OLED Burn-In Prevention via AI Object Duration Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
OLED displays face burn-in issues due to varying luminance of pixels over time, leading to afterimages, especially when fixed images are continuously displayed, and existing methods either react after the afterimage occurs or require periodic display turn-ons/off, making them unsuitable for streaming images.
Innovation Solution
A display method that predicts the display duration of objects like logos, words, or text bars in images using deep neural networks and adjusts the luminance of corresponding pixels based on the predicted duration to prevent burn-in, allowing for continuous streaming image display without afterimages.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If luminance is maintained at high levels for displaying fixed images, then image quality is improved, but burn-in phenomenon occurs due to pixel degradation
Solution Approach 1:
The system performs preliminary action by predicting the display duration of objects before they cause burn-in. It calculates anticipated display times for logos, text bars, and other static elements, then proactively adjusts luminance values before degradation occurs. This prevents burn-in by reducing luminance of pixels that are expected to display the same content for extended periods, while maintaining high luminance for dynamic content that won't cause burn-in.
2Reliability
If image processing is performed to prevent afterimage, then burn-in is reduced, but processing time increases
Solution Approach 1:
The system segments the image processing task by identifying and separating static objects (logos, text bars, fixed elements) from dynamic content. It applies different processing strategies to each segment: predictive luminance adjustment for static objects based on their expected display duration, and normal display processing for dynamic content. This segmentation enables efficient processing by focusing computational resources only where burn-in prevention is needed.
3Reliability
If periodic display turn-on and off is implemented, then afterimage is reduced, but continuity of streaming display is disrupted
Solution Approach 1:
The system applies local quality by implementing selective luminance adjustment only for specific regions containing static objects that are prone to burn-in, rather than uniformly adjusting the entire display. It identifies objects like logos and text bars, predicts their display duration, and applies luminance reduction only to the pixels displaying these static elements. This localized approach maintains display continuity for streaming content while preventing burn-in in specific high-risk areas.
Data Source
AI summary
The present disclosure relates to a display device and a display method for reducing a burn-in phenomenon by performing an on-board artificial intelligence (AI) algorithm and/or machine learning algorithm to predict a display duration of an object in an image, and processing the image based on the predicted display duration. The display method according to one embodiment of the present disclosure may include extracting a predetermined type of object from an inputted image, predicting a display duration of the extracted object, and adjusting a luminance value of a pixel in an area corresponding to the object in the image based on the predicted display duration, and displaying the image in which the adjusted luminance value of the pixel is reflected.