Image Sensor Readout Alignment for Single-Conversion Depth Sensing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing image sensors face challenges in efficiently converting multiple pixel signals into depth information signals, leading to prolonged readout times and reduced frame rates.

Innovation Solution

An image sensor design that allows simultaneous reading of pixel signals from multiple rows, with aligned signals undergoing a single analog-to-digital conversion operation, using an alignment circuit and signal conversion circuits to generate depth information signals based on voltage level differences.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If multiple pixel signals are converted into depth information signals through sequential A/D conversion operations, then each pixel signal can be accurately converted, but the readout time is prolonged and frame rate is reduced

Engineering Contradiction:
Improvedepth information signal accuracyVSAvoidreadout time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent merges multiple A/D conversion operations into a single simultaneous conversion operation. Multiple pixel signals from different rows are aligned and fed into a single A/D converter at the same time, converting them into depth information signals concurrently. This combining approach maintains conversion accuracy while dramatically reducing the total readout time by eliminating sequential processing delays.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent implements preliminary signal alignment before the A/D conversion operation. Pixel signals from multiple rows are pre-aligned in the alignment circuit to ensure they are properly synchronized and ready for simultaneous conversion. This preliminary preparation enables the subsequent single A/D conversion operation to process multiple signals efficiently without compromising accuracy.

Inventive Principle:
Principle #10Preliminary action

2Productivity

If multiple A/D conversion circuits are used to convert multiple pixel signals simultaneously, then readout time is reduced and frame rate is improved, but device complexity and manufacturing cost increase

Engineering Contradiction:
Improveframe rateVSAvoidnumber of A/D conversion circuits
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent makes a single A/D conversion circuit universal by using it to convert multiple pixel signals from different rows simultaneously. The alignment circuit prepares signals from multiple rows in a standardized format that the single A/D converter can process concurrently. This multi-functionality approach achieves high frame rate without requiring multiple A/D conversion circuits, thereby reducing device complexity and manufacturing cost.

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

Solution Approach 2:

The alignment circuit serves as an intermediary between the pixel array and the single A/D conversion circuit. It receives pixel signals from multiple rows, aligns them properly, and presents them to the A/D converter in a format suitable for simultaneous conversion. This intermediary structure enables a single A/D converter to handle multiple signals efficiently, avoiding the need for multiple converters.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Productivity

If pixel signals from multiple rows are read out simultaneously, then readout efficiency is improved and frame rate is increased, but signal alignment and processing complexity increases

Engineering Contradiction:
Improvereadout efficiencyVSAvoidsignal alignment circuit
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent segments the pixel array into multiple rows, with each row having dedicated switches that can independently select and output pixel signals. This segmentation allows signals from different rows to be read out simultaneously through separate column lines, improving readout efficiency while maintaining organized signal paths that facilitate alignment in the subsequent circuit.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS12382199B2Image sensor and operating method thereof
Publication Date: 2025.08.05 SK HYNIX INC
  • US12382199B2 patent drawing
  • US12382199B2 patent drawing
  • US12382199B2 patent drawing

AI summary

Disclosed is an image sensor including first and second pixels arranged in a first row, for outputting first and second pixel signals through first and third column lines during a unit row time; third and fourth pixels arranged in a second row, for outputting third and fourth pixel signals through second and fourth column lines during the unit row time; an alignment circuit for aligning the first to fourth pixel signals for each row, and outputting first to fourth alignment signals, according to control signals; a first signal conversion circuit for generating a first depth information signal corresponding to a difference in voltage levels between the first and second alignment signals, through one A/D conversion operation; and a second signal conversion circuit for generating a second depth information signal, corresponding to a difference in voltage levels between the third and fourth alignment signals, through one A/D conversion operation.