Toroidal Segmentation of Repeating Patterns for Clean Vector Objects
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vectorization techniques for images with repeating patterns often create unwanted boundaries and Bezier paths that pierce objects, making the output unusable without significant manual effort, especially when the minimal cell is treated as a single unit with high density or artistic arrangements.
Innovation Solution
A method involving toroidal segmentation to isolate and join adjacently placed segments, unwrapping toroidally connected regions, and converting the minimal cell into independent graphical objects, avoiding unwanted artifacts and extra paths.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the minimal cell is treated as a single unit for vectorization, then computing resources are saved and processing is simplified, but unwanted boundaries and Bezier paths are created that pierce objects and make the output unusable without significant manual effort
Solution Approach 1:
The patent divides the minimal cell into multiple segments based on detected objects within the cell. Each object is segmented into its own region, allowing independent vectorization. This segmentation prevents the creation of unwanted boundaries and Bezier paths that would occur if the entire minimal cell was treated as a single unit, while still maintaining the repeating pattern structure.
Solution Approach 2:
The patent introduces a dimensional approach by creating a graph representation where nodes represent image regions and edges represent adjacencies. This graph structure adds a relational dimension to the vectorization process, enabling the system to understand spatial relationships between objects and correctly handle toroidal connections without creating artifacts.
2Manufacturing precision
If toroidal segmentation is applied to isolate and join adjacently placed segments, then accurate vector representations are achieved, but the process complexity increases compared to simple minimal cell treatment
Solution Approach 1:
The patent segments the minimal cell into multiple regions based on object detection, then uses graph theory to manage the complexity of joining these segments. The graph structure organizes the segmentation process systematically, making the complexity manageable through structured processing rather than brute-force approaches.
Solution Approach 2:
The patent introduces a graph data structure as an intermediary between the segmented regions and the final vector representation. This graph acts as a mediator that tracks adjacencies and toroidal connections, simplifying the complex task of joining segments by providing a clear relational framework that guides the vectorization process.
3Productivity
If simple minimal cell vectorization is used, then processing is fast and simple, but seam lines and artifacts are created that require significant manual editing
Solution Approach 1:
The patent performs preliminary segmentation and graph construction before the actual vectorization. By pre-identifying objects, regions, and their relationships in the minimal cell, the system prepares all necessary information in advance. This preliminary action ensures that when vectorization occurs, the process is guided by pre-computed structures, reducing the need for manual editing to fix artifacts.
Solution Approach 2:
The patent incorporates feedback mechanisms where the graph structure continuously refines the vectorization process. The system monitors the vectorization output and adjusts the segmentation and joining processes based on detected issues, ensuring clean vector representations without seam lines or artifacts that would require manual correction.
Data Source
AI summary
Certain aspects and features of the present disclosure relate to segmenting repeating patterns in images to provide representations of individual objects according to certain embodiments. For example, a method involves segmenting an input image including a repeating pattern to identify input image regions of a minimal cell. The method further involves representing image regions using nodes of a graph, and defining edges between nodes that represents adjoining image regions. The method also involves identifying, using the graph, portions of toroidally connected image regions and portions of mergeable adjoining image regions. The method additionally involves joining each portion of each mergeable adjoining image region, and unwrapping each portion of each toroidally connected image region. The method also involves rendering or storing each of the resultant image regions as an object from the repeating pattern.