Row and column feature detection in binary images

Details Row and column feature detection in binary images Row and column feature detection in binary images

2000-01-07

2003-03-04

Tran, Phuoc (Department: 2621)

Image analysis

Image segmentation

C382S190000, C382S149000

Reexamination Certificate

active

06529628

ABSTRACT:

FIELD OF THE INVENTION
This invention pertains to computer image processing and more particularly to identification and labeling of features in a binary image.
BACKGROUND OF THE INVENTION
Computer image processing is a large and growing field. The ability to process images to isolate features of practical interest is an important recognition tool. Feature extraction from images has a wide application in Computer Vision, Robotics, and other areas. Images are typically drawn using sensory devices such as Charge-Coupled Devices (CCDs), or range, proximity, and touch sensors. Feature extraction on these images is often divided into three components, namely: preprocessing, feature extraction, and feature detection. Problems associated with this type of analysis include object recognition in a three dimensional scene, character recognition, and geometrical characteristics in a given binary shape obtained using touch sensors such as center, and orientation angles.
Feature extraction is also applicable to other areas, such as silicon manufacturing. Memory devices in microprocessor products form a rectangular grid of memory cells, in which any given cell can fail. After testing such a device, information is extracted about the pass/fail status of each cell, yielding a binary image where a “one” represents a failed status, and a “zero” represents a pass status. The type of defects affecting memory cells exhibit themselves in full or partial rows or columns and in bit signatures such as single, double, or multiple bits. When looking at an input image, these types of defects have a very distinct shape. Morphology or image labeling techniques can be extended to isolate these features in the image.
In the past, heuristics were used to identify defects in the memory module from features of the binary image. For example, larger defects were assumed to be a collection of many adjacent smaller defects. Heuristics are difficult to implement and expensive to execute. Where a binary image has many small objects, such as a transformed memory map, using heuristics is impractical.
SUMMARY OF THE INVENTION
The invention is a method and system for raster reduction of a binary image. The binary image includes a number of features. A software program decomposes the binary image separately into row- and column-oriented rectangles. A subset of the row-oriented rectangles and column-oriented rectangles are then selected that cover the features of the binary image.


REFERENCES:
patent: 5153444 (1992-10-01), Maeda et al.
patent: 5793932 (1998-08-01), Kuratomi et al.
patent: 5999647 (1999-12-01), Nakao et al.
patent: 6014450 (2000-01-01), Heilper et al.
Berthold Klaus Paul Horn, “Binary Images: Topological Properties”, Robot Vision, The MIT Press, 1986, pp. 65-85.

Affiliated with

Also associated with

Rating

Row and column feature detection in binary images does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Row and column feature detection in binary images, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Row and column feature detection in binary images will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3009633