Background Element Detection in Financial Document Image Processing

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing image processing technologies fail to accurately discriminate between intended images and background patterns in financial documents, as they are not designed to determine the presence of background elements, leading to incomplete removal of background patterns.

Innovation Solution

An image processing apparatus and method that generates a decimated image, extracts edge pixels, creates an isolated histogram, and uses this data to decide whether an input image contains a background element, allowing for precise discrimination and removal of background patterns.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If existing image processing technologies are used to remove background patterns, then some background removal capability is achieved, but the ability to accurately determine the presence of background elements is lost

Engineering Contradiction:

Improveaccuracy of background element determinationVSAvoidcapability to handle both documents with and without background patterns

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent applies preliminary action by performing pixel decimation and edge pixel extraction before determining the presence of background elements. The system pre-processes the image data to create a decimated image and identify edge pixels, then uses these pre-processed features to accurately determine whether background elements exist before proceeding with background removal operations.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent applies segmentation by dividing the image processing into distinct stages: pixel decimation to create a reduced-resolution image, edge pixel extraction to identify boundary elements, isolated edge pixel identification to find specific characteristic pixels, and histogram generation to analyze the distribution of these pixels. This segmented approach enables accurate background element determination.

Inventive Principle:

Principle #1Segmentation

2Productivity

If background pattern removal is applied to all documents, then documents with background patterns are processed, but documents without background patterns may be incorrectly processed

Engineering Contradiction:

Improveprocessing throughputVSAvoidaccuracy of intended image extraction

Core Design Contradiction:

ProductivityVSManufacturing precision

Solution Approach 1:

The patent applies feedback by generating a histogram from isolated edge pixels and using this histogram information to determine whether background elements are present before applying background removal operations. The system continuously monitors the image characteristics through histogram analysis and adjusts its processing approach based on the detected presence or absence of background patterns, ensuring high accuracy in intended image extraction.

Inventive Principle:

Principle #23Feedback

3Measurement precision

If simple binarization is used, then processing speed is maintained, but background patterns cannot be accurately removed from documents with varying density

Engineering Contradiction:

Improveaccuracy of background pattern discriminationVSAvoidcomplexity of image processing operations

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies local quality by focusing analysis on specific pixels with particular characteristics - namely, isolated edge pixels that exhibit specific density relationships with their neighbors. Rather than analyzing all pixels uniformly, the system identifies and analyzes only those pixels that are likely to indicate the presence of background elements, maintaining measurement precision while managing complexity.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The patent applies parameter changes by transforming the original image through pixel decimation to create a decimated image with different resolution characteristics. This parameter change in image resolution enables more effective edge pixel detection and histogram generation, improving the accuracy of background pattern discrimination.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS8824796B2Image processing apparatus, image processing method, and computer-readable, non-transitory medium

Publication Date: 2014.09.02 PFU LTD

AI summary

There are provided an image processing apparatus, image processing method, and a computer-readable, non-transitory medium that can determine with high accuracy whether or not an input image contains a background element. The image processing apparatus includes a decimated image generator for generating a decimated image through pixel decimation from an input image, an edge pixel extractor for extracting edge pixels from the decimated image, an edge class extractor for extracting an isolated edge pixel from among the edge pixels, a histogram generator for generating an isolated histogram based on the isolated edge pixel, and a decision unit for making, based on the isolated histogram, a decision as to whether or not the input image contains a background element.