Multi-resolution Image Segregation for Shadow Edge Discrimination

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional image processing techniques struggle to accurately distinguish between shadows and material object edges, leading to significant false positives and false negatives due to the penumbra effect, where shadows can form sharp boundaries and material edges can be soft.

Innovation Solution

The method employs spatio-spectral information derived from multi-resolution representations, such as a scale-spaced pyramid, to segregate illumination and material aspects of an image, using spatio-spectral operators and constraints to identify intrinsic images, including material and illumination components.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If conventional brightness boundary analysis is used to detect object edges, then the algorithm can be accurately executed by a computer, but the results will often be incorrect due to false positives and false negatives in shadow recognition

Engineering Contradiction:
Improvealgorithm execution accuracyVSAvoidshadow and object edge recognition accuracy
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The patent segments the image processing task by creating multiple selectively varied representations of the image (e.g., different resolutions, color spaces, or feature representations) and performing image segregation operations on each representation. This allows the system to analyze brightness boundaries across multiple views and contexts, improving the reliability of shadow versus object edge discrimination while maintaining computational accuracy.

Inventive Principle:
Principle #1Segmentation

2Reliability

If multi-resolution representations are used to accurately identify illumination and material characteristics, then false positives and negatives are reduced, but the device complexity increases

Engineering Contradiction:
Improveillumination and material identification accuracyVSAvoidmulti-resolution representation complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent applies partial action by selectively performing image segregation operations on preselected representations from the set of selectively varied representations. Rather than processing all possible representations equally, the system identifies and processes only those representations most relevant to detecting illumination and material characteristics, thereby reducing computational complexity while maintaining high reliability in the results.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP2435956B1Multi-resolution analysis in an image segregation
Publication Date: 2019.03.06 TANDENT VISION SCIENCE INC
  • EP2435956B1 patent drawingFigure 1
  • EP2435956B1 patent drawingFigure 2A
  • EP2435956B1 patent drawingFigure 2B~2C

AI summary

In a first exemplary embodiment of the present invention, an automated, computerized method is provided for processing an image. According to a feature of the present invention, the method comprises the steps of providing an image file depicting an image, in a computer memory, forming a set of selectively varied representations of the image file and performing an image segregation operation on at least one preselected representation of the image of the image file, to generate intrinsic images corresponding to the image. According to a feature of the exemplary embodiment of the present invention, the selectively varied representations comprise multi-resolution representations such as a scale-spaced pyramid of representations. In a further feature of the exemplary embodiment of the present invention, the intrinsic images comprise a material image and an illumination image.