Sub-pixel Rendering for Display Devices to Reduce Power and Complexity

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing display technologies face challenges in reducing power consumption and improving display quality when using a number of pixels smaller than the input image resolution, often resulting in color errors and blurring due to complex computational processes and the need for multiple line memories, which increases product costs.

Innovation Solution

An image processing method that separates and processes R and B data differently from G data, using a 2x1 simple filtering scheme to compute sharpness and luminance, reducing the number of bits required and eliminating sharpness filtering for G data, while combining inverse-gamma-converted R and B data with input G data according to a sub-pixel structure, thereby simplifying computations and reducing power consumption.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If a diamond filter is used to determine gray scale values of sub-pixels using five sub-pixel values, then display resolution can be reduced, but computational complexity increases and power consumption reduction is small

Engineering Contradiction:
Improvenumber of pixelsVSAvoidcomputational process complexity
Core Design Contradiction:
Quantity of substanceVSDevice complexity

Solution Approach 1:

The patent segments the image processing into distinct stages: edge detection phase followed by sharpening phase. This segmentation allows different processing strategies to be applied at different stages, reducing overall computational complexity while maintaining display quality with fewer pixels.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs edge detection as a preliminary action before sharpening. By identifying edges first, the subsequent sharpening process can be targeted and optimized, reducing the computational burden compared to applying complex filters uniformly across the entire image.

Inventive Principle:
Principle #10Preliminary action

2Manufacturing precision

If diamond filter and sharpness processing using G data are applied, then image processing is enhanced, but color errors occur and contour blurring occurs

Engineering Contradiction:
Improveimage processing qualityVSAvoidcolor error and contour blurring
Core Design Contradiction:
Manufacturing precisionVSObject-generated harmful factors

Solution Approach 1:

The patent separates edge detection from sharpening operations, allowing each to be optimized independently. Edge detection identifies boundaries without introducing color errors, while subsequent sharpening enhances contours without the blurring effects of conventional diamond filters applied to all data.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies different processing treatments to different regions of the image based on edge detection results. Edge regions receive targeted sharpening while non-edge regions are handled differently, preventing color errors in uniform areas and avoiding contour blurring in edge areas.

Inventive Principle:
Principle #3Local quality

3Ease of operation

If three line memories are provided to determine data values of pixels in particular rows, then image processing is enabled, but product unit cost increases

Engineering Contradiction:
Improveimage processing capabilityVSAvoidproduct unit cost
Core Design Contradiction:
Ease of operationVSEase of manufacture

Solution Approach 1:

The patent extracts only the essential data needed for edge detection and sharpening operations, processing pixels in a optimized sequence that requires fewer line memories. By taking out only the necessary previous row data for computation, the system reduces memory requirements from three line memories to a smaller number.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent implements dynamic data processing where the required memory buffer is adjusted based on the current processing stage and pixel position. This dynamic approach allows the system to function with fewer line memories by efficiently managing data flow and reuse, reducing product cost while maintaining processing capability.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP2388769B1Image processing method and display device using the same
Publication Date: 2016.01.06 LG DISPLAY CO LTD
  • EP2388769B1 patent drawingFigure 1~2
  • EP2388769B1 patent drawingFigure 3~4
  • EP2388769B1 patent drawingFigure 5

AI summary

Disclosed is an image processing method, in which three primary color data of an input RGB data format are rendered on a display panel according to a sub-pixel structure of the display panel, the display panel having as many G sub-pixels as the display resolution input G data and as many R and B sub-pixels as half the display resolution of input R and B data, respectively, the method comprising: (A) separating the R and B data and the G data from the input data; (B) loading data corresponding to respective odd rows of gamma-converted R and B data, and storing data corresponding to respective even rows of the R and B data adjacent to the loaded odd rows; (C) loading two R data of the even row, along with two R data of the odd row corresponding to a first display position, so as to form a 2x2 R pixel area, and loading two B data of the even row, along with two B data of the odd row corresponding to a second display position, so as to form a 2x2 B pixel area; (D) computing the sharpness of the corresponding display data by comparing the data in each of the R and B pixel areas column by column and row by row; (E) computing the luminance of the display data by taking the average value of the data corresponding to the odd row of each of the R and B pixel areas; (F) determining the gray scale value of output R data by adding the sharpness to the luminance of the R data, and determining the gray scale value of output B data by adding the sharpness to the luminance of the B data; and (G) combining the inverse-gamma-converted R and B data and the input G data and then outputting the combined data according to the sub-pixel structure of the display panel.