Vision Sensor Timestamp Regeneration for Bad-Pixel Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Image sensors, particularly vision sensors, suffer from issues with bad pixels that affect yield and image quality, and existing methods are limited in improving this through manufacturing processes alone.
Innovation Solution
An image processing device that regenerates timestamps for events occurring at bad pixels using a vision sensor and a processor, classifying events into groups based on timestamp values and determining abnormal events to replace them with adjacent pixel timestamps.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If manufacturing process improvement is used to address bad pixels, then pixel quality improves, but manufacturing complexity and cost increase
Solution Approach 1:
The patent introduces an intermediary processing system that detects bad pixels and regenerates their timestamps using data from adjacent pixels. This mediator layer handles the correction of defective pixels without requiring complex manufacturing process changes, thus resolving the contradiction between improving pixel quality and avoiding manufacturing complexity.
Solution Approach 2:
The patent changes the temporal parameter (timestamp) of bad pixels by regenerating it from adjacent pixels' timestamp data. Instead of trying to physically repair the bad pixels through complex manufacturing processes, the system modifies the timestamp parameter to make bad pixels appear normal, thereby improving effective pixel quality without increasing manufacturing complexity.
2Productivity
If all pixels including bad pixels are processed normally, then data processing volume increases, but image quality deteriorates
Solution Approach 1:
The patent extracts and identifies bad pixels from the overall pixel array, then processes them differently through timestamp regeneration. By separating bad pixels from normal pixels and applying a specific correction method, the system maintains high image quality while avoiding unnecessary processing of defective data, thus resolving the contradiction between processing volume and image quality.
Data Source
AI summary
An image processing device includes a vision sensor and a processor. The vision sensor generates a plurality of events in which an intensity of light changes and generates a plurality of timestamps depending on times when the events occur. In addition, the processor may regenerate a timestamp of a pixel where an abnormal event occurs, based on temporal correlation of the events.