Multi-binarization Image Processing for Character Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing methods, particularly in document processing, face inefficiencies in character detection due to global binarization failing to preserve all relevant information and local binarization requiring excessive computing resources and time.
Innovation Solution
A method involving binarization of an input image to create a first binary image, inverting and binarizing it to create a second binary image, detecting connected components in both, combining labels, and arranging them in a family tree with alternating layer classifications to select and remove labels based on character definition, thereby enhancing character detection efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If global binarization is used, then processing speed is improved, but character detection accuracy deteriorates
Solution Approach 1:
The patent divides the binarization process into multiple stages: first binarization creates a preliminary binary image, then connected component analysis identifies character regions, followed by second binarization focused on those regions. This segmentation allows different binarization strategies to be applied to different parts of the image, maintaining both speed and accuracy.
Solution Approach 2:
The patent performs preliminary binarization and connected component analysis before final character detection. This preliminary action identifies potential character regions that can then be processed with more accurate binarization methods, ensuring that no characters are missed while maintaining overall processing efficiency.
2Measurement precision
If local binarization is used, then character detection accuracy is improved, but computational resources and time increase significantly
Solution Approach 1:
The patent applies local binarization only to specific regions identified as potential characters, rather than to the entire image. This local quality approach ensures high detection accuracy for characters while avoiding the computational overhead of applying local binarization to all image regions, thus maintaining processing efficiency.
Solution Approach 2:
The patent performs binarization operations on only the necessary portions of the image - specifically on regions identified as potential characters through connected component analysis. This partial action avoids the excessive computational resources required for full local binarization while ensuring all characters are detected.
3Reliability
If local binarization is used, then character detection completeness is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary global binarization and connected component analysis to identify all potential character regions before applying more thorough binarization. This ensures that no characters are missed (improving completeness) while avoiding the time penalty of applying intensive binarization to the entire image from scratch.
Solution Approach 2:
The patent segments the processing into phases: initial global binarization to identify candidate regions, connected component analysis to refine region identification, and targeted second binarization on those regions. This segmentation ensures complete character detection while minimizing processing time by applying intensive methods only where needed.
Data Source
AI summary
An image encoded with character information can be created by binarizing an input image followed by connected component labeling, and then repeating the binarization and connected component labeling on an inverted version of the input image. This results in identification of connected components. Related connected components are arranged in a family tree in which successive generations of the connected components alternate between two tree layer classifications. One of the tree layer classifications is selected based on whether certain connected components define characters. A label image is created which includes labels for the connected components except for the connected components in the selected tree layer classification.


