Haar Wavelet Pre-processing for 4K Video Compression

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current video compression techniques struggle to keep up with the increasing demands of high-resolution video data traffic, particularly with the advent of 4K and 8K UHD content, leading to quality declines and costly infrastructure investments in cable and fiber systems, especially in rural areas where infrastructure development is challenging.

Innovation Solution

The implementation of Haar Wavelet-based pre-processing and post-processing methods that reduce video frame size before compression, using a special purpose computer to apply Haar Wavelet Transform for size reduction and Inverse Haar Transform for expansion, enhancing compression efficiency without degrading video quality, allowing for high-quality video transmission at significantly lower bitrates.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If conventional video compression techniques are used for 4K and 8K content, then video quality can be maintained, but transmission bandwidth and storage requirements increase significantly

Engineering Contradiction:
Improvevideo qualityVSAvoiddata transmission volume
Core Design Contradiction:
Manufacturing precisionVSQuantity of substance

Solution Approach 1:

The patent applies Haar Wavelet Transform to segment video frames into different frequency sub-bands (approximation and detail coefficients). This segmentation allows selective compression of less important high-frequency detail components while preserving low-frequency approximation components, achieving significant compression ratios while maintaining perceptual video quality for 4K and 8K content

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transforms video data from spatial domain to frequency domain using wavelet transform, changing the representation parameters. This parameter transformation enables more efficient compression by exploiting the energy compaction property of wavelets, where most signal energy is concentrated in fewer coefficients, reducing the data volume required for 4K and 8K video transmission

Inventive Principle:
Principle #35Parameter changes

2Productivity

If cable and fiber infrastructure is expanded to handle increased video data traffic, then transmission capacity improves, but infrastructure investment and complexity increase

Engineering Contradiction:
Improvetransmission capacityVSAvoidinfrastructure complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent performs pre-processing of video content using wavelet transform before transmission. This preliminary action compresses the video data in advance, reducing the bandwidth requirements for cable and fiber infrastructure, thereby avoiding the need for expensive infrastructure expansion while maintaining transmission capacity for 4K and 8K content

Inventive Principle:
Principle #10Preliminary action

3Manufacturing precision

If video frame resolution is increased for 4K and 8K content, then video quality improves, but compression efficiency decreases

Engineering Contradiction:
Improvevideo resolutionVSAvoidcompression efficiency
Core Design Contradiction:
Manufacturing precisionVSProductivity

Solution Approach 1:

The patent segments high-resolution 4K and 8K video frames into wavelet sub-bands, allowing differential compression where coarse approximation coefficients are retained at full precision while fine detail coefficients are coarsely quantized or discarded. This segmentation enables maintaining high resolution for perceptually important components while compressing less critical details, preserving compression efficiency at high resolutions

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP3170306B1Multilevel video compression, decompression, and display for 4k and 8k applications
Publication Date: 2020.01.29 YAMZZ IP BV
  • EP3170306B1 patent drawingFigure 1A~1C
  • EP3170306B1 patent drawingFigure 2A~2C
  • EP3170306B1 patent drawingFigure 3A~3F

AI summary

Embodiments of this invention include computer-implemented methods for use of Haar wavelet transforms (HWT) to pre-process video frames that can then be compressed using a codec to produce a compressed video frame. A compressed video frame can then be transmitted, decompressed, post-processed and displayed in their original size and quality, thereby producing real-time high-quality reproduction of video sequences. Devices can implement methods of this invention and include mainframe computers, desktop computers, personal computers, laptop computers, tablet computers, wireless computers, television sets, set top boxes, cellular telephones, and computer readable media. Computer implemented steps can be easily and efficiently implemented in multicore parallel hardware architectures.