space required to store chroma information while maintaining acceptable image quality due to human visual system characteristics
Data Source
AI summary
An image compression device includes a floating-point texture-loading unit, a shader, a RGB to YCrCb compressor, and a parameter adjusting unit. The floating-point texture-loading unit receives raw image data that includes N number of RGB pixels. The shader receives the raw image data and shades the N number of RGB pixels. The RGB to YCrCb compressor receives the shaded image data and calculates the brightness/chroma value of each of the RGB pixels, wherein the chroma value of the RGB pixel having the smallest brightness value among all the RGB pixels is selected as a common chroma value for all the RGB pixels. The parameter adjusting unit receives and adjusts N number of brightness values of the RGB pixels and the common chroma value and stores them in a form of compressed image data into a memory.