Image Reading Shading Correction via Singular Point Pixel Interpolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image reading apparatuses face challenges in generating accurate shading correction data due to the influence of dirt and dust, particularly when using a white reference chart, leading to reduced effectiveness of shading correction, especially when the chart is dirty or has stripes caused by dust during conveyance.
Innovation Solution
The apparatus employs a determination unit to identify singular point pixels by comparing image data with threshold values, interpolates data from surrounding pixels, and uses a sampling unit to select valid shading correction data, thereby reducing the impact of dust and dirt, and controlling the number of effective sampling lines to improve data accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a white reference chart is used for generating shading correction data, then shading correction can be performed, but the accuracy of correction data is reduced when the chart is dirty or has dust stripes
Solution Approach 1:
The patent extracts and removes singular point pixels (dust and dirt) from the image data by comparing each pixel with threshold values. This extraction process separates the harmful dust particles from the valid image data, allowing the system to generate accurate shading correction data without being affected by chart contamination.
Solution Approach 2:
The patent converts the harmful effect of dust and dirt into a beneficial process by using threshold comparison to identify singular points. The dust particles, while initially harmful, become useful markers that guide the interpolation process to recover accurate shading information from surrounding clean pixels.
2Measurement precision
If all pixels are used for sampling shading correction data, then data collection is simple, but accuracy is reduced due to inclusion of dust-affected pixels
Solution Approach 1:
The patent segments the image data into two categories: singular point pixels (dust-affected) and normal pixels (clean). By dividing the data processing into these segments, the system can selectively process only the necessary pixels for accurate shading correction, improving precision without excessive complexity.
Solution Approach 2:
The patent applies local quality by treating different pixels differently based on their characteristics. Clean pixels are used directly for shading correction data, while dust-affected pixels are identified and replaced through interpolation from surrounding pixels, ensuring each pixel contributes optimally to the correction accuracy.
3Measurement precision
If singular point pixels are removed and interpolated from surrounding pixels, then accuracy of shading correction data is improved, but processing time increases
Solution Approach 1:
The patent applies partial action by performing interpolation only on the singular point pixels (dust-affected areas) rather than processing the entire image. This selective approach maintains high accuracy where needed while minimizing unnecessary processing time on already-clean pixels.
Data Source
AI summary
An image reading apparatus that makes it possible to reduce the influences of dirt on a white reference chart itself, stripes caused by dust, and the like, and thereby generate accurate data for shading correction. An image processor performs shading correction on image data read by a reader unit. The read image data is compared with threshold values set for each pixel, and a pixel which is out of a range of the threshold values is determined as a singular point pixel. Data of a pixel determined as a singular point pixel is interpolated from data around the singular point pixel. Data of a pixel which is not determined as a singular point pixel is adopted as valid shading correction data. An operation controller controls the number of effective sampling lines from start to termination of sampling to be performed.