Dynamic Region Scaling for Digital Image Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional digital image compression systems for industrial radiography result in undesirable quantization effects and slow reconstruction due to scaling of raw volumetric data from 32-bit to 8-bit or 16-bit formats, leading to loss of data/image quality and inefficient storage and transmission.
Innovation Solution
The proposed method involves identifying and scaling image data based on characteristics such as gray value range, noise, and contrast to assign appropriate bit values (8-bit or 16-bit) to regions of the image, allowing for dynamic rescaling to 32-bit values for reconstruction, thereby reducing quantization and improving storage efficiency and reconstruction speed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If raw volumetric data is scaled from 32-bit to 8-bit or 16-bit formats for compression, then storage requirements and transmission time are reduced, but quantization effects occur and image quality is lost
Solution Approach 1:
The image data is divided into multiple regions based on gray value ranges, and each region is compressed using different bit depths (8-bit or 16-bit) depending on its characteristics. This segmentation allows optimal compression without uniform quality loss across the entire image.
Solution Approach 2:
Different regions of the image are assigned different compression qualities based on their specific characteristics. Regions with narrower gray value ranges use 8-bit compression, while regions with wider ranges use 16-bit compression, ensuring appropriate quality preservation where needed.
2Quantity of substance
If raw volumetric data is scaled from 32-bit to 8-bit or 16-bit formats for compression, then storage requirements are reduced, but reconstruction speed becomes slow
Solution Approach 1:
The image is segmented into regions with different compression levels, allowing for selective decompression. This segmentation enables faster reconstruction by processing only the necessary regions at appropriate resolutions rather than uniformly decompressing the entire image.
3Device complexity
If uniform bit depth compression is applied to the entire image, then processing complexity is reduced, but dynamic range is limited and data loss occurs
Solution Approach 1:
The patent applies different bit depths to different regions based on their gray value characteristics. This local differentiation preserves dynamic range and prevents data loss in regions that require higher precision while maintaining simplicity in regions where lower precision is sufficient.
Solution Approach 2:
The compression parameters (bit depth) are dynamically changed based on the characteristics of each image region. This parameter adaptation allows the system to optimize between compression efficiency and data preservation without using a single fixed parameter for the entire image.
Data Source
AI summary
Described herein are examples of imaging systems for digital image processing. For example, techniques are disclosed for dynamic compression of individual regions representing pixels and/or voxels of high-resolution digital images. During image data compression, first regions may be scaled based on a first scaling value, whereas second region may be scaled based on a second scaling value. During image data decompression and image reconstruction, a third scaling value is applied to both first and second regions.