Image Processing Device Near-Infrared Pattern Removal
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing technologies face challenges in enhancing the readability of documents with ground patterns, as they often improperly remove essential information due to insufficient image quality or overlapping patterns, leading to issues like color shift and erroneous removal of text or codes.
Innovation Solution
The use of near-infrared light and image sensors that distinguish between black and color toners based on their optical characteristics in the invisible wavelength range, allowing for accurate extraction and removal of ground patterns and imprints while preserving the readability of text and codes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If color dropout processing is used to remove ground patterns and ruled lines, then the readability of text information is improved, but color shift occurs and essential information may be improperly removed when image quality is insufficient
Solution Approach 1:
The patent segments the image processing into distinct stages: first performing color dropout processing to remove color-based ground patterns and ruled lines, then performing black pattern removal processing to remove black imprints and patterns. This segmentation allows each processing stage to target specific types of interference patterns without affecting the other, thereby improving text readability while preserving essential information.
Solution Approach 2:
The patent changes the processing parameters by switching between different removal methods based on the characteristics of the interference patterns. For color-based patterns (ground patterns, ruled lines), color dropout processing is applied. For black patterns (imprints, stamps), black pattern removal processing is applied. This parameter change approach ensures accurate removal of various pattern types while maintaining the reliability of essential information.
2Measurement precision
If ground patterns are removed to enhance document readability, then text recognition is improved, but security features and authentication elements may be improperly removed
Solution Approach 1:
The patent applies different processing qualities to different regions and types of patterns locally. Color dropout processing is applied specifically to color-based ground patterns and ruled lines, while black pattern removal is applied to black imprints and stamps. This local quality approach ensures that security features with specific characteristics are preserved while removing only the interfering patterns that match the processing criteria.
Solution Approach 2:
Instead of removing all patterns uniformly, the patent inverts the approach by selectively removing only specific types of patterns based on their color and characteristics. Essential security features that do not match the removal criteria (such as black security patterns or color-coded authentication elements) are preserved, while interfering patterns are removed.
3Measurement precision
If aggressive pattern removal is applied to eliminate all ground patterns and imprints, then document readability is enhanced, but color shift and distortion of original information occur
Solution Approach 1:
The patent applies partial action by performing color dropout processing first to remove color-based patterns, then applying black pattern removal processing to remove black patterns. This partial application of removal techniques ensures that only necessary patterns are removed while preserving essential information, avoiding the excessive action that would cause color shift and distortion.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach effectively enhances the readability of documents by accurately identifying and removing secondary information, such as ground patterns and imprints, without affecting the quality of the content information, even in cases of overlapping or low-quality scans, thereby improving the recognition of text and codes.
Implementation Method 1
A light source 2 irradiates an object with invisible light, and an image sensor 9 serving as an image reader reads first information and second information included in the object and obtainable in response to the object being irradiated with the invisible light
Implementation Method 2
black and color toners are distinguished from one another according to a difference in optical characteristics in an invisible wavelength range
Data Source
Figure 1~2
Figure 3
Figure 4~5
AI summary
An image processing device (101) includes a light source (2), an image reader (9), and an image selecting unit (21). The light source (2) is configured to irradiate an object at least with invisible light. The image reader (9) is configured to read first information and second information. The first information is included in the object and indicating a first characteristic in response to the object being irradiated with the invisible light. The second information is included in the object and indicating a second characteristic in response to the object being irradiated with the invisible light. The image selecting unit (21) is configured to selectively output the first information read by the image reader (9).