JPEG Bit Rate Controller Using Subset Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current JPEG compression methods require multiple iterations to achieve a preset file size, often leading to inefficient use of storage space due to fixed quantization factors and lack of bit rate control, resulting in either excessive compression or wasted space.
Innovation Solution
The method involves selecting a representative subset of an image for initial compression, using clustering analysis to determine a scale factor and Bit Rate Control (BRC) curve, which is then applied to compress the entire image efficiently, eliminating the need for multiple compressions and ensuring the final file size meets the preset target without exceeding it.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If multiple compressions are performed to achieve preset file size, then storage space utilization is improved, but compression time increases and device complexity increases
Solution Approach 1:
The patent performs preliminary compression of a representative subset of the image to estimate the scale factor and generate a BRC curve before compressing the entire image. This preliminary action allows the system to determine optimal compression parameters in advance, enabling single-pass compression of the full image that achieves the target file size without requiring multiple iterative compressions.
Solution Approach 2:
The patent uses feedback from the preliminary compression of the subset to adjust and determine the scale factor and BRC curve for the full image compression. The BRC curve generated from subset compression results provides feedback information that guides the compression of the entire image, ensuring the final output meets the target file size requirements.
2Device complexity
If fixed quantization factor is used throughout the image, then device complexity is reduced, but bit rate control capability is lost
Solution Approach 1:
The patent segments the image into a representative subset and the full image for separate processing. By analyzing the subset separately, the system can determine optimal scale factors and generate BRC curves tailored to specific regions or characteristics, enabling bit rate control while maintaining manageable complexity through structured segmentation of the compression process.
Solution Approach 2:
The patent introduces dynamic bit rate control through the BRC curve, which allows the quantization process to adapt during compression based on the compression progress and target file size. This dynamic adjustment capability enables the system to control the bit rate throughout the compression process, transforming the fixed quantization approach into a dynamic, controllable process.
3Quantity of substance
If multiple compressions are performed, then storage space utilization is improved, but device complexity increases
Solution Approach 1:
The patent performs preliminary compression of a representative subset to estimate compression characteristics and determine the scale factor before compressing the full image. This preliminary action provides the necessary information to achieve the target storage size in a single pass, eliminating the need for multiple iterative compression operations and reducing overall process complexity.
Solution Approach 2:
The patent uses a representative subset as a copy or sample of the full image to determine compression parameters. By analyzing this representative copy, the system can infer the compression characteristics needed for the full image, avoiding the complexity of multiple full-image compressions while still achieving accurate storage space utilization.
Data Source
AI summary
Embodiments are directed towards compressing an image to substantially a preset file size using statistical information obtained from a single subset of the image and an initial compression of the single subset. A representative subset portion of the image is selected based in part on a clustering analysis of the image. The representative subset is then compressed, in one embodiment, twice, in order to obtain statistics useable for the entire image. A scale factor is then determined that may be used in the quantization and for creating a Bit Rate Control (BRC) curve that represents an amount of accumulated bits per Minimal Codec Unit (MCU). During the compression process, the BRC curve is used to prevent accumulating bits from over shooting a final preset file size target.