Image Reading Apparatus Shading Correction via Color Ratio Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image reading apparatuses face inaccuracies in shading correction due to chromatic aberration and sensitivity differences in light receiving elements, leading to unnatural variations in read images when foreign particles or flaws are present on the reference member.
Innovation Solution
An image reading apparatus with a plurality of light sources emitting colors according to color separation technology, utilizing a ratio calculator to determine abnormal data in white level information and a data corrector to replace it with correction data from other colors, ensuring accurate shading correction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If white level information is obtained by reading the reference member with multiple color light sources, then color accuracy is improved, but foreign particles or flaws on the reference member cause abnormal data that degrades measurement precision
Solution Approach 1:
The patent introduces an intermediary processing step that calculates ratios among white level information of different colors and uses this ratio relationship to identify and correct abnormal data caused by foreign particles. Instead of directly using raw white level data, the system mediates through ratio calculation to detect and eliminate harmful effects of foreign particles on measurement precision.
Solution Approach 2:
The patent transforms the white level information by calculating ratios among different color channels (R, G, B) to create a new parameter set that is more robust against foreign particle interference. This parameter transformation allows the system to maintain measurement precision even when foreign particles are present on the reference member.
2Stability of the object's composition
If shading correction is performed using white level information from all colors, then color balance is improved, but chromatic aberration and sensitivity differences cause unnatural variations in the corrected image
Solution Approach 1:
The patent implements a feedback mechanism where the system calculates ratios among white level information of different colors, compares these ratios to detect abnormalities, and uses this feedback to selectively replace abnormal data. This feedback loop ensures that shading correction maintains color balance while eliminating unnatural variations caused by chromatic aberration and sensitivity differences in light receiving elements.
3Device complexity
If abnormal data replacement is performed without considering color ratios, then data correction is simplified, but color accuracy deteriorates due to chromatic aberration differences
Solution Approach 1:
The patent changes the parameter representation from absolute white level values to relative ratios among color channels. This parameter transformation enables the system to maintain color accuracy during abnormal data replacement by preserving the relative relationships among R, G, and B components, thereby compensating for chromatic aberration differences without increasing process complexity.
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
The solution effectively corrects white level information abnormalities, eliminating shading lines and improving image accuracy by using the white level information of other colors not affected by foreign particles or flaws.
Implementation Method 1
The light sources emit light of a plurality of colors, namely, red (R), green (G), and blue (B), toward the platen glass so that the emitted light of respective colors is transmitted through the platen glass and reflected by a surface of the document. The reflected light is received by the light receiving elements that convert the received light of the respective colors RGB into electrical signals.
Data Source
AI summary
An image reading apparatus for reading an image with respect to a plurality of colors, including: a reference member; an image reading device including a plurality of light receiving elements, the image reading device reading the reference member to obtain white level information for the respective colors, the white level information for each of the colors being a set of data of a quantity obtained by the light receiving elements, the read image being corrected based on at least the white level information; a ratio calculator calculating a ratio among the information of the respective colors; an abnormal-data determiner determining that a first piece of data in the level information of any particular color is abnormal data, where the first piece of data is different from a second specific piece of data in the level information of the same color which second specific piece of data is in a predetermined relationship with the first specific data, by an amount not smaller than a threshold, or where the first specific data is outside a range; and a data corrector obtaining correction data obtained by multiplying, by the ratio, corresponding data which corresponds to the abnormal data according to a predefined principle and included in the information of at least one color determined to be not including the abnormal data, and replaces the abnormal data with the correction data.