CMOS Image Sensor Blackening Correction via Selective D-Phase Counting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
CMOS image sensors face a blackening phenomenon where pixels irradiated with high illumination light, such as sunlight, experience charge leakage, leading to inaccurate reset component counting and small output values, resulting in incorrect pixel signal representation, especially in optical black areas where uniform correction can distort the entire image color.
Innovation Solution
A solid-state imaging device with a pixel array, a pixel signal generation part, and a control part that includes a comparator and counter to calculate pixel signals by comparing voltages with a reference voltage, where the counter's D-phase count value is set as a limit value regardless of the comparator's output when the P-phase count value is a limit value, with an enable signal controlling the process to prevent blackening correction in optical black areas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the count value of the clock signal in the P-phase period is monitored to set flags and uniformly correct the count value in the D-phase period to a maximum value, then the blackening phenomenon is prevented, but the color of the entire image may be distorted when applied to optical black areas
Solution Approach 1:
The patent applies different correction strategies to different regions of the pixel array. For non-optical black areas, uniform correction is applied when blackening is detected. For optical black areas, the correction is selectively applied only when necessary, preserving the natural color balance of these reference regions while still preventing blackening artifacts in the imaging areas.
Solution Approach 2:
The pixel array is segmented into optical black areas and non-optical black areas, with different correction logic applied to each segment. This segmentation allows the system to prevent blackening in imaging regions while maintaining color accuracy in reference regions used for image processing.
2Measurement precision
If charges photoelectrically converted by photodiodes leak to floating diffusions under high illumination light, then the reset component counting becomes inaccurate, but uniformly correcting all pixels including optical black areas preserves simplicity at the cost of image color integrity
Solution Approach 1:
The patent implements region-specific correction where optical black areas and non-optical black areas are treated differently. In optical black areas, correction is applied only when blackening is detected, preserving the natural color reference. In non-optical black areas, uniform correction ensures accurate pixel signal representation without compromising overall image color fidelity.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This solution effectively corrects the blackening phenomenon by ensuring accurate pixel signal generation, maintaining natural and beautiful image quality without uniformly correcting count values in optical black areas, thus preserving the image color integrity.
Implementation Method 1
charges photoelectrically converted by photodiodes
Data Source
AI summary
Disclosed is a solid-state imaging device including a pixel array, a pixel signal generation part, and a control part. The pixel signal generation part includes a comparator and a counter. In a case where an enable signal is supplied from the control part, a count value of the counter in a D-phase period where a signal level is detected is set as a limit value regardless of an output of the comparator when a count value of the counter in a P-phase period where a reset level is detected is a limit value.