CMOS Image Sensor Black Level Correction via Unidirectional Pixel Arrangement

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional CMOS image sensors face issues with fixed pattern noise and color correction errors due to variations in pixel characteristics, leading to suboptimal image quality and errors between color corrections at the same pixel position.

Innovation Solution

The implementation of a unidirectional arrangement of photoelectric conversion elements for each color, with a black correction unit that subtracts independent output results from the same pixel's photoelectric conversion elements to correct the black level, ensuring accurate color correction without inter-color crosstalk.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If conventional area sensors with spectral filters are used, then multiple colors can be captured at different positions, but correction errors occur between colors at the same pixel position due to variation of black correction circuits

Engineering Contradiction:
Improvecolor data capture capabilityVSAvoidblack level correction accuracy
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The sensor is divided into multiple pixel groups, where each group contains photoelectric conversion elements for different colors (R, G, B) that are spatially adjacent. This segmentation allows each color to have its own dedicated black correction circuit, eliminating cross-color correction errors while maintaining the ability to capture multiple colors at the same pixel position through the Bayer arrangement

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Each pixel group is assigned dedicated black correction circuits specific to that local region, rather than using a single shared correction circuit for all colors. This local quality approach ensures that each color channel receives customized correction tailored to its specific characteristics, preventing correction errors that would occur with shared circuits

Inventive Principle:
Principle #3Local quality

2Manufacturing precision

If fixed pattern noise correction is applied, then image quality improves, but correction errors occur between colors at the same pixel position

Engineering Contradiction:
Improveimage qualityVSAvoidcolor correction consistency
Core Design Contradiction:
Manufacturing precisionVSMeasurement precision

Solution Approach 1:

The correction system is segmented by color channel, with separate black correction circuits for R, G, and B pixels within each pixel group. This allows independent correction of each color's fixed pattern noise without introducing errors from other color channels, while the Bayer arrangement enables synthesis of corrected multi-color data at the same pixel position

Inventive Principle:
Principle #1Segmentation

3Measurement precision

If unidirectional arrangement of photoelectric conversion elements is used, then color correction errors are prevented, but device complexity increases

Engineering Contradiction:
Improveblack level correction accuracyVSAvoidphotoelectric conversion element arrangement
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

Within each pixel group, photoelectric conversion elements for different colors are arranged unidirectionally (e.g., horizontally adjacent), creating a localized structured pattern. This unidirectional arrangement within groups simplifies the correction circuit design for each group while maintaining the overall Bayer arrangement capability for multi-color capture

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The unidirectional arrangement pattern used within each pixel group can be applied universally across all color channels and all pixel groups in the sensor array. This universal application of the same arrangement principle reduces overall device complexity compared to requiring different complex arrangements for different colors, while still achieving accurate color-specific correction

Inventive Principle:
Principle #6Universality (Multi-functionality)

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 approach effectively prevents color correction errors between colors at the same pixel position, maintaining image quality and reducing the impact of temperature changes over time, while enhancing the productivity of continuous scanning.

Implementation Method 1

a plurality of photoelectric conversion elements that are arranged unidirectionally for each color of light to be received, and that perform photoelectric conversion of reflected light from an approximately same position of an object to be read for the each color sequentially pixel by pixel

Methodology Applied
Scientific EffectPhotoelectric conversion: Photoelectric Effect

Data Source

PatentUS9118873B2Imaging device, image reading apparatus, image forming apparatus, and image reading method
Publication Date: 2015.08.25 RICOH CO LTD
  • US9118873B2 patent drawing
  • US9118873B2 patent drawing
  • US9118873B2 patent drawing

AI summary

An imaging device includes photoelectric conversion elements arranged unidirectionally for each color of light to be received, and performing photoelectric conversion of reflected light from an approximately same position of an object for the each color sequentially pixel by pixel; and a black correction unit that corrects a black level of the object with respect to a pixel group including some photoelectric conversion elements in a way that an output result of the photoelectric conversion of reflected light by a photoelectric conversion element for a pixel, independent of the reflected light, is subtracted from another output result of photoelectric conversion of reflected light by the photoelectric conversion element for the same pixel.