Text vectorization using ocr and stroke structure modeling
AssigneeXerox Corporation
InventorsZhigang Fan, Francis K. Tse, Bingfeng Zhou, Yadong Mu, Tao He
Life of this patent
Estimating the term…
Abstract
Systems and methods are described that facilitate dominant point detection for text in a scanned document. The dominant points are classified as “major” (e.g., structural) and “minor” (e.g., serif). A set of rules or parameters for each character is determined off-line. During the text vectorization, OCR is performed and the rules (parameters) associated with the recognized character are selected. Both major and minor dominant points are detected as a maximization process with the parameter set. For minor dominant points, additional processes are optionally employed.
Claims14
A method of detecting dominant points on text characters in a scanned document, comprising:
- executing an optical character recognition (OCR) algorithm that recognizes a text character in the scanned document;
- performing a table lookup that identifies one or more dominant point detection rules for detecting the dominant points;
- applying the one or more identified dominant point detection rules to the text character to detect dominant points on the text character;
- fitting one or more curves to each text character outline between neighboring dominant points to convert the text character from a bitmap to a vector representation using the detected dominant points;
- and
- storing the vectorized text character to memory for use in a vector representation of the scanned document.
A vectorization system for scanned document, comprising:
- a memory that stores computer-executable instructions for:
- performing an optical character recognition (OCR) algorithm that detects a text character in the scanned document;
- performing a table lookup of a rule database that identifies one or more dominant point detection rules for detecting dominant pointr;
- applying the one or more identified dominant point detection rules to the text character to detect dominant points on the text character;
- fitting one or more curves to each text character outline between neighboring dominant points using the detected dominant points;
- and
- storing a vectorized text character for use in a vector representation of the scanned document;
- and a processor that executes the computer-executable instructions.
An apparatus for vectorization on text characters in a scanned document, comprising:
- means for executing an optical character recognition (OCR) algorithm that recognizes a text character in the scanned document;
- means for performing a table lookup that identifies one or more dominant point detection rules for detecting dominant points;
- means for applying the one or more identified dominant point detection rules to the text character to detect major and minor dominant points on the text character;
- means for fitting one or more curves to each text character outline between neighboring dominant points using the detected dominant points;
- and
- means for storing the vectorized text character to memory for use in a vector representation of the scanned document.
Dependencies are inferred from the claim text.