Edge Pair Encoding for Graphical Data Stream Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current printing systems face inefficiencies in transmitting graphical elements due to large data stream sizes and processing overhead, particularly when using formats like PostScript or fully rendered rasters, which require significant resources and struggle with compression and decompression, especially in bandwidth-limited channels.
Innovation Solution
The introduction of 'edge pairs' and 'two-dimensional run-length encoding (RLE)' formats for decomposing vector objects, allowing for more compact representations and efficient processing, particularly suitable for hardware implementation without the need for a CPU, and optimizing stream sizes through techniques like nibble and bit-based encoding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If graphical elements are transmitted in PDL format (PostScript, PCL), then the representation is flexible and comprehensive, but the data stream size becomes large and processing requires significant CPU resources
Solution Approach 1:
The patent segments graphical elements into distinct categories (text, shapes, images) and processes each segment with appropriate encoding methods. Vector objects are decomposed into edge pairs and encoded with compact representations, while raster images use specialized compression, avoiding the need to process entire pages as monolithic PDL streams.
Solution Approach 2:
The patent changes the parameter representation by transitioning from high-level PDL commands to low-level geometric primitives (edge pairs with explicit coordinates). This parameter transformation enables more efficient compression and hardware processing while maintaining visual fidelity.
2Device complexity
If graphical elements are fully rendered into raster format, then processing requirements are reduced, but the data stream size increases significantly and compression becomes difficult
Solution Approach 1:
The patent implements a dynamic processing pipeline where the rendering stage is separated from transmission. Vector objects are rendered to raster format on-demand at the receiver end, allowing the transmitted data to remain in compact vector form while still providing raster output when needed by the marking engine.
Solution Approach 2:
The patent introduces an intermediary representation layer between vector input and raster output. Edge pairs serve as this intermediary, providing a compact intermediate format that can be efficiently transmitted and then converted to raster format only when and where needed, avoiding premature full-page rasterization.
3Device complexity
If vector objects are decomposed into simple graphical primitives like triangles or trapezoids, then hardware processing becomes easier, but the data stream size increases and compression ratios deteriorate
Solution Approach 1:
The patent changes the dimensional representation by using edge pairs that define graphical elements through boundary coordinates rather than filling entire primitive shapes. This edge-based representation reduces redundancy compared to triangle/trapezoid decomposition while remaining suitable for hardware processing.
Solution Approach 2:
The patent extracts only the essential boundary information (edge pairs) needed to define graphical elements, discarding redundant internal structure data that would be present in full triangle or trapezoid decompositions. This extraction maintains hardware processability while minimizing data stream size.
4Quantity of substance
If compression is applied to reduce data stream size, then transmission efficiency improves, but decompression requires processing overhead that slows down rendering
Solution Approach 1:
The patent uses simple, lightweight encoding schemes (edge pairs with explicit coordinates, run-length encoding for raster data) that require minimal decompression processing. These compact representations can be decoded quickly by simple hardware logic without requiring complex decompression algorithms or significant CPU intervention.
Data Source
AI summary
Encoding a graphical element for processing can utilize an edge pair format in which the graphical element is divided into multiple objects, where each object contains exactly two edges which do not cross or self-intersect. Another format is a run-length encoding (RLE) format in which the graphical element is divided into multiple lines, where the RLE format includes an X start position, a Y start position, a length of a first of the lines, and, for each subsequent line, indications of right and left edge steps relative to the immediately proceeding line.