USA flag logo/image

An Official Website of the United States Government

Design and Development of a Video Database With Automated Indexing Mechanisms

Award Information

Agency:
Department of Defense
Branch:
Air Force
Award ID:
40831
Program Year/Program:
1998 / SBIR
Agency Tracking Number:
40831
Solicitation Year:
N/A
Solicitation Topic Code:
N/A
Solicitation Number:
N/A
Small Business Information
Azima Advanced Technologies,
3409 Morgan Street West Lafayette, IN 47906
View profile »
Woman-Owned: Yes
Minority-Owned: Yes
HUBZone-Owned: No
 
Phase 1
Fiscal Year: 1998
Title: Design and Development of a Video Database With Automated Indexing Mechanisms
Agency / Branch: DOD / USAF
Contract: N/A
Award Amount: $99,966.00
 

Abstract:

We propose to develop automated video indexing and retrieval mechanisms based on comprehensive representations of video data, and efficient processing for automated in-dexing. Content-based video data retrieval has enormous potential in defense, industry and the government. However, a number of stubborn problems have impeded progress in developing fully automated video database systems. These problems include difficulty of specification of video queries and huge computational requirements. Our effort will focus on developing viable solutions for these two problems and utilizing the solutions in building an automated video database indexing and management system. We plan to employ wavelet-based multiresolution pattern matching techniques for recognition of video objects. The precision of object modeling and recognition can be tuned to the level of available computational resources and time constraints thus allowing trade-offs between levels of computation and accuracy of pattern matching. We will develop pragmatic knowledge-based models of video data to capture complex video semantics. These models will provide efficient Indexing and retrieval mechanisms for a variety of scenarios. Two models, based on object motion trajectories and spatio-temporal logic will be developed for this purpose. The system will provide a graphic user-interface for the users to formulated complex video queries.

Principal Investigator:

M. Farrukh Khan
7654973330

Business Contact:


0
Small Business Information at Submission:

Azima Advanced Technologies,
3409 Morgan Street West Lafayette, IN 47906

EIN/Tax ID:
DUNS: N/A
Number of Employees:
Woman-Owned: No
Minority-Owned: No
HUBZone-Owned: No