Handwritten Character Segmentation via Sweeping Shape Parameters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Handwritten character recognition faces challenges in character segmentation due to factors like character intersection in cursive handwriting, which degrades the recognition process.
Innovation Solution
A method and system for determining handwritten character segmentation shape parameters by sweeping images with predefined shapes and parameters, such as line slopes, to segment and recognize characters, with the ability to store and reuse these parameters for improved accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional character segmentation methods are used, then the ICR process can be performed, but character segmentation accuracy degrades due to character intersection in cursive handwriting
Solution Approach 1:
The patent applies segmentation by dividing the handwritten image into multiple candidate character segments using horizontal slicing at detected baseline positions. This creates discrete character regions that can be individually processed, effectively separating intersecting characters in cursive handwriting through vertical cutting at identified separation points.
Solution Approach 2:
The patent introduces baseline detection as an intermediary step between image input and character recognition. By first identifying baseline positions and then using these as intermediaries to guide vertical cutting operations, the system creates a mediating structure that facilitates accurate segmentation of intersecting characters.
2Measurement precision
If multiple parameters are tested for character segmentation, then segmentation accuracy can be determined, but the processing time and complexity increase
Solution Approach 1:
The patent performs preliminary baseline detection and candidate segment identification before final character segmentation. By pre-identifying baseline positions and potential cutting points, the system prepares the data structure in advance, allowing for more efficient processing when multiple parameters need to be evaluated for optimal segmentation accuracy.
Solution Approach 2:
The patent employs dynamic parameter adjustment by testing multiple baseline positions and cutting points, then selecting the combination that yields optimal segmentation accuracy. This dynamic approach allows the system to adapt parameters based on the specific characteristics of each handwritten sample rather than using fixed predetermined values.
Data Source
AI summary
A system and method for determining handwritten character segmentation shape parameters for a user in automated handwriting recognition by prompting the user for a training sample; obtaining an image that includes handwritten text that corresponds to the training sample; sweeping the image with shapes corresponding to parameters to determine coordinates of the shapes in the image; segmenting the image into segmented characters based on the coordinates of the shapes; determining character segmentation accuracies of the parameters; and storing an association between the user and the parameters. The system and method can further include receiving a writing sample from the same user and utilizing the stored parameters to segment characters in the writing sample for use in automated handwriting recognition of the writing sample.