Document Processing System Using Text Row Structure Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for processing documents, such as optical character recognition and image registration, are time-consuming and processor-intensive, requiring significant storage and processing power, and often need prior knowledge of data location, making them ineffective for efficient document processing.
Innovation Solution
The system analyzes the physical structure of text rows in a document image by creating character blocks, determining their alignments, and classifying text rows based on these structures, using pattern matching to group similar rows and interpolate vector data for efficient document processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If optical character recognition is performed on the entire page, then text identification is achieved, but processing time and processor intensity increase significantly
Solution Approach 1:
The patent divides the document processing into distinct segments: first analyzing the physical structure of text rows and character blocks to identify document layout patterns, then using these patterns to guide selective OCR processing only on relevant regions. This segmentation avoids applying OCR to the entire page, thereby reducing processing time while maintaining text identification accuracy.
Solution Approach 2:
The patent performs preliminary analysis of the document's physical structure, including text row positions, character block alignments, and column formations, before conducting OCR. This preliminary structuring creates a framework that enables subsequent targeted text extraction without requiring full-page OCR, thus reducing processing time while preserving identification accuracy.
2Measurement precision
If image registration is used to compare documents, then document differences are identified, but storage requirements and processing power increase significantly
Solution Approach 1:
The patent extracts only the essential structural features from document images, such as text row positions, character block alignments, and column patterns, rather than storing and processing complete document images. This extraction approach enables document comparison and difference identification while dramatically reducing storage requirements compared to traditional image registration methods.
Solution Approach 2:
The patent creates simplified structural representations (copies) of document layouts using binary data to represent text row positions and character block alignments. These lightweight structural copies enable efficient document comparison without requiring storage of large image files, thus reducing storage requirements while maintaining comparison accuracy.
3Productivity
If conventional document processing methods are used, then document analysis is performed, but prior knowledge of data location is required
Solution Approach 1:
The patent enables the processing system to automatically analyze and determine the physical structure of documents, including text row positions and character block alignments, without requiring external input about data locations. The system self-generates the structural framework needed for processing, eliminating the need for prior knowledge of data locations while maintaining processing efficiency.
Solution Approach 2:
The patent performs preliminary automatic analysis of document physical structure, creating a framework of text rows and character blocks that identifies data locations without requiring prior knowledge. This preliminary structuring action enables subsequent processing steps to proceed efficiently with automatically determined data locations, reducing system complexity.
Data Source
AI summary
Systems and methods analyze the physical structure of text rows in a document image, including the positions of one or more alignments of one or more character blocks in one or more text rows of the document image. The systems and methods determine one or more groups of text rows that are placed into a class based on the structures of the text rows, such as the positions of the one or more alignments of the one or more character blocks in each text row. A pattern matching system then determines if one or more classes should be further combined into a combined class.