Image Sensor with Nested Color, Absorption and Polarization Filters

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current RGB image sensors struggle to reliably detect hazards like puddles, water, and ice on road surfaces due to limitations in distinguishing between absorption and polarization effects, leading to unreliable localization and classification results.

Innovation Solution

An image sensor with a grid pattern of pixels, incorporating multiple filter elements including color filters, absorption filters with different optical bandwidths, and polarization filters with distinct characteristics, which provide additional information beyond traditional color data to enhance detection capabilities.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If RGB image sensors are used for road condition detection, then the device remains compact and simple, but the ability to distinguish between water, snow, and ice is poor

Engineering Contradiction:
Improvedetection reliabilityVSAvoidsensor complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent divides each image element into multiple sub-elements, with each sub-element containing different filter combinations (RGB filters, absorption filters, polarization filters). This segmentation allows the sensor to capture multiple types of optical information simultaneously without requiring separate sensor arrays, thereby improving detection reliability while maintaining reasonable device complexity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Each image element is designed to perform multiple functions by incorporating different filter types (color filters for RGB information, absorption filters for material identification, polarization filters for surface property detection). This multi-functionality allows a single sensor array to provide comprehensive road condition analysis, resolving the contradiction between detection capability and device complexity.

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

2Loss of information

If multiple filter elements are added to each image element, then absorption and polarization information is obtained, but the surface area and resolution are affected

Engineering Contradiction:
Improveinformation completenessVSAvoidsensor surface area
Core Design Contradiction:
Loss of informationVSArea of stationary object

Solution Approach 1:

The patent nests multiple filter elements within each image element, creating a hierarchical structure where absorption filters, polarization filters, and color filters are arranged in nested patterns. This nesting allows multiple filter types to occupy the same spatial footprint, capturing comprehensive optical information without proportionally increasing the overall sensor surface area.

Inventive Principle:
Principle #7Nested doll (Nesting)

Solution Approach 2:

The patent transitions from traditional 2D filter arrangements to a multi-dimensional filter configuration where filters are arranged in multiple rows and columns within each image element. This dimensional expansion allows more filter elements to be packed into the same area, maintaining resolution while increasing information completeness.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Loss of information

If multiple filter elements are added to each image element, then absorption and polarization information is obtained, but the manufacturing complexity increases

Engineering Contradiction:
Improveinformation completenessVSAvoidmanufacturing simplicity
Core Design Contradiction:
Loss of informationVSEase of manufacture

Solution Approach 1:

The patent applies different filter combinations to different regions of the sensor array, with each image element containing locally optimized filters based on the specific detection requirements. This local quality approach allows standardized manufacturing processes to be used while achieving customized detection capabilities, thereby reducing overall manufacturing complexity.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent uses parameter changes in filter arrangement and configuration to optimize performance. By varying the orientation, density, and type of filters in different image elements, the system achieves comprehensive information capture while maintaining compatibility with existing manufacturing processes, thus balancing information completeness with manufacturing simplicity.

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 enhanced image sensor enables improved detection of road hazards by differentiating between water, snow, and black ice, and identifying organic materials, offering a compact and efficient solution for improved road condition recognition.

Implementation Method 1

The color filter is designed to filter light according to color, thus enabling the image sensor to detect color information

Methodology Applied
Scientific EffectColor filtering: Filter (optical)

Implementation Method 2

Absorption information makes it easier to find specific materials with certain absorption properties in a sensor image

Methodology Applied
Scientific EffectAbsorption: Absorption (EM radiation)

Implementation Method 3

sensor images with polarization filters make it possible to analyze the light rays reflected from a surface

Methodology Applied
Scientific EffectPolarization: Polarisation

Data Source

PatentEP4184932A1An image sensor, a method for operating an image sensor, a method for producing an image sensor and a stationary device or a drone with an image sensor
Publication Date: 2023.05.24 FRAUNHOFER GESELLSCHAFT ZUR FORDERUNG DER ANGEWANDTEN FORSCHUNG EV
  • EP4184932A1 patent drawingFigure 1
  • EP4184932A1 patent drawingFigure 2
  • EP4184932A1 patent drawingFigure 3

AI summary

Exemplary embodiments of the invention provide an image sensor (100, 410, 650, 720) with an image element structure (100). The image element structure (100) comprises a plurality of image elements (120) arranged in a grid pattern in a first direction (113, 213) and in a second direction 116, which is orthogonal to the first direction (113, 213). Each image element (120) of the plurality of image elements (120) comprises a plurality of spatially adjacent filter elements (130, 133, 136, 242, 244, 246, 253, 256, 263, 266). The majority of filter elements (130, 133, 136, 242, 244, 246, 253, 256, 263, 266) have at least one color filter (133, 242, 244, 246) and at least one additional filter (136, 253, 256, 263, 266) from a filter group.The filter group comprises a first absorption filter (136, 253, 256) with a first optical bandwidth, a second absorption filter (136, 253, 256) with a second optical bandwidth that differs from the first optical bandwidth, a first polarization filter (136, 263, 266) with a first polarization characteristic, a second polarization filter (136, 263, 266) with a second polarization characteristic that differs from the first polarization characteristic, and a filter element (136) without absorption or polarization effect.