System for capturing and using expert's knowledge for image processing

• US 6,804,394 B1
• Filed: 10/11/2001
• Issued: 10/12/2004
• Est. Priority Date: 04/10/1998
• Status: Expired due to Term

• Current Assignee: Shin-Yi Hsu
• Original Assignee: Shin-Yi Hsu
• Inventors: Hsu, Shin-yi
• Primary Examiner(s): Boudreau, Leo
• Assistant Examiner(s): SHERALI, ISHRAT I
• Application Number:
  US09/975,228
• Time in Patent Office: 1,097 Days
• Field of Search: 707/1-6, 707/7-10, 707/100-102, 707/104.1, 382/173, 382/155-159, 382/113, 382/305, 382/199, 382/203, 382/162, 382/164, 382/165, 382/108, 382/224, 382/191, 382/195, 382/197, 382/201, 382/242, 382/243, 382/109, 706/55-59, 706/45-48
• US Class Current: 382/173
• CPC Class Codes: G06F 18/40 Software arrangements specially adapted for pattern recognition, e.g. user interfaces or toolboxes therefor

Abstract

The present invention relates to an apparatus and a method for object detection in an image. The apparatus for this invention includes a preprocessor, a detector, a segmentor, a classifier, a classifier systems integrator, a system output and a post processor. The method for object detection allows the user to identify an object by using three approaches: (1) a segmentation detector, (2) a pixel-based detector, and (3) a grid-cell and mesotexture based detector. All three of the aforementioned approaches allows the user to use a pseudo-English programming language in the processing system for object detection. This invention allows the user to use an expert'"'"'s knowledge and convert it to object based content retrieval algorithms. The user can preserve the segmented scenes of the original data and perform a raster-to-vector conversion which preserves the size and shape of the original objects. Further, the object based image data can be converted into geographic information system (GIS) layers which can be analyzed using standard GIS software such as ARC/Info or ARC/View.

Claims

1. A method for identifying an object in a region of an image, the steps comprising:

a) storing spectral based objects in a first library;

b) storing spatial based objects in a second library;

c) storing rule based objects in a third library; and

d) integrating said spectral based objects with said spatial based objects and with said rule based objects to increase confidence level of object identification.

2. A method for identifying an object in a region of an image in accordance with claim 1, the steps comprising:

a) providing a pixel-based classification map comprising a plurality of regions of interest; and

b) using integrating means to process said map and to generate a complex object that comprises said plurality of regions of interest, using information provided by a data expert.

3. The method for identifying an object in a region of an image in accordance with claim 2, wherein said integrating means comprises a human language program.

4. A method for identifying an object in a region of an image in accordance with claim 1, further comprising generating a geographic information system (GIS) database, the steps comprising:

a) creating a base map;

b) converting said base map into a segmented raster image, including a feature attribute table; and

c) incorporating additional data in said attribute table.

5. A method for identifying an object in a region of an image in accordance with claim 4, wherein said database is updatable and multi-media for use in an object identification and extraction program.

6. A method for identifying an object in a region of an image in accordance with claim 5, wherein said database comprises text.

7. A method for identifying an object in a region of an image in accordance with claim 5, wherein said database comprises annotated images.