Image Feature Separation Using Principal Inertia Axis for Shadow Removal
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques struggle to accurately separate an object from its shadow in captured images, especially in dark environments or with backlighting, leading to inaccurate object identification.
Innovation Solution
An information processing device that detects a feature area using a detector and separates it along a principal inertia axis, utilizing a separator to divide the area into distinct objects based on the feature area's orientation and an evaluation of the separation's appropriateness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If shadow removal is performed based on reference luminance value, then shadow removal is achieved in normal lighting, but shadow cannot be removed accurately in dark environment or backlighting
Solution Approach 1:
The patent changes the separation parameter from luminance value to spatial distribution characteristics (principal inertia axis). By using the principal inertia axis which reflects the spatial orientation and shape of the feature area, the method adapts to different lighting conditions where luminance-based separation fails, achieving reliable shadow removal across varied environments.
Solution Approach 2:
The patent introduces the principal inertia axis as an intermediary parameter that bridges the gap between raw image data and accurate shadow separation. This intermediary represents the spatial orientation of the detected feature area and enables separation decisions that are independent of lighting conditions, resolving the contradiction between reliability in normal lighting and adaptability to dark/backlit environments.
2Measurement precision
If feature area is separated along principal inertia axis, then separation accuracy is improved, but device complexity increases due to additional calculations
Solution Approach 1:
The patent performs preliminary detection of the feature area and calculation of its principal inertia axis before the separation decision. By pre-computing the spatial orientation characteristics, the actual separation process becomes simpler and more accurate, as the principal inertia axis provides a ready-made reference for division without requiring complex real-time calculations during separation.
3Productivity
If feature area occupies large proportion of circumscribing rectangle, then separation may create unnecessary division, but without separation shadow identification accuracy decreases
Solution Approach 1:
The patent uses feedback from the principal inertia axis calculation to determine whether separation is necessary. By analyzing the orientation and shape characteristics provided by the principal inertia axis, the system can identify cases where separation would be meaningless (when feature area already occupies most of the circumscribing rectangle) and avoid unnecessary processing, while maintaining accurate shadow identification when separation is appropriate.
Data Source
AI summary
An information processing device includes a detector that detects a feature area with a predetermined feature from a captured image, and a separator that separates the feature area along a principal inertia axis of the feature area.