Dirty Region Image Processing for Display Power Reduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current screen refresh rates of 60Hz and 90Hz in electronic devices are insufficient for handling multiple applications or large applications, leading to inefficiencies in image processing and increased power consumption.

Innovation Solution

An image processing method that determines the dirty region of the display, compares image data using a similarity detection algorithm like dHash, and intercepts the Vsync signal when similarity between frames is below a threshold, reducing GPU and CPU load and conserving power.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If the screen refresh rate is increased to 60Hz or 90Hz to meet user requirements for screen response, then the screen response performance is improved, but the power consumption of the GPU and CPU increases

Engineering Contradiction:
Improvescreen response speedVSAvoidpower consumption
Core Design Contradiction:
SpeedVSUse of energy by moving object

Solution Approach 1:

The patent divides the screen into multiple regions and identifies only the 'dirty regions' (areas that have changed) for updating. Instead of refreshing the entire screen at high refresh rates, the system segments the update task to only process changed regions, thereby maintaining responsive performance while reducing the overall processing load and power consumption of the GPU and CPU.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies partial action by updating only the necessary dirty regions rather than the entire screen. By using similarity detection algorithms to identify and update only the changed portions, the system achieves the required screen response performance with reduced processing effort and lower power consumption compared to full-screen updates at high refresh rates.

Inventive Principle:
Principle #16Partial or excessive action

2Stability of the object's composition

If the entire screen is updated at high refresh rates to ensure smooth display, then the display smoothness is improved, but the processing load on GPU and CPU increases

Engineering Contradiction:
Improvedisplay smoothnessVSAvoidprocessing load
Core Design Contradiction:
Stability of the object's compositionVSDevice complexity

Solution Approach 1:

The patent segments the screen update process by identifying and processing only the dirty regions that have changed. This segmentation approach maintains display smoothness by ensuring that changed areas are updated promptly, while avoiding the unnecessary processing load of updating the entire screen, thereby reducing GPU and CPU complexity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system employs self-service mechanisms through automated dirty region detection and similarity comparison. The device automatically identifies which regions need updating and processes only those areas, eliminating the need for full-screen processing and reducing the overall processing load while maintaining smooth display performance.

Inventive Principle:
Principle #25Self-service

3Use of energy by moving object

If similarity detection algorithms are used to compare dirty regions and reduce updates, then power consumption is reduced, but the processing time for each frame increases

Engineering Contradiction:
Improvepower consumptionVSAvoidframe processing time
Core Design Contradiction:
Use of energy by moving objectVSLoss of time

Solution Approach 1:

The patent extracts only the dirty regions from the full screen for comparison and processing. By taking out and processing only the changed portions rather than the entire frame, the similarity detection algorithm reduces the amount of data to be processed, thereby minimizing the increase in processing time while still achieving power savings through selective updating.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies local quality by focusing processing resources only on the dirty regions where changes have occurred. The similarity detection is performed locally on these specific regions rather than globally across the entire screen, which reduces the overall processing time per frame while still enabling power consumption reduction through intelligent selective updating.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP3916709B1Image processing method and device, electronic device, and storage medium
Publication Date: 2023.10.18 BEIJING XIAOMI MOBILE SOFTWARE CO LTD
  • EP3916709B1 patent drawingFigure 1
  • EP3916709B1 patent drawingFigure 2
  • EP3916709B1 patent drawingFigure 3

AI summary

The disclosure relates to an image processing method and device, an electronic device, and a storage medium. The method includes: a dirty region of a display region is determined, and a percentage of the dirty region in the display region is calculated (S11); first image data of the dirty region in an image frame to be updated for displaying and second image data of the dirty region in a presently displayed image frame are acquired, and similarity detection is performed on the first image data and the second image data to generate a similarity detection result (S12); whether to update the image frame to be updated for displaying to the display region is determined according to the similarity detection result and the percentage of the dirty region in the display region, and if NO, an updating request for the image frame to be updated for displaying is shielded (SI3).