Parallel Band Rendering Without Data Division
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing band parallel rendering techniques require dividing image data across multiple bands, increasing processing costs due to the need to generate intermediate data for each band and perform rendering separately, which is inefficient for images spanning multiple bands.
Innovation Solution
An image processing apparatus with a first and second rendering unit that accesses and renders image data across areas of a page, allowing parallel rendering without pre-decompression and division of image data, using a shared decompression process across band areas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If image data is divided for each band and intermediate data is generated for each divided image, then band parallel processing can be performed, but processing cost increases
Solution Approach 1:
The patent divides the page area into multiple bands and assigns different rendering units to process different bands in parallel. Each rendering unit processes only the scan lines within its assigned band range, enabling parallel processing without dividing the image data itself. This resolves the contradiction by segmenting the processing task rather than the data.
Solution Approach 2:
The patent uses a single decompression unit that serves all rendering units. The decompression unit decompresses the entire image data once, and all rendering units access the decompressed data as needed. This multi-functional approach eliminates the need for separate decompression processes for each band, reducing processing cost while maintaining parallel rendering capability.
2Productivity
If image data is pre-decompressed and divided for each band, then rendering can be performed, but processing cost increases
Solution Approach 1:
The decompression unit performs preliminary decompression of the compressed image data before rendering. However, instead of dividing the image data afterward, the system allows rendering units to access the fully decompressed data and process only their assigned band portions. This preliminary action enables parallel processing without the overhead of subsequent data division.
Solution Approach 2:
Multiple rendering units access the same decompressed image data simultaneously, each copying only the portion needed for their assigned band. This approach avoids the need to create multiple copies of divided image data, reducing memory usage and processing overhead while enabling parallel rendering.
3Productivity
If image data is accessed by multiple rendering units, then parallel rendering is enabled, but memory access conflict may occur
Solution Approach 1:
Each rendering unit is assigned a specific band range and processes only the scan lines within that range. This local processing approach ensures that while multiple rendering units access the same decompressed image data, they access different portions simultaneously, minimizing memory access conflicts and ensuring stable parallel operation.
Data Source
AI summary
An image processing apparatus uses a first rendering processor that performs rendering of a first area within a page based on print data and a second rendering processor that performs rendering of a second area within the page based on the print data. The image processing apparatus includes a controller, having a processor which executes instructions stored in a memory or having circuitry, being configured to acquire image data arranged across the first area and the second area based on the print data, cause the first rendering processor having accessed the image data to acquire all pixels of the image data and to perform rendering with pixels within the first area of all the pixels of the image data, and cause the second rendering processor having accessed the image data to acquire all the pixels of the image data and to perform rendering with pixels within the second area of all the pixels of the image data.