Image Processing Apparatus Bidirectional Path Selection for Area Separation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing techniques face challenges in effectively separating areas from images, particularly in identifying the correct contours and paths to segment characters from overlapping patterns, which often require complex calculations and storage of pixel positions, leading to increased data processing and potential inefficiencies.
Innovation Solution
The proposed image processing apparatus employs a path convergence method using first and second path information calculating units and a path selecting unit to determine the shortest path as the contour for separating areas, which involves calculating forward and backward paths and selecting the optimal path based on convergence, reducing the need to store pixel positions between start and end points and adjusting the number of round-trips for efficient separation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional area separation techniques are used to segment characters from overlapping patterns, then separation accuracy can be maintained, but data processing complexity and storage requirements increase significantly
Solution Approach 1:
The patent segments the area separation problem into two independent directional processes: forward direction (left to right) and backward direction (right to left). Each direction processes path information independently, calculating minimum cost paths separately. This segmentation allows the system to avoid complex simultaneous multi-directional calculations while maintaining accurate character separation from overlapping patterns.
Solution Approach 2:
The patent applies inversion by processing path information in both forward and backward directions. Instead of attempting to solve the separation problem from a single perspective, the system calculates paths from both directions and integrates the results. This bidirectional approach simplifies the overall calculation by breaking down the complex overlapping pattern separation into two manageable directional passes.
2Measurement precision
If pixel positions between start and end points are stored for path calculation, then accurate contour identification is achieved, but memory storage requirements increase
Solution Approach 1:
The patent extracts and processes only the essential path information needed for contour identification, rather than storing all pixel positions between start and end points. By focusing on calculating minimum cost paths in forward and backward directions, the system extracts only the critical separation paths needed for accurate contour identification, significantly reducing memory storage requirements while maintaining precision.
Solution Approach 2:
The patent performs partial action by calculating paths in two directions (forward and backward) rather than attempting to store and process all possible pixel positions. This partial approach to path calculation achieves sufficient contour identification accuracy without the excessive memory storage that would be required to maintain all intermediate pixel data.
3Measurement precision
If complex threshold settings are applied for area separation, then segmentation accuracy improves, but processing time increases
Solution Approach 1:
The patent changes the parameter approach from threshold-based segmentation to cost-based path calculation. Instead of applying complex threshold settings to pixel intensities, the system calculates minimum cost paths where the cost function incorporates both directional information and separation quality. This parameter change maintains high segmentation accuracy while avoiding the time-consuming iterative threshold adjustment process.
Solution Approach 2:
The patent performs preliminary action by pre-calculating path information in both forward and backward directions before final separation. By preparing the minimum cost path data in advance through directional passes, the system eliminates the need for complex threshold settings during the actual segmentation process, thereby reducing processing time while maintaining accuracy.
Data Source
AI summary
An image processing apparatus includes a first path information calculating unit, a second path information calculating unit, and a path selecting unit. The first path information calculating unit calculates first path information which is information representing a first path for separating areas from an image. The second path information calculating unit calculates second path information representing a second path for separating the areas from the image, the second path being the reverse of the first path. The path selecting unit selects one of the first path information calculated by the first path information calculating unit and the second path information calculated by the second path information calculating unit.