Document Processing System White Space Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current image processing systems, both hardware and software-based, face challenges in optimizing performance due to unnecessary processing of 'white space' regions, leading to prolonged execution times and potential artifacts from error diffusion in reassembling image parts.

Innovation Solution

A method and system that determine the background level of white space in a document, identify and exclude pixels belonging to it, generate blocks of image content, and process these blocks in parallel, eliminating the need for processing white space pixels, thereby reducing computational load and improving output quality.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If pixel-by-pixel processing is performed on every region including white space, then complete image processing is achieved, but execution time is prolonged

Engineering Contradiction:
Improveimage processing completenessVSAvoidexecution time
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The image is segmented into white space regions and image content regions. By identifying and separating these regions, the system processes only the image content regions, eliminating unnecessary processing of white space areas while maintaining complete processing of all image content.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent extracts and identifies white space regions from the image, then excludes these regions from further processing. This extraction of unnecessary processing targets reduces computational load and execution time while preserving processing of essential image content.

Inventive Principle:
Principle #2Taking out (Extraction)

2Productivity

If parallel processing is used to reduce print time, then processing speed is improved, but artifacts may appear at boundaries of processed parts

Engineering Contradiction:
Improveprocessing speedVSAvoidoutput quality
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

By extracting white space regions and excluding them from processing, the patent eliminates the boundary regions where artifacts typically appear in parallel processing. This approach maintains processing speed while preventing quality degradation at boundaries.

Inventive Principle:
Principle #2Taking out (Extraction)

3Manufacturing precision

If computational processing is performed on all pixels, then thorough image processing is achieved, but computational load is excessive

Engineering Contradiction:
Improveprocessing thoroughnessVSAvoidcomputational load
Core Design Contradiction:
Manufacturing precisionVSLoss of energy

Solution Approach 1:

The patent segments the image into processable regions (image content) and non-processable regions (white space). This segmentation enables thorough processing of essential content while avoiding unnecessary computational expenditure on white space areas.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent extracts white space regions and removes them from the processing pipeline, reducing computational load by eliminating operations on pixels that do not require processing while maintaining thorough processing of image content regions.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentUS11087126B2Method to improve performance in document processing
Publication Date: 2021.08.10 XEROX CORP
  • US11087126B2 patent drawing
  • US11087126B2 patent drawing
  • US11087126B2 patent drawing

AI summary

A document processing system produces a digital representation of an original image by determining a background level for white space of the original image. Using the background level, the system identifies pixels not belonging to the white space. The system generates a binary image of the original image based from the identified pixels. The system then identifies image content surrounded by the white space in the original image using the binary image. Next, the system generates a block for each image content and extracts each block from the original image. After processing each block, the system reapplies the processed blocks to a blank canvas to generate the digital representation.