Character Recognition Region Partitioning for Abnormal Layouts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current information processing systems face challenges in accurately recognizing characters within images due to incorrect partitioning of regions, leading to erroneous character recognition, especially when characters have abnormal sizes or uncommon arrangements.
Innovation Solution
An information processing apparatus is developed with a character recognition unit and a partitioning unit that performs layout analysis, optical character recognition (OCR), and detects out-of-range value characters by calculating size distributions, and re-partitions regions to improve recognition accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single partitioning method is used for the entire image, then the processing is simple and fast, but character recognition accuracy deteriorates when characters have abnormal sizes or uncommon arrangements
Solution Approach 1:
The patent applies segmentation by dividing the image into multiple regions and performing partitioning separately for each region rather than applying a single partitioning method to the entire image. This allows the system to adapt to local variations in character size and arrangement, improving recognition accuracy for regions with abnormal characters while maintaining simplicity in normal regions.
Solution Approach 2:
The patent implements dynamic partitioning by adjusting partitioning parameters based on the specific characteristics of each region. The system dynamically selects partitioning methods and parameters according to the detected character distribution, size variations, and arrangement patterns in different regions, enabling adaptive optimization without requiring a completely complex fixed system.
2Measurement precision
If the entire image is re-processed to improve recognition accuracy, then recognition accuracy improves, but processing time increases significantly
Solution Approach 1:
The patent extracts and identifies only the specific regions containing abnormal characters through size distribution analysis and partitioning evaluation. Instead of re-processing the entire image, the system isolates and re-processes only the problematic regions, significantly reducing the time cost while maintaining improved recognition accuracy for the affected areas.
Solution Approach 2:
The patent applies partial action by performing re-partitioning and re-processing only on specific regions where character recognition failures are detected, rather than applying the same operation to the entire image. This selective approach reduces computational overhead and processing time while still achieving the goal of improving overall recognition accuracy.
3Measurement precision
If partitioning parameters are fixed for all regions, then the processing is efficient and simple, but recognition accuracy deteriorates for regions with varying character characteristics
Solution Approach 1:
The patent applies local quality by using different partitioning parameters and methods for different regions based on their specific characteristics. The system analyzes character size distributions and arrangement patterns in each region and adjusts partitioning parameters locally to match the regional characteristics, thereby improving recognition accuracy without requiring complete re-processing of the entire image.
Solution Approach 2:
The patent implements parameter changes by dynamically adjusting partitioning parameters such as grid size, threshold values, and partitioning methods based on the detected characteristics of each region. The system modifies these parameters according to character size variations and arrangement patterns, enabling adaptive optimization that balances accuracy improvement with processing efficiency.
Data Source
AI summary
An information processing apparatus includes a character recognition unit that performs a character recognition process to recognize a character included in an image, and a partitioning unit that partitions a specific region if a character string recognized through the character recognition process performed by the character recognition unit on a specific region included in the image includes a character or a character string, each of which satisfies a predetermined condition.