Image Processing Apparatus Background Density Distribution Noise Elimination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image processing methods fail to effectively eliminate noise from background areas in images, especially when the density distribution of the background is wide and overlaps with the density distribution of black pixels caused by entry, leading to misjudgment in character recognition.
Innovation Solution
An image processing apparatus and method that estimates the density distribution of background and designated areas, using a background density distribution estimating part and an area density distribution estimating part, and an entry existence judging part to determine the presence of predetermined information such as characters, allowing for accurate noise elimination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional noise elimination methods (averaging filter, Gaussian filter, morphology operation, median filter) are used, then noise can be reduced in simple cases, but when background density distribution is wide and overlaps with entry density, these methods cannot eliminate noise effectively and lead to misjudgment
Solution Approach 1:
The invention changes the parameter of density distribution analysis from simple threshold-based methods to statistical distribution comparison. By comparing the density distribution of a designated area against the background density distribution, the system can reliably detect entries even when their density ranges overlap with background noise, thus improving both measurement precision and reliability simultaneously
Solution Approach 2:
The invention creates a reference copy of the background density distribution from areas known to contain no entries. This background density distribution serves as a template for comparison, allowing the system to identify entries by detecting deviations from the expected background pattern, thereby maintaining high reliability in entry detection despite background noise
2Productivity
If simple threshold-based methods are used for entry detection, then processing is fast and simple, but misjudgment occurs when background noise is prominent
Solution Approach 1:
The invention transitions from single-threshold parameter detection to multi-parameter statistical distribution analysis. By evaluating density distribution characteristics rather than relying on a single threshold value, the system achieves higher measurement precision while maintaining computational efficiency through algorithmic optimization
Data Source
AI summary
An image processing apparatus of the present invention includes: a background density distribution estimating part 12 that estimates density distribution of background pixels; an area density distribution estimating part 13 that estimates density distribution in each of areas into which an input image is divided; an entry existence judging part 14 that judges the existence or not of the entry of predetermined information based on the density distribution of the background pixels and the density distribution in the relevant area; and a judgment result output part 15 that outputs a judgment result indicating whether or not there exists the entry of the predetermined information in the area being the judgment target of the entry existence judging part 14.