Image Processing Device Edge Block Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Dual gamma correction in image processing leads to poor compression efficiency and high distortion, particularly affecting edge pixels where the edge information may disappear during the mapping process.

Innovation Solution

An image processing method that separates edge blocks from other blocks, skipping dual gamma correction and specific encoding/decoding processes for edge blocks, allowing them to maintain their characteristics and preventing edge disappearance, while improving encoding and decoding efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Illumination intensity

If dual gamma correction is performed on all blocks, then viewing angle is improved, but compression efficiency deteriorates and edge information is lost

Engineering Contradiction:
Improveviewing angleVSAvoidcompression efficiency
Core Design Contradiction:
Illumination intensityVSProductivity

Solution Approach 1:

The patent segments the image into different block types (edge blocks and non-edge blocks) and applies different processing methods to each. Edge blocks skip dual gamma correction while non-edge blocks undergo the full correction process, thereby maintaining compression efficiency for most areas while preserving viewing angle improvements where applicable.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies local quality by treating different regions of the image differently based on their characteristics. Edge blocks are identified and processed separately without dual gamma correction, while other blocks receive the correction. This localized approach optimizes both compression efficiency and viewing angle without forcing the same treatment on all blocks.

Inventive Principle:
Principle #3Local quality

2Productivity

If mapping process is performed to improve compression efficiency, then compression efficiency is improved, but edge texture disappears

Engineering Contradiction:
Improvecompression efficiencyVSAvoidedge texture
Core Design Contradiction:
ProductivityVSLoss of information

Solution Approach 1:

The patent segments blocks into edge and non-edge categories and applies mapping process selectively. Only non-edge blocks undergo the mapping process to improve compression efficiency, while edge blocks skip this step to preserve edge texture and geometric information.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent extracts edge blocks from the general processing pipeline. By identifying and separating edge blocks, the system excludes them from the mapping process that would otherwise distort their characteristics, thereby preserving edge information while still benefiting from compression efficiency improvements in other regions.

Inventive Principle:
Principle #2Taking out (Extraction)

3Measurement precision

If dual gamma correction is applied to correct color deviation, then color accuracy is improved, but pixel value differences between adjacent subpixels increase

Engineering Contradiction:
Improvecolor accuracyVSAvoidpixel value uniformity
Core Design Contradiction:
Measurement precisionVSLoss of information

Solution Approach 1:

The patent applies local quality by differentiating between edge blocks and non-edge blocks. Non-edge blocks receive dual gamma correction to improve color accuracy, while edge blocks skip this process to maintain pixel value uniformity and prevent artificial edge creation from varying gamma voltages.

Inventive Principle:
Principle #3Local quality

Data Source

PatentUS12080036B2Image processing device and image processing method thereof
Publication Date: 2024.09.03 NOVATEK MICROELECTRONICS CORP
  • US12080036B2 patent drawing
  • US12080036B2 patent drawing
  • US12080036B2 patent drawing

AI summary

An image processing method including the following steps is provided. An image information of an image is received, wherein the image includes a plurality of blocks and the image information includes a plurality of pixel information of each block. A dual gamma correction is performed on a first group of blocks of the image to obtain one or more corrected blocks and the dual gamma correction is skipped on a second group of blocks of the image to obtain a plurality of uncorrected blocks. A first encoding process is performed on the one or more corrected blocks to obtain a plurality of first encoded blocks. A second encoding process different from the first encoding process is performed on the plurality of uncorrected blocks to obtain a plurality of second encoded blocks.