Band Area Rasterization with Selective Decompression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional parallel drawing processing in units of bands requires significant hardware resources, leading to increased costs due to the need for ample memory and multiple decompression units, which limits processing speed and efficiency.
Innovation Solution
An image processing method that performs rasterization in units of band areas, where raster data is compressed and stored in memory, allowing for concurrent processing and selective decompression only for necessary areas, thereby reducing the hardware resources required for parallel operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If drawing processing is performed in parallel in units of bands with sufficient memory capacity, then processing speed is improved, but hardware resources and cost increase
Solution Approach 1:
The patent divides the drawing processing into segments by performing rasterization in units of band areas rather than processing entire pages at once. This segmentation allows parallel processing of multiple bands simultaneously while reducing the memory capacity required for each processing unit, thereby improving processing speed without proportionally increasing hardware resources.
Solution Approach 2:
The patent changes the processing parameter from page-level parallel processing to band-level parallel processing. By adjusting the granularity of processing units from entire pages to smaller band areas, the system achieves better utilization of limited memory resources while maintaining parallel processing capabilities, thus improving productivity without linearly increasing device complexity.
2Productivity
If memory capacity is increased to support parallel band processing, then processing efficiency is improved, but cost increases
Solution Approach 1:
By segmenting the processing into smaller band units, the patent reduces the amount of memory each processing unit needs to hold at once. Multiple bands can be processed in parallel with smaller individual memory allocations, achieving high processing efficiency without requiring proportionally large total memory capacity.
Solution Approach 2:
The patent performs preliminary rasterization of drawing objects into band-specific intermediate data before parallel processing. This preliminary action prepares the data in a format that requires less memory during the actual parallel rendering phase, improving processing efficiency while controlling memory capacity requirements.
3Quantity of substance
If fallback process is used to divide drawing processing into multiple pieces, then memory capacity requirement is reduced, but processing time increases
Solution Approach 1:
The patent implements segmentation by processing drawing objects in units of bands rather than requiring complete fallback processing of entire pages. This segmentation enables parallel execution of multiple band processing tasks simultaneously, reducing the overall processing time while maintaining reduced memory capacity requirements through the band-wise division of work.
Data Source
AI summary
An image processing method of performing rasterization of a page in units of band area includes, among other things, performing, in units of band areas, rasterization based on second intermediate data stored in a first memory area for storing intermediate data and compressed raster data stored in a second memory area. The rasterization of the plurality of band areas is performed in parallel. In addition, the performing of the rasterization based on the second intermediate data and the compressed raster data includes selectively decompressing, in units of predetermined areas, a piece of the compressed raster data stored in the second memory area, in accordance with which band areas, of a plurality of band areas, are to be subject to the rasterization.