Character Chopper Segments Bitmap Words Using Curved Chop Lines
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional Optical Character Recognition (OCR) engines face challenges in segmenting words into individual characters due to poor image quality, font weight, and italic text, especially when images are scanned from varying orientations and illumination, leading to degraded performance and increased processing time.
Innovation Solution
An image processing apparatus with a Character Chopper component that generates and filters curved chop-lines to segment words into individual characters by considering glyph and background colors, using a fitness function that maximizes path lightness and minimizes intersections, and employs heuristics to select and refine the chop-lines for accurate separation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional OCR engines process all pixels in low-quality images, then complete image analysis is achieved, but processing time increases and performance degrades
Solution Approach 1:
The patent segments the image processing task by first identifying text lines and then dividing them into individual character candidates using chop lines. This segmentation allows the system to focus processing resources only on relevant text regions rather than analyzing all pixels in the entire image, thereby reducing processing time while maintaining recognition accuracy.
Solution Approach 2:
The patent performs preliminary actions by pre-identifying text lines and generating candidate chop lines before actual character recognition. This preliminary segmentation creates a structured framework that guides subsequent recognition processes, enabling faster and more accurate OCR by eliminating the need to process irrelevant non-text areas.
2Productivity
If simple segmentation methods are used, then processing speed is maintained, but segmentation accuracy deteriorates due to poor image quality, font weight, and italic text
Solution Approach 1:
The patent employs dynamic adaptation by adjusting segmentation strategies based on local image characteristics. The system generates multiple candidate chop lines and selects the most appropriate ones based on actual image features such as font weight, italics, and character spacing, allowing the segmentation process to adapt dynamically to varying text styles and quality conditions.
Solution Approach 2:
The patent changes segmentation parameters adaptively by considering multiple factors including glyph colors, background colors, and local image characteristics. The system adjusts chop line positions and characteristics based on these parameter variations, enabling accurate segmentation across different font weights, styles, and image qualities without sacrificing processing speed.
3Measurement precision
If multiple candidate chop lines are generated and filtered, then segmentation accuracy improves, but computational complexity increases
Solution Approach 1:
The patent segments the complex filtering task into distinct stages: first generating candidate chop lines based on image features, then filtering them using multiple criteria including glyph-background color relationships and spatial constraints. This staged segmentation of the filtering process manages computational complexity by breaking down the overall task into smaller, more manageable sub-tasks.
Solution Approach 2:
The patent applies different filtering criteria and quality standards to different regions and contexts within the image. The system considers local characteristics such as glyph colors, background colors, and local text density when evaluating candidate chop lines, allowing the algorithm to maintain high accuracy while adapting computational effort to local image complexity rather than applying uniform processing throughout.
Data Source
Figure 1~2
Figure 3a~4
Figure 5~6
AI summary
An image processing apparatus is provided that includes a character chopper component that segments words into individual characters in a bitmap of a textual image undergoing an OCR process. The Character chopper component is configured to produce a set of (possibly curved) chop-lines which divide a bitmap of any given word into its individual character or glyph candidates. Cases where an input bitmap contains two separate words are handled by marking a place where those words should be split. The character segmentation algorithm computes the set of vertically oriented, curved chop-lines by considering glyph and background colors in a given word bitmap. The set is filtered afterwards using various heuristics, in order to preserve those lines that indeed do separate a word's glyphs and minimize the number of those that do not.