Dynamic Vision Sensor Optical Path Actuation for Static-Scene Calibration

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Dynamic vision sensors struggle to calibrate effectively in static scenes due to the detection of noise events and insufficient events, leading to poor image capture.

Innovation Solution

An apparatus and method for a dynamic vision sensor that includes an actuator to change the optical path of light relative to the pixel array, causing the array to detect events, and a control circuit to adjust the sensor based on these events, enhancing image capture in static scenes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the dynamic vision sensor operates in a static scene, then the sensor can continuously monitor the environment, but the sensor detects mainly noise events and insufficient events for calibration

Engineering Contradiction:
Improvecalibration effectivenessVSAvoidnumber of detectable events
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The actuator performs preliminary action by actively changing the optical path before calibration is needed. The control circuit activates the actuator to shift the optical path, creating artificial light changes that guarantee sufficient events are detected for calibration, even when the scene itself is static.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system introduces dynamics into a static scene by using the actuator to dynamically change the optical path. This creates artificial motion and light changes that enable the DVS to detect sufficient events for calibration, transforming a static calibration problem into a dynamic solution.

Inventive Principle:
Principle #15Dynamics

2Measurement precision

If the optical path is changed to create calibration events, then sufficient events are generated for calibration, but the device complexity increases due to the actuator

Engineering Contradiction:
Improvecalibration accuracyVSAvoidoptical path adjustment mechanism
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system uses the DVS's own pixel array to detect the calibration events created by the actuator. The control circuit processes events from the same sensor that is being calibrated, eliminating the need for separate calibration sensors or external reference devices.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The actuator serves multiple functions: it creates calibration events for the DVS, and can potentially serve other optical adjustment functions. The control circuit performs multiple tasks including detecting calibration events, distinguishing them from noise events, and processing calibration data.

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

3Object-generated harmful factors

If the sensor sensitivity is adjusted during calibration, then noise events can be reduced, but the focus and optical aberration correction must also be adjusted

Engineering Contradiction:
Improvenoise eventsVSAvoidmultiple adjustment parameters
Core Design Contradiction:
Object-generated harmful factorsVSDevice complexity

Solution Approach 1:

The control circuit performs preliminary separation of noise events from meaningful events by analyzing event characteristics before full calibration processing. This preliminary filtering reduces the impact of noise events on calibration accuracy while the system processes multiple adjustment parameters in an organized sequence.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system systematically changes multiple parameters including sensitivity, focus, and optical aberration correction settings during calibration. The control circuit adjusts these parameters based on the detected events and calibration results, optimizing each parameter to improve overall sensor performance.

Inventive Principle:
Principle #35Parameter changes

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

The solution allows for effective calibration and improved image quality by distinguishing noise events from meaningful events, adjusting sensitivity, focus, and correcting optical aberrations, thereby enhancing the sensor's performance in static environments.

Implementation Method 1

an actuator configured to change an optical path of the light relative to the pixel array of the DVS

Methodology Applied
Scientific EffectOptical path change:

Implementation Method 2

a dynamic vision sensor configured to detect changes of light impinging on a pixel array of the DVS

Methodology Applied
Scientific EffectPhotoelectric effect: Photoelectric Effect

Data Source

PatentUS12445737B2Apparatus for a dynamic vision sensor and method for adjusting a dynamic vision sensor
Publication Date: 2025.10.14 SONY SEMICON SOLUTIONS CORP
  • US12445737B2 patent drawing
  • US12445737B2 patent drawing
  • US12445737B2 patent drawing

AI summary

Embodiments of the present disclosure provide an apparatus for a dynamic vision sensor and a method for adjusting a dynamic vision sensor. The apparatus comprises an actuator configured to change an optical path of light impinging on a pixel array of the DVS relative to the pixel array to cause the pixel array to detect one or more events resulting from the actuator changing the optical path of the light relative to the pixel array. Also, the apparatus comprises a control circuit configured to adjust the DVS based on the one or more events.