Vector Graphics Rasterization Using B-Spline Stencil Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for rasterizing vector graphics on devices with low processing capacity, such as mobile telephones and car navigation systems, result in increased storage and processing costs due to the need for high approximate accuracy using numerous triangles, leading to rough raster graphics when enlarged.
Innovation Solution
An image processing apparatus comprising a first unit to generate triangular data from vector format graphic data, a second unit to generate stencil data based on the drawing order of vertices, and a third unit to produce raster format graphic data using the stencil data, allowing for efficient rasterization based on the non-zero rule, independent of resolution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If numerous triangles are used to approximate curves for high rasterization accuracy, then the smoothness of raster graphics is improved, but the storage capacity and processing cost increase
Solution Approach 1:
The patent segments the curve approximation process by using B-spline basis functions to represent curves analytically rather than using numerous triangular segments. This allows high accuracy curve representation with minimal data storage, as the curve is defined by a few control points and mathematical functions rather than many discrete triangle vertices.
Solution Approach 2:
The patent changes the parameter representation from geometric approximations (triangles) to mathematical parameters (B-spline basis functions). This parameter transformation enables exact curve representation through analytical computation, eliminating the need to increase the quantity of geometric elements to improve accuracy.
2Manufacturing precision
If numerous triangles are used to approximate curves for high rasterization accuracy, then the smoothness of raster graphics is improved, but the processing cost increases
Solution Approach 1:
The patent segments the rendering process into two independent stages: (1) analytical curve evaluation using B-spline basis functions to determine exact pixel coverage, and (2) efficient rasterization using the pre-computed curve data. This segmentation eliminates the need for expensive real-time triangle approximation while maintaining high accuracy.
Solution Approach 2:
The patent performs preliminary analytical computation of curve properties (intersections with scanlines, pixel coverage) before the actual rasterization process. By pre-computing these mathematical evaluations, the subsequent rasterization becomes much faster and less computationally intensive, as it only needs to apply the pre-determined curve characteristics rather than perform complex geometric approximations.
3Ease of operation
If vector graphics are rasterized using traditional methods on low-processing-capacity devices, then the image can be displayed, but the arithmetic costs become too high
Solution Approach 1:
The patent replaces the mechanical/traditional rasterization approach (triangle-based pixel-by-pixel evaluation) with a mathematical/analytical system (B-spline basis function evaluation). This substitution transforms the computation from intensive arithmetic operations on geometric data to more efficient mathematical function evaluations, making the process feasible for low-processing-capacity devices.
Data Source
AI summary
An image processing apparatus includes, a first data generation unit configured to generate triangular data by analyzing straight line contours and curved contours with respect to vector format graphic data and making the straight line contours and the curved contours triangles, a second data generation unit configured to generate stencil data from the triangular data on the basis of a drawing order of vertices of the triangular data generated from straight line contours and curved contours, and a third data generation unit configured to generate raster format graphic data with reference to the stencil data.