Ambiguity Detection in 2D Structure Matrices

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for detecting ambiguities in 2D structures, such as images, are inefficient and power-intensive, particularly in real-time systems like driver assistance systems, as they often rely on complex transformations like Fourier transforms and require significant computational resources.

Innovation Solution

A method that compares a matrix of a 2D structure with a comparison matrix using a similarity algorithm, forming a similarity matrix and evaluating it to generate an ambiguity measure, which can be implemented in hardware with low power consumption, enabling efficient processing of large numbers of images in real-time.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If complex transformation methods like Fourier transforms are used for ambiguity detection, then measurement precision is improved, but use of energy increases and productivity decreases

Engineering Contradiction:
Improveambiguity detection precisionVSAvoidpower consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent extracts only the essential information needed for ambiguity detection from the full image data by comparing selected regions (first image region and second image region) rather than processing entire images through complex transforms. This selective extraction maintains detection precision while significantly reducing computational energy requirements.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent segments the image processing task into specific comparable regions rather than processing the entire image. By dividing the task into comparing a first image region with a second image region, the method reduces the data volume requiring complex transformation while preserving the ability to detect ambiguities in the relevant areas.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If complex transformation methods like Fourier transforms are used for ambiguity detection, then measurement precision is improved, but productivity decreases

Engineering Contradiction:
Improveambiguity detection precisionVSAvoidimages processed per second
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent extracts only the essential information needed for ambiguity detection from the full image data by comparing selected regions (first image region and second image region) rather than processing entire images through complex transforms. This selective extraction maintains detection precision while significantly reducing computational energy requirements.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent segments the image processing task into specific comparable regions rather than processing the entire image. By dividing the task into comparing a first image region with a second image region, the method reduces the data volume requiring complex transformation while preserving the ability to detect ambiguities in the relevant areas.

Inventive Principle:
Principle #1Segmentation

3Productivity

If simple comparison algorithms are used, then use of energy decreases and productivity increases, but measurement precision worsens

Engineering Contradiction:
Improveprocessing speedVSAvoidambiguity detection accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent introduces an intermediary evaluation step that assesses the comparison results between the first image region and second image region. This intermediary evaluation function analyzes the similarity or differences between regions to determine ambiguity, bridging the gap between simple comparison operations and accurate ambiguity detection without requiring complex transformations.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP3327629B1Method and apparatus for detecting ambiguities in a matrix of a 2d structure
Publication Date: 2023.08.16 ROBERT BOSCH GMBH
  • EP3327629B1 patent drawingFigure 1~2
  • EP3327629B1 patent drawingFigure 3a~3b
  • EP3327629B1 patent drawingFigure 3c

AI summary

The present invention relates to a method for detecting ambiguities in a structure, comprising several steps, and a device, for example a chip, such as in a driver assistance system, for detecting ambiguities in a structure. In a first step (S100) of the method, a matrix of a 2D structure is provided. The matrix has several matrix elements. In a further step (S200), a comparison matrix with several comparison matrix elements is provided. A step (S300) comprises comparing the matrix with the comparison matrix using a comparison algorithm. In a subsequent step (S400), a similarity matrix with similarity elements is formed from the result of the comparison. A step (S500) comprises evaluating the similarity matrix using an evaluation function along a path that includes a plurality of similarity elements.One step (S600) involves generating an evaluation result as a measure of any ambiguity present in the matrix of the first 2D structure.