Multi-Layered OCR Document Creation via Morphological Dilation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing OCR techniques generate a single OCR layer that fails to accurately select and copy text from documents with multiple columns and rows, often resulting in the unintended selection of undesired content.
Innovation Solution
The method involves generating a multi-layered OCR document by performing morphological dilation operations using horizontal and vertical structuring elements to create text groups, applying OCR to each group, and combining the layers for precise text selection and copying.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a single OCR layer is generated from a scanned document, then the document can be converted into editable and searchable form, but text selection from multi-column documents results in unintended selection of content from other columns
Solution Approach 1:
The patent divides the document into multiple text groups based on spatial coordinates, where each text group corresponds to a logical section (e.g., address field, date field, product description). By applying segmentation to the OCR layer, the system creates separate selectable regions that match the document's logical structure, preventing unintended selection across columns while maintaining accurate text extraction.
2Adaptability or versatility
If OCR text is formatted in a single left-to-right, top-to-bottom layer, then the OCR process is simple, but it does not work well with documents having information in multiple columns and rows
Solution Approach 1:
The patent introduces a new dimension to the OCR layer by organizing text into multiple groups with associated spatial coordinates (x1, y1, x2, y2) that define rectangular regions. This dimensional enhancement allows the system to represent multi-column and multi-row layouts accurately, enabling proper text selection in complex document structures while maintaining a manageable layer structure through coordinate-based organization.
Data Source
AI summary
The present disclosure discloses methods and systems for creating a multi-layered Optical Character Recognition (OCR) document, the multi-layered OCR document facilitates selection of the desired text from the multi-layered OCR document. The method includes receiving a scanned image corresponding to a document, the document includes text information. A binary image is generated from the scanned image. Then, a morphological dilation operation is performed to create one or more text groups, using a horizontal structuring element and a vertical structuring element. Thereafter, OCR operation is applied on each text group to generate a corresponding OCR layer. The one or more OCR layers are then combined while creating a multi-layered OCR document. Finally, the combined OCR layers are superimposed as invisible text layers over the scanned image to create the multi-layered OCR document.