High-Dynamic Range Image Compression via Floating-Point to Integer Conversion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image compression techniques fail to efficiently compress and decompress high-dynamic range images while preserving significant original image quality, particularly in formats like OpenEXR, which requires efficient storage and transmission solutions for digital content.
Innovation Solution
The method involves dividing images into channels, converting floating-point pixel values into integer representations that preserve sign, calculating and applying offsets to store differences between adjacent pixels, and padding blocks for even divisibility, allowing for efficient compression and reconstruction of high-dynamic range images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If floating point representations of pixel values are used in high-dynamic range images, then image quality and dynamic range are improved, but data size and storage requirements increase
Solution Approach 1:
The patent changes the parameter representation from floating point to integer format. Specifically, it converts floating point pixel values to integer representations by calculating offsets based on perceptual differences, thereby reducing data size while maintaining image quality through intelligent parameter transformation
Solution Approach 2:
The patent segments the image into blocks and processes each block independently. It divides the image channel into multiple blocks, calculates offsets for each block based on local pixel value distributions, and stores only the offset values rather than complete pixel data, thereby reducing overall data quantity while preserving quality
2Quantity of substance
If compression algorithms reduce data size significantly, then storage and transmission efficiency improve, but image quality and reconstruction accuracy deteriorate
Solution Approach 1:
The patent introduces offset values as intermediary representations between the original floating point pixel values and the compressed integer data. These offsets act as mediators that capture the essential perceptual differences while enabling efficient compression, allowing reconstruction of high-quality images from compressed data
Solution Approach 2:
The patent transforms the compression approach by changing from direct pixel value compression to offset-based compression. By representing only the differences (offsets) between adjacent pixel values rather than absolute values, it achieves better compression ratios while maintaining reconstruction accuracy through the offset correction process
3Speed
If image data is compressed to enable near real-time transmission, then transmission speed improves, but processing complexity and computational requirements increase
Solution Approach 1:
The patent segments the image processing into independent block operations. Each block can be processed and compressed independently using simple offset calculation algorithms, which reduces the complexity of any single operation and enables parallel processing, thereby achieving near real-time transmission without excessive computational burden
Solution Approach 2:
The compression algorithm uses the image data itself to generate the compression parameters. By calculating offsets based on local pixel value distributions and perceptual differences within each block, the system performs self-optimizing compression without requiring complex external processing or heavy computational overhead
Data Source
AI summary
A computer-implemented method includes dividing an image into one or more image channels for image compression. The method also includes dividing one or more of the image channels into one or more blocks. At least one of the blocks includes floating point representations of pixel values included in the block. The method also includes converting the floating point representations of pixel values into integer representations such that the sign of each floating point representation is preserved. The method also includes storing the difference of adjacent integer representations as a compressed version of the image.