Vision Sensor Spatial Sensing via Periodic Color Filtering

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional vision sensors struggle to effectively sense spatial information of both moving and stationary objects, as they primarily rely on temporal changes in light intensity, failing to capture shape and color details of stationary objects.

Innovation Solution

The apparatus and method employ a shutter and actuator to adjust light transmittance, combine color filters for wavelength-specific imaging, and utilize multiple vision sensors to calculate disparity, extracting 3D spatial information and compensating for movement-induced optical flow, enabling the recognition of spatial information across various object states.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If a conventional vision sensor relies on temporal changes in light intensity, then it can detect moving objects, but it fails to capture shape and color details of stationary objects

Engineering Contradiction:
Improvecapability to detect both moving and stationary objectsVSAvoidshape and color details of stationary objects
Core Design Contradiction:
Adaptability or versatilityVSLoss of information

Solution Approach 1:

The patent applies periodic action by sequentially switching between multiple color filters (e.g., red, green, blue) in front of the vision sensor. This periodic filtering allows the sensor to capture spatial information including shape and color details of stationary objects by taking multiple measurements at different wavelengths, while still maintaining the ability to detect temporal changes for moving objects.

Inventive Principle:
Principle #19Periodic action

Solution Approach 2:

The patent segments the spectral information by using multiple color filters to divide the full spectrum into separate wavelength bands. Each filter captures specific spatial information at its particular wavelength, and these segmented measurements are then combined to reconstruct complete spatial information including color details that would be lost with a single temporal-change-based measurement.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If multiple vision sensors are used to calculate disparity for 3D spatial information, then depth perception improves, but device complexity increases

Engineering Contradiction:
Improve3D spatial information accuracyVSAvoidnumber of vision sensors and changers
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent makes each vision sensor and changer combination multi-functional by enabling it to capture both temporal change information (for motion detection) and spatial information (for shape and color) through sequential color filtering. This universality allows the system to achieve 3D spatial information using fewer sensors than would traditionally be required, as each sensor performs multiple measurement functions.

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

Solution Approach 2:

The patent introduces color filters as intermediary elements between the light source and the vision sensor. These filters act as mediators that enable the extraction of spatial information including color details by sequentially blocking different wavelength bands, thereby allowing a single vision sensor to gather information that would otherwise require multiple specialized sensors.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Speed

If the vision sensor moves during imaging, then it can track moving objects, but the captured spatial information becomes distorted due to optical flow

Engineering Contradiction:
Improvetracking speed of moving objectsVSAvoidspatial information accuracy
Core Design Contradiction:
SpeedVSMeasurement precision

Solution Approach 1:

The patent applies feedback by using the detected temporal changes and optical flow information to adjust and compensate for sensor movement. The system uses the measured optical flow from sequential color filter measurements to correct spatial information, thereby maintaining measurement precision even when the sensor is moving to track objects.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent applies preliminary anti-action by measuring optical flow using sequential color filtering before final spatial information extraction. This preliminary measurement of movement-induced optical flow allows the system to pre-compensate for distortion effects, counteracting the harmful impact of sensor movement before it degrades the final spatial information accuracy.

Inventive Principle:
Principle #9Preliminary anti-action

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 allows for the accurate sensing of spatial information, including position, distance, shape, size, and color, of both moving and stationary objects, enhancing the capability to reconstruct color images and compensate for sensor movement, thereby improving the overall spatial information extraction process.

Implementation Method 1

the vision sensor includes a photoelectric conversion device in the form of an integrated circuit

Methodology Applied
Scientific EffectPhotoelectric conversion: Photoelectric Effect

Implementation Method 2

at least one color filter configured to change transmittance according to a wavelength band of the light being input to the vision sensor

Methodology Applied
Scientific EffectOptical filtering: Filter (optical)

Data Source

PatentEP2800353B1Method and apparatus for sensing spatial information based on vision sensor
Publication Date: 2021.06.09 SAMSUNG ELECTRONICS CO LTD
  • EP2800353B1 patent drawingFigure 1
  • EP2800353B1 patent drawingFigure 2
  • EP2800353B1 patent drawingFigure 3

AI summary

Apparatuses and methods for sensing spatial information based on a vision sensor are disclosed. The apparatus and method recognize the spatial information of an object sensed by the vision sensor that senses a temporal change of light. The light being input into the vision sensor is artificially changed using a change unit configured to change the light being input to the vision sensor.