Lossless Image Compression Using Adaptive Spatial Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current lossless image compression methods, such as JPEG-LS, are inefficient for high-definition images due to their focus on processing speed rather than maximum compression ratio, and new standards like MPEG for ultra-high definition images involve loss and complexity, necessitating a low-complexity lossless compression method.
Innovation Solution
A lossless image compression method that extracts residual information by decoding a bit stream, generates prediction values based on neighbor pixel correlations, and updates lookup tables according to frequency, allowing for efficient encoding and decoding of high-definition images through directional and self-correlation spatial predictions, adaptive variable length coding, and image segmentation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If JPEG-LS compression method is used, then processing speed is improved, but compression ratio is insufficient for high-definition images
Solution Approach 1:
The patent implements adaptive prediction mode selection that dynamically adjusts the prediction method based on image characteristics and processing stage. The system switches between different prediction modes (spatial prediction, temporal prediction, adaptive prediction) to optimize both speed and compression ratio for high-definition images
Solution Approach 2:
The patent changes the prediction parameter selection based on image quality requirements. For high-definition images, the system uses more sophisticated prediction parameters and updates lookup tables adaptively to achieve higher compression ratios while maintaining processing speed
2Speed
If conventional compression algorithms are used for high-definition images, then processing speed is maintained, but compression efficiency deteriorates
Solution Approach 1:
The patent performs preliminary actions by pre-calculating prediction values and pre-updating lookup tables during the encoding process. This allows the decoder to operate more efficiently without requiring complex real-time calculations, thus maintaining speed while improving compression efficiency for high-definition images
Solution Approach 2:
The patent implements feedback mechanisms where the encoder and decoder continuously adjust prediction modes and lookup table entries based on the actual image data and compression performance. This adaptive feedback loop optimizes compression efficiency while maintaining processing speed
3Loss of information
If loss-based compression standards like MPEG are used for ultra-high definition, then compression ratio is improved, but complexity increases
Solution Approach 1:
The patent extracts only the essential prediction and compression functions from complex loss-based standards like MPEG. By isolating the core prediction mechanisms and removing unnecessary complex lossy processing, the system achieves high compression ratios for ultra-high definition images with reduced complexity
Solution Approach 2:
The patent uses copying strategies by creating simplified versions of prediction models that replicate the effectiveness of complex standards without their computational burden. The system copies the predictive power of MPEG-like standards while using lighter-weight algorithms that reduce device complexity
4Loss of information
If lossless compression is implemented with high compression ratio, then information is preserved, but processing complexity increases
Solution Approach 1:
The patent segments the compression process into distinct prediction modes and processing stages. By dividing the complex task into manageable segments (spatial prediction, temporal prediction, adaptive prediction), the system achieves high compression ratios while keeping individual processing steps simple and manageable
Data Source
AI summary
A lossless image compression method for a high definition image is provided. The image compression method generates a minimum value of values produced by subtracting a current virtual pixel value from an average of neighbor pixels designated by a plurality of directions, as the prediction value of a current pixel. Thus, the image compression more efficient than the current JPEG-LS can be achieved.