Image Sensor Fixed-Pattern Noise Correction Using Light Shielded Pixels
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image capturing apparatuses face challenges in accurately correcting fixed-pattern noise when switching between different reading methods, particularly due to variations in manufacturing processes and temperature changes, which affect the S/N ratio of correction values.
Innovation Solution
An image capturing apparatus with a processing unit and correction unit that can handle and correct image signals using different reading methods, employing a light shielded pixel area for noise detection and correction across multiple frames to generate accurate correction values for effective pixel areas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If correction values are calculated from the beginning when fixed-pattern shapes change, then correction accuracy is maintained, but processing time increases and productivity decreases
Solution Approach 1:
The system performs preliminary detection of fixed-pattern noise shapes and pre-calculates correction values before actual image capture. When the reading method changes, the system has already prepared correction values for multiple potential reading methods in advance, allowing immediate application without recalculation delay
Solution Approach 2:
The correction system dynamically adapts by storing multiple sets of correction values corresponding to different reading methods (combined and non-combined). The system selectively applies the appropriate correction set based on the current reading method, enabling rapid switching without recalculation while maintaining accuracy
2Productivity
If detection is performed in advance to generate correction values, then processing speed improves, but correction accuracy deteriorates when fixed-pattern noise shapes change due to temperature change or the like
Solution Approach 1:
The system periodically detects fixed-pattern noise shapes and updates correction values at scheduled intervals or when triggered by temperature changes. This periodic refresh ensures correction values remain accurate while maintaining overall processing efficiency through batch processing
Solution Approach 2:
The system incorporates feedback mechanisms that monitor temperature changes and reading method switches. When temperature varies beyond a threshold or reading method changes, the system triggers recalculation of correction values, ensuring accuracy is maintained under changing conditions while avoiding unnecessary recalculations
3Measurement precision
If the number of frames is increased to improve correction pattern accuracy, then correction precision improves, but reading out speed from the image sensor decreases
Solution Approach 1:
The system uses a small number of frames (even a single frame) for fixed-pattern noise detection by leveraging light shielded pixel areas. This partial action approach extracts sufficient correction information without requiring multiple frames, thereby maintaining high reading out speed while achieving adequate correction accuracy
Data Source
AI summary
An image capturing apparatus includes an image sensor having a plurality of pixels and being capable of outputting a plurality of image signals read out using different methods for individually reading out the signals of the plurality of pixels and correction signals for correcting the plurality of image signals, a processing unit for processing the correction signals so as to handle the reading methods for the plurality of image signals, and a correction unit for correcting the plurality of image signals using the correction signals processed by the processing unit.