HDR LED Flicker Detection Using Exposure-Varied Pixel Groups
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing imaging systems struggle to differentiate between LED flicker artifacts and moving objects in High Dynamic Range (HDR) imagery, leading to image blurriness and ghosting, which compromises image quality and object identification.
Innovation Solution
A method using spatially multiplexed image sensors with varying exposure settings to capture multiple frames, combined with low pass filters and intensity pattern analysis, distinguishes between LED flicker and moving objects by analyzing pixel intensity patterns across frames.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If long-exposure frames are used to mitigate LED flicker, then LED flicker artifacts are reduced, but moving objects become blurry and ghosting occurs
Solution Approach 1:
The image frame is segmented into multiple regions: first region containing LED light sources (processed with long-exposure data to eliminate flicker), second region containing moving objects (processed with short-exposure data to maintain clarity), and third region containing static scenes (processed with fused data). This spatial segmentation allows different exposure settings to be applied selectively to different areas, resolving the contradiction between flicker mitigation and image clarity.
Solution Approach 2:
Different quality characteristics are applied to different regions of the image. The LED light source regions use long-exposure processing for flicker elimination, while moving object regions use short-exposure processing for sharpness preservation. This local differentiation of processing quality enables simultaneous optimization of both flicker mitigation and image clarity in their respective regions.
2Object-affected harmful factors
If long-exposure settings are applied to the entire frame, then LED flicker is compensated, but moving objects exhibit ghosting and blurriness
Solution Approach 1:
The frame is divided into distinct processing zones based on content type. Moving objects are identified and isolated in a second region where short-exposure data is used, ensuring reliable identification without ghosting. LED light sources are identified in a first region where long-exposure data eliminates flicker. This segmentation ensures reliable object identification while compensating for LED flicker in appropriate regions.
Solution Approach 2:
A third region acts as an intermediary or transition zone between the LED light source region and moving object region. This intermediate region handles areas with mixed characteristics or boundary conditions, allowing smooth transitions and ensuring that neither flicker nor ghosting artifacts dominate the final image, thereby maintaining overall reliability.
3Manufacturing precision
If short-exposure frames are used, then moving objects remain clear, but LED flicker artifacts persist
Solution Approach 1:
The image processing systematically segments the frame to apply appropriate exposure data to each region. While short-exposure data is used for moving objects to maintain clarity, long-exposure data is specifically applied to LED light source regions to eliminate flicker artifacts. This segmentation strategy resolves the contradiction by allowing short-exposure clarity where needed while applying long-exposure flicker mitigation where applicable.
Solution Approach 2:
The final image is generated by merging data from both short-exposure and long-exposure frames according to region-specific requirements. Moving object regions are populated from short-exposure data for clarity, while LED light source regions are populated from long-exposure data to eliminate flicker. This selective merging combines the advantages of both exposure types, achieving both clarity and flicker mitigation in their respective domains.
Data Source
AI summary
This application describes method and apparatus for detecting LEDs and moving objects using spatially multiplexed image sensors. The sensor may include various pixel groups, each containing multiple pixels set to different exposure levels. Correspondingly, each frame captured reflects these diverse exposure settings. An example method involves extracting a vector of pixel intensities from a specific cluster of pixels in each frame, resulting in a collection of intensity vectors, one from each frame. These vectors are used to create a pattern that shows how the intensity varies across the series of frames. This intensity-changing pattern may be used to determine whether the cluster of pixels is recording a moving object or the flicker of an LED light source.