Image Binarization for Faint Blue Characters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional binarization techniques fail to effectively preserve faint characters, such as carbon-copied blue characters and handwritten text, during the image processing of scanned documents, often converting them into white pixels and causing them to disappear.
Innovation Solution
An image processing apparatus and method that extracts pixels within a specific color range, adjusts their gradation values to lower brightness, generates a histogram, sets a threshold value based on the adjusted brightness values, and applies a binarization process to ensure that these characters are converted to black pixels rather than white, using a scanner and personal computer system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional binarization technique is applied to image data containing faint characters, then the binarization process converts pixels to black or white based on threshold comparison, but the faint characters are converted into white pixels and disappear
Solution Approach 1:
The patent applies preliminary color-based pixel adjustment before binarization. Pixels within the blue color range have their brightness values reduced in advance, ensuring that faint blue characters are darkened sufficiently before the threshold comparison occurs. This preliminary action prevents the loss of character information during subsequent binarization.
Solution Approach 2:
The patent applies different brightness adjustment strategies to different color ranges. Specifically, pixels within the blue color range undergo brightness reduction, while other pixels maintain their original brightness. This localized quality adjustment ensures that only the relevant faint blue characters are enhanced without affecting the rest of the image.
2Loss of information
If the threshold value is set to preserve faint characters, then character visibility improves, but the overall binarization quality and contrast may deteriorate
Solution Approach 1:
The patent changes the brightness parameter of pixels within the blue color range before binarization. By reducing the brightness values of blue pixels, the patent effectively shifts the distribution of brightness values for character pixels, allowing the threshold to be set at an optimal point that preserves both character visibility and overall image contrast.
3Loss of information
If brightness values of all pixels are adjusted uniformly, then character visibility may improve, but the natural color distribution and image quality are distorted
Solution Approach 1:
The patent applies brightness adjustment selectively to pixels within the blue color range while leaving other pixels unchanged. This localized approach preserves the natural color distribution and brightness characteristics of the rest of the image, maintaining overall image quality while enhancing only the relevant faint blue characters.
Data Source
AI summary
An image processing apparatus has a storage configured to store image data representing a color image, the image data being constituted by multiple pixels, each of the multiple pixels having a gradation value, and a controller. The controller is configured to extract pixels, from among the multiple pixels, within a specific color range as target pixels, adjust the gradation values of the target pixels such that brightness values of the target pixels are lowered, generate a histogram of index values corresponding to quantities of the brightness values of the multiple pixels constituting the image data after the gradation values of the target pixels are adjusted, set a first threshold value based on the histogram as generated, and apply a binarizing process to the image data using the first threshold value.