Image Data Correction Using Pixel Distance Coefficients
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing systems require large memory capacity to correct toner density variations, leading to increased circuit scale and costs due to the need for extensive observation windows to detect edges and correct image data, especially in electro-photographic processes where toner sweeping causes blur and uneven density.
Innovation Solution
An image processing apparatus that shifts image data one pixel at a time, detects edges using a minimal observation window, and calculates correction values based on distance coefficients and edge strength, allowing for pixel-by-pixel correction without needing to observe the entire image region simultaneously, thereby reducing memory requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If a large observation window is used to detect edges and correct image data, then image quality correction is improved, but memory capacity and circuit scale increase
Solution Approach 1:
The patent divides the image correction process into two independent stages: edge detection using a small observation window (3-5 pixels) and correction value calculation using distance coefficients. This segmentation allows the system to achieve accurate correction without requiring a large observation window to hold both edge detection and correction regions simultaneously in memory.
Solution Approach 2:
The patent performs preliminary edge detection using a minimal observation window to identify edge positions, then uses these edge positions to calculate correction values based on distance coefficients. By performing the detection action first and storing only the edge position information, the system avoids the need to maintain large observation windows in memory throughout the entire correction process.
2Manufacturing precision
If a large observation window is used to detect edges and correct image data, then image quality correction is improved, but circuit scale increases
Solution Approach 1:
The patent separates the correction process into edge detection (using small window) and correction value calculation (using distance coefficients from edge positions). This segmentation reduces the circuit scale by eliminating the need for large observation windows that would be required to hold both detection and correction regions simultaneously.
Solution Approach 2:
The patent introduces distance coefficients as an intermediary mechanism that translates edge position information into correction values. This intermediary allows the system to achieve accurate correction without directly processing large observation windows, thereby reducing circuit complexity.
3Productivity
If pipeline process is used to connect image processes, then processing efficiency is improved, but memory requirements increase due to inability to reuse resources
Solution Approach 1:
The patent segments the image processing into independent stages (edge detection, correction value calculation, correction application) that can be processed sequentially through the pipeline. Each stage uses minimal memory resources, allowing the pipeline to maintain high processing efficiency while avoiding the memory requirements of traditional approaches that need to hold entire observation windows across all stages.
Data Source
AI summary
An image processing apparatus includes: a pipeline processor that inputs image data shifted 1 pixel at a time in a sub-scanning direction; a detector that inputs and holds each pixel of the image data, compares each pixel value of the held plurality of pixels, and detects an edge of a front end or a rear end of an object; a counter that counts a number of pixels that each pixel of the image data shifted in the sub-scanning direction; a distance coefficient determiner that uses a count value of the counter to determine a distance coefficient; a correction value calculator that calculates a correction value; a corrector that adds the correction value of each pixel to the pixel value of each pixel of the image data; and a position where the correction value is added is a position shifted at least N+1 pixels.