Image Data Processing Using Wavelet Transform and Tree Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image compression methods are inadequate for achieving significant reductions in data rates for high-definition images, especially for applications requiring low bit rates below 10Mb/s, as they only partially alleviate the bandwidth reduction issue and do not scale well with varying motion and resolutions.
Innovation Solution
The use of a wavelet transform-based compression system that employs a tree structure for encoding data bits, allowing for efficient lossless or lossy compression by combining data bits and excluding redundant information, along with a color space transformation from RGB to YUV to take advantage of human vision characteristics, enabling scalable and real-time compression.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If traditional compression methods (interlaced images, reduced chrominance) are used, then bandwidth is partially reduced, but compression ratio is insufficient for high-definition images at low bit rates
Solution Approach 1:
The image is divided into 8x8 pixel blocks, and each block is processed independently through the compression algorithm. This segmentation allows the system to apply compression techniques to manageable units while maintaining overall image quality and enabling parallel processing for real-time performance.
Solution Approach 2:
The patent applies two-dimensional Discrete Cosine Transform (DCT) to convert spatial domain image data into frequency domain coefficients. This dimensionality change transforms the representation from pixel values to frequency coefficients, enabling efficient compression by concentrating image energy into fewer coefficients that can be quantized and encoded more compactly.
2Loss of energy
If aggressive compression is applied to achieve bit rates below 10Mb/s, then bandwidth is reduced, but image quality may deteriorate
Solution Approach 1:
The patent applies different quantization levels to different frequency coefficients based on their importance. Low-frequency coefficients (which represent the main image content) are quantized with higher precision, while high-frequency coefficients (which represent fine details) are quantized with lower precision. This local quality approach maintains perceived image quality while achieving aggressive compression ratios.
Solution Approach 2:
The system dynamically adjusts quantization parameters and compression settings based on the complexity of the image content and the target bit rate. By changing parameters such as quantization step size, transformation depth, and encoding precision, the system can adapt to achieve the desired balance between bit rate and image quality for different high-definition formats.
3Loss of time
If real-time processing is required for high-definition images, then latency is minimized, but processing complexity increases
Solution Approach 1:
By dividing the high-definition image into 8x8 blocks, the patent enables parallel processing of multiple blocks simultaneously. This segmentation reduces the computational burden on any single processing unit and allows for pipelined implementation, achieving real-time processing of high-definition video streams with minimal latency.
Solution Approach 2:
The patent replaces complex spatial-domain processing operations with more efficient frequency-domain operations using DCT. The transform concentrates image energy into fewer coefficients, reducing the number of operations needed for compression and enabling real-time processing of high-definition content with acceptable hardware complexity.
4Loss of information
If color space transformation from RGB to YUV is applied, then compression efficiency is improved by exploiting human vision characteristics, but processing overhead increases
Solution Approach 1:
The patent transforms the color representation from RGB to YUV, changing the parameter space to separate luminance (Y) from chrominance (U, V) components. This parameter change exploits the human visual system's higher sensitivity to luminance variations, allowing for differential compression where the luminance channel is processed at higher quality than the chrominance channels, improving overall compression efficiency.
Data Source
Figure 1~3
Figure 4~7
Figure 5
AI summary
Methods of data encoding using trees formed with logic gates are described which lead to spatial compression of image data. Data encoding is achieved using a five-level wavelet transform, such as the Haar or the 2/10 transform. A dual transform engine is used, the first and engine being used for the first part of the first-level transform, the second part of the first-level transform and the subsequent-level transforms being performed by the second transform engine within a time interval which is less than or equal to the time taken by the first transform engine to effect the part-transform. Each bit plane of the resulting coefficients is then encoded by forming a tree structure from the bits and OR logical combinations thereof. Redundant data are removed from the resulting tree structure, and further data can be removed by using a predetermined compression profile. The resulting blocks of compressed data are of variable length and are packaged with sync words and index words for transmission so that the location and identity of the transformed data blocks can be determined from the received signal.