Imaging Sensor Signal Combination for Fast Focus Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing imaging apparatuses face challenges in increasing the speed of reading out signals from multiple photoelectric conversion units while maintaining data reduction and focus detection accuracy, particularly in measuring phase differences.

Innovation Solution

The imaging apparatus employs a configuration where signals from two photoelectric conversion units per pixel are combined and read out in a way that reduces the number of data points to be scanned, using an adding circuit to generate combined signals (A+A and A+B) specifically for focus detection, allowing for faster horizontal scanning and enhanced signal-to-noise ratio.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If signals from all photoelectric conversion units are read out individually, then measurement precision is maintained, but read-out speed decreases and data volume increases

Engineering Contradiction:
Improvefocus detection accuracyVSAvoidread-out speed
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The image sensor is divided into a focus detection region and other regions. Only photoelectric conversion units within the focus detection region are used for phase difference measurement, while units in other regions are excluded from readout. This segmentation allows selective processing that maintains measurement precision while reducing overall data volume and readout time.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent extracts and processes only the necessary signals from photoelectric conversion units located within the focus detection region. By taking out only the relevant data needed for focus detection rather than reading all signals, the system achieves faster readout speeds while preserving the precision required for accurate focus measurement.

Inventive Principle:
Principle #2Taking out (Extraction)

2Productivity

If signals from multiple photoelectric conversion units are combined, then read-out speed increases and data volume reduces, but signal-to-noise ratio may deteriorate

Engineering Contradiction:
Improveread-out speedVSAvoidsignal-to-noise ratio
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent segments the image sensor into a focus detection region and other regions, combining signals only from units within the focus detection region. This selective combination maintains signal-to-noise ratio by excluding irrelevant data while still achieving data reduction and faster readout speeds compared to processing all signals.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Different processing quality is applied to different regions: signals from the focus detection region are combined for fast readout, while signals from other regions are excluded. This local quality approach ensures that combination operations are performed only where necessary for focus detection, preserving signal integrity and noise characteristics in the critical measurement region.

Inventive Principle:
Principle #3Local quality

3Measurement precision

If all pixels output both individual and combined signals, then measurement precision is maintained, but device complexity and data processing load increase

Engineering Contradiction:
Improvephase difference measurement accuracyVSAvoidsignal processing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the image sensor into a focus detection region and other regions. Only pixels within the focus detection region output both individual signals (for phase difference measurement) and combined signals (for focus detection). This segmentation reduces device complexity by limiting the number of pixels that perform dual output operations, while maintaining measurement precision in the critical focus detection area.

Inventive Principle:
Principle #1Segmentation

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 enables faster read-out of signals with improved signal-to-noise ratio and focus detection accuracy by reducing the amount of data to be read and combining signals only in the focus detection region, thereby increasing operational speed.

Implementation Method 1

a plurality of pixels, arranged in rows and columns, each including a plurality of photoelectric conversion units generating an electric charge based on an incident light

Methodology Applied
Scientific EffectPhotoelectric effect: Photoelectric Effect

Data Source

PatentEP2843939B1Imaging apparatus, imaging system, and method for driving imaging apparatus
Publication Date: 2017.05.10 CANON KK
  • EP2843939B1 patent drawingFigure 1
  • EP2843939B1 patent drawingFigure 2
  • EP2843939B1 patent drawingFigure 3

AI summary

The imaging apparatus has a plurality of pixels each of which has a plurality of photoelectric conversion units; generates a plurality of first combined signals obtained by combining signals based on electric charges of photoelectric conversion units in one side with each other, and a plurality of second signals obtained by combining signals based on electric charges of the plurality of photoelectric conversion units with each other; and outputs a part of the first combined signals out of the plurality of first combined signals.