Handwriting Document Updating with Local Stroke Re-Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Handwriting recognition in computing devices is a highly intensive process, and local changes to digital documents can lead to irrelevant modifications throughout the document, making efficient and reliable updating of digital content challenging.
Innovation Solution
A computing device and method that processes handwriting input by defining update, context, and extended zones to selectively classify and segment strokes, allowing efficient updating of digital documents by overwriting or erasing content based on predefined categories.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If document content analysis is performed globally on the entire document whenever content is added, then recognition accuracy is improved, but processing time and computational intensity increase significantly
Solution Approach 1:
The patent divides the document into multiple zones (update zone, context zone, extended zone) based on the location of handwriting input. Only strokes within these specific zones are selected for classification and segmentation, rather than processing the entire document. This spatial segmentation allows the system to maintain recognition accuracy for relevant areas while dramatically reducing processing time by excluding unrelated document portions.
Solution Approach 2:
The patent applies different processing quality levels to different regions of the document. The update zone receives full classification and segmentation processing, the context zone receives partial processing for spatially or temporally related strokes, while the rest of the document remains unprocessed. This local quality approach ensures accurate recognition where needed while minimizing unnecessary computational overhead.
2Reliability
If global document analysis is performed to ensure complete content interpretation, then recognition reliability is improved, but irrelevant modifications throughout the document increase
Solution Approach 1:
By segmenting the document into update, context, and extended zones, the patent limits the scope of modifications to only those areas that are spatially or temporally related to the handwriting input. This prevents irrelevant modifications in distant or unrelated portions of the document while maintaining recognition reliability in the affected zones through proper contextual analysis.
Solution Approach 2:
The patent applies comprehensive analysis only locally to the update zone and context zone where handwriting input occurs, rather than globally across the entire document. This localized quality assurance ensures recognition reliability where it matters (near the input) while avoiding harmful irrelevant modifications in unrelated document regions.
3Stability of the object's composition
If the entire document is reprocessed to accommodate local changes, then content consistency is improved, but processing efficiency decreases
Solution Approach 1:
The patent segments the document processing into distinct zones based on the location of changes. Only the update zone (where handwriting input occurs) and context zone (spatially or temporally related strokes) are reprocessed to maintain content consistency. The extended zone includes intersecting initial blocks for boundary consistency. This segmented approach preserves content consistency in affected areas while maintaining high processing efficiency by excluding unrelated document portions.
4Measurement precision
If comprehensive stroke classification and segmentation is performed on the whole document, then recognition precision is improved, but computational complexity increases
Solution Approach 1:
The patent reduces computational complexity by segmenting the document into update, context, and extended zones. Stroke classification and segmentation are performed only on strokes within these zones rather than the entire document. The selecting module identifies strokes enclosed in the extended zone, the classifying module classifies them according to predefined categories, and the segmenting module segments the classified strokes - all within the limited scope of the zones, significantly reducing computational requirements while maintaining precision where needed.
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
The disclosure relates to a device and method for updating a digital document. The method comprises: displaying an initial content comprising strokes classified and segmented into initial blocks; defining an update zone (UZ) around a detected handwriting input interaction (IN); defining a context zone comprising the update zone (UZ) and strokes related with the strokes of the update zone; defining an extended zone (EZ) including the context zone and initial blocks intersecting with the update zone; selecting all the strokes at least partially included into the extended zone (EZ); reclassifying the selected strokes of the extended zone; re-segmenting the reclassified strokes of the extended zone; retrieving the re-segmented strokes of the update zone; retrieving the classified and segmented strokes of the initial content; and overwriting the classified and segmented strokes of the update zone with the re-segmented strokes of the update zone to generate updated content (MC).