Image Processing for 3D Video Alignment and Target Object Separation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
3D display technologies face issues with 3D errors due to misalignment of stereoscopic objects in image pairs, leading to reduced video quality and increased visual fatigue, as existing methods fail to differentiate and correct stereoscopic objects from target objects effectively.
Innovation Solution
An image processing method that determines whether stereoscopic objects are aligned on the same horizontal line, identifies and separates target objects based on geometric features, and performs distinct image processing for stereoscopic and target objects to prevent new 3D errors, using techniques such as morphological, temporal, and geometric feature analysis, and compensates for image holes to maintain object alignment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If image processing is performed on image pairs without differentiating stereoscopic objects from target objects, then processing is simplified, but 3D errors occur and video quality deteriorates
Solution Approach 1:
The patent segments the image processing task by first identifying and separating target objects from stereoscopic objects based on geometric features. Target objects are extracted using morphological operations and geometric property analysis, then processed independently from the stereoscopic objects. This segmentation allows different processing strategies for different object types, resolving the contradiction between processing simplicity and 3D quality.
Solution Approach 2:
The patent extracts target objects from the image pair by detecting their geometric properties (area, perimeter, aspect ratio) and separating them from the stereoscopic content. By taking out target objects before processing, the system avoids applying incorrect corrections to non-stereoscopic elements, thereby maintaining 3D video quality while managing processing complexity through structured separation.
2Ease of manufacture
If all objects in image pairs are treated as stereoscopic objects for correction, then processing is uniform, but target objects acquire new 3D errors
Solution Approach 1:
The patent applies local quality by treating target objects and stereoscopic objects differently based on their geometric properties. Target objects are identified through specific geometric criteria (area thresholds, aspect ratios) and excluded from stereoscopic correction processing. This localized differentiation ensures that only appropriate objects receive appropriate processing, maintaining alignment accuracy while preserving processing efficiency.
3Measurement precision
If target objects are separated from image pairs, then processing precision improves, but image holes are generated requiring compensation
Solution Approach 1:
The patent extracts target objects using geometric feature analysis and morphological operations, creating segmented regions that are then processed separately. This extraction achieves high identification accuracy by relying on distinct geometric properties of target objects versus stereoscopic objects, while the subsequent hole compensation uses standard interpolation techniques that add minimal complexity.
4Manufacturing precision
If geometric feature analysis is performed to identify target objects, then processing accuracy improves, but computational complexity increases
Solution Approach 1:
The patent applies partial action by using geometric feature analysis only for initial target object identification, rather than performing comprehensive analysis on all image elements. By focusing computational resources on calculating area, perimeter, and aspect ratio for region identification, the system achieves sufficient accuracy without excessive computational complexity. The geometric features provide an efficient first-pass separation that minimizes further processing needs.
Data Source
AI summary
An image processing method and apparatus are provided. The image processing method may include determining whether stereoscopic objects that are included in an image pair and that correspond to each other are aligned on the same horizontal line. The method includes determining whether the image pair includes target objects having different geometric features from those of the stereoscopic objects if the stereoscopic objects are not aligned on the same horizontal line. The method includes performing image processing differently for the stereoscopic objects and for the target objects if the image pair includes the target objects.