Raster Image Segmentation With Half-Edge Scanline Boundaries
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional raster image segmentation algorithms are inefficient and inflexible, particularly in handling complex images, due to their reliance on computationally expensive flood fill operations and limited multithreading capabilities.
Innovation Solution
A color segmentation system utilizing half-edge mesh structures and scanline operations to generate oriented polyline boundary loops, enabling efficient and flexible raster image segmentation, including multithreading capabilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional flood fill algorithms are used for raster image segmentation, then segmentation can be performed, but processing time and computational resources increase significantly
Solution Approach 1:
The patent segments the image processing task into scanline-based processing units, where each scanline is processed independently to identify boundaries. This divides the complex segmentation problem into smaller, manageable units that can be processed efficiently and parallelized.
Solution Approach 2:
The patent performs preliminary actions by pre-processing the image to identify and store boundary information during scanline operations. Boundaries are identified and stored in advance during the scanline pass, eliminating the need for expensive post-processing flood fill operations.
2Adaptability or versatility
If conventional segmentation algorithms are used, then basic segmentation is achieved, but flexibility for multithreading and complex images is limited
Solution Approach 1:
The patent divides the image into scanlines that can be processed independently, enabling multithreading. Each scanline or group of scanlines can be handled by separate threads, improving parallel processing capability while maintaining manageable algorithm complexity through structured processing.
Solution Approach 2:
The patent implements dynamic adaptation to different image complexities through the scanline processing framework. The same basic scanline algorithm can handle simple images efficiently and scale to complex images by processing boundaries and zones dynamically during the scanline pass.
3Measurement precision
If naive segmentation algorithms are used, then implementation is simple, but accuracy and efficiency for complex raster images deteriorate
Solution Approach 1:
The patent performs preliminary boundary identification during scanline operations, storing boundary information in advance. This preliminary action enables accurate segmentation without requiring complex post-processing, achieving both high accuracy and efficiency by doing the work during the initial scanline pass.
Solution Approach 2:
The patent maintains continuous useful action by processing boundaries and zones continuously during the scanline operations. Instead of discrete flood fill operations, the algorithm continuously identifies and stores boundary information as it scans through the image, improving both accuracy and efficiency.
Data Source
AI summary
Methods, systems, and non-transitory computer readable storage media are disclosed for generating segmentations of a raster image via a half-edge mesh structure with scanline operations. The disclosed system determines, during scanline operations on a raster image, a plurality of sets of adjacent pixels having a common color value in the raster image. The disclosed system determines, during the scanline operations on the raster image, a plurality of half-edges at edges of pixels along a boundary of a set of adjacent pixels of the plurality of sets of adjacent pixels with next half-edge directions indicating directions of subsequent half-edges along the boundary of the set of adjacent pixels. The disclosed system generates one or more oriented polyline boundary loops representing the boundary of the set of adjacent pixels from the plurality of half-edges and the next half-edge directions of the set of adjacent pixels.