Image Compression via Adaptive Coding Mode Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image compression techniques incur significant system overhead and time, especially in applications with limited memory, such as laser printing, where lossy compression is often necessary to achieve substantial data reduction without compromising image quality.
Innovation Solution
A method that determines the number and locations of non-zero AC coefficients in a transform block to select an appropriate coding mode, allowing for bit allocation optimization, thereby compressing images efficiently using a 4×4 transform that approximates DCT and performs quantization through bit-shifts, eliminating the need for multiplication or division, and ensuring near-orthogonal de-correlation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If lossy compression techniques are used to achieve substantial data reduction, then the data compression ratio is improved, but the system overhead and processing time increase significantly
Solution Approach 1:
The image is divided into multiple blocks, and each block is processed independently through transform and quantization operations. This segmentation allows the compression system to handle data in manageable units, reducing overall system complexity while maintaining high compression ratios through localized processing of transform coefficients.
Solution Approach 2:
The patent applies quantization by scaling transform coefficients and uses bit-shift operations instead of multiplication or division. By changing the mathematical operations to simpler parameter-based transformations, the system achieves substantial data reduction while minimizing processing overhead and computational complexity.
2Quantity of substance
If complex compression algorithms are used to achieve higher compression ratios, then the data reduction is improved, but the processing time increases
Solution Approach 1:
The patent replaces complex multiplication and division operations with simpler bit-shift operations during quantization and decompression. This substitution of mathematical mechanisms significantly reduces processing time while maintaining the same data reduction effectiveness, as bit-shifts are computationally less intensive than traditional arithmetic operations.
Solution Approach 2:
The transform and quantization operations are performed in advance during the compression phase, organizing data into a format that enables rapid decompression. By performing the computationally intensive transformations beforehand, the system achieves both high compression ratios and fast decompression speeds when needed.
3Volume of stationary object
If memory is reduced to meet system constraints, then the device size is improved, but the ability to store uncompressed images deteriorates
Solution Approach 1:
The patent transforms image data from spatial domain to frequency domain using transform operations, then applies quantization to reduce the precision of coefficient representations. This parameter transformation allows the system to store significantly compressed image data in limited memory while preserving essential visual information, effectively increasing storage capacity within fixed memory constraints.
Data Source
AI summary
In a method for compressing an image, a number of non-zero AC coefficients for a component transform and the locations of the non-zero AC coefficients in a transform block are determined. One of a plurality of coding modes is selected in response to the number of non-zero AC coefficients and the locations of the non-zero AC coefficients satisfying predetermined criteria, where the plurality of coding modes comprise different bit allocations from each other. In addition, the locations and values of the non-zero AC coefficients and a DC coefficient for the component transform are coded according to the selected coding mode, to thereby compress the image.