Image Processing Apparatus Shadow Correction Grid
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing methods for correcting shadows in bound documents, such as books or magazines, require large and costly circuits due to the need for extensive luminance correction calculations, and lack effective optimization of image quality by not properly determining block sizes and aspect ratios for luminance correction.
Innovation Solution
A method that divides the shadow into rectangular blocks using grid lines perpendicular to the document's scanning and sub-scanning directions, calculates correction values for grid points, and stores these values in memory to correct luminance irregularities while preserving images, using interpolation and weighting coefficients to optimize memory usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If luminance correction values are calculated for all pixel positions using interpolation calculation, then image quality after correction is high, but circuit size and cost increase
Solution Approach 1:
The patent divides the image into multiple blocks and calculates luminance correction values only at grid points (block corners) rather than for all pixels. This segmentation approach reduces the number of correction values that need to be stored and processed, thereby reducing circuit complexity while maintaining correction effectiveness through subsequent interpolation for non-grid pixels.
Solution Approach 2:
Instead of calculating correction values for every pixel position, the patent performs partial action by calculating values only at discrete grid points. The remaining pixel corrections are obtained through interpolation from these grid point values, achieving adequate correction with reduced computational and storage requirements.
2Ease of operation
If square grid is used for dividing the shadow region, then calculation is simple, but luminance correction accuracy is reduced
Solution Approach 1:
The patent employs asymmetric rectangular blocks instead of symmetric square grids for dividing the shadow region. This asymmetric segmentation better adapts to the actual geometry and luminance distribution characteristics of bound document shadows, improving correction accuracy while maintaining a systematic grid-based approach for manageable complexity.
Solution Approach 2:
The patent applies different block dimensions (asymmetric rectangular shapes) to different regions of the shadow based on local luminance characteristics. This local adaptation allows the grid structure to better match the actual shadow patterns in different areas, improving overall correction precision without requiring completely complex local analysis.
3Device complexity
If correction value memory capacity is limited, then device cost is reduced, but correction precision is compromised
Solution Approach 1:
By segmenting the image into blocks and storing correction values only at grid points (block corners) rather than for all pixels, the patent dramatically reduces the number of memory storage requirements. This segmentation strategy enables precise correction functionality to be implemented within limited memory capacity constraints.
Solution Approach 2:
The patent uses interpolation calculation as an intermediary process to generate correction values for non-grid pixels based on the stored grid point values. This intermediary approach allows the system to achieve comprehensive pixel-level correction precision while only storing a limited set of correction values at grid points, effectively bridging the gap between limited memory capacity and high correction precision requirements.
Data Source
AI summary
An image correcting method for an image processing apparatus that reads a bound document and corrects a shadow in a bound portion in obtained document image data includes extracting shadow image data from the document image data, generating a grid with which a ratio of changes in luminance between the grid lines in directions perpendicular to each other is smaller than in a square grid and which has grid points within a capacity of a memory for correction values, calculating first correction values that corrects luminance of pixels in the shadow image data corresponding to grid points, storing the first correction values in the memory, calculating second correction values for luminance of the pixels in the shadow image data based on first correction values of grid points close to each pixel, and correcting luminance of the document image data using the first and second correction values.