Browse Prior Art Database

DYNAMIC REGION OF INTEREST (ROI) PLACEMENT ALGORITHM FOR LOCATING CHARACTERS IN AN IMAGE

IP.com Disclosure Number: IPCOM000008098D
Original Publication Date: 1997-Mar-01
Included in the Prior Art Database: 2002-May-17
Document File: 2 page(s) / 78K

Publishing Venue

Motorola

Related People

Anil D. Mandalia: AUTHOR

Abstract

In machine vision applications, an algorithm is applied within a region in an image (512 by 480 pixels) so that it operates on pixels within this region to obtain an output. This region within the image is called REGION OF INTEREST or ROI. The ROI provides a bounding box within which all of the operations on pixels are performed. The ROI can be specified as the entire or partial image. If the ROI is partial image, the computational time is fast due to less number of pixels. The ROI bounding box is represented by starting row position (n), starting column position (y), height (high) and width (wide).

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 63% of the total text.

Page 1 of 2

MOTOROLA Technical Developments

DYNAMIC REGION OF INTEREST (ROI) PLACEMENT ALGORITHM FOR LOCATING CHARACTERS IN AN IMAGE

by Anil D. Mandalia

BACKGROUND

  In machine vision applications, an algorithm is applied within a region in an image (512 by 480 pixels) so that it operates on pixels within this region to obtain an output. This region within the image is called REGION OF INTEREST or ROI. The ROI provides a bounding box within which all of the operations on pixels are performed. The ROI can be specified as the entire or partial image. If the ROI is partial image, the computational time is fast due to less number of pixels. The ROI bounding box is represented by starting row position (n), starting column position (y), height (high) and width (wide).

  For an application, such as Optical Character Recognition (OCR) where characters have to be found in an image (shown in figure l), the require- ment is to locate the characters instantly. If the entire image is chosen as ROI (512x480), then the algorithm for finding the characters would take longer. But If the ROI was defined in such a way that it is enveloped around the characters, then the computational time would be less. Since the charac- ters in the image can shift up or down or sideways, the ROI can be calculated dynamically to keep the

n' = .m.,, - X-Deviation Y' = yCnOr

     - Y-Deviation high= constant
wide= constant

where .rrhwr is row character position, ych, is column character position, X-Deviation and Y-Deviat...