Automatic detection of targets using hierarchical spatial/temporal processing

Eliezer Haber; Yehuda Sylman

doi:10.1117/12.281401

22 September 1997 Automatic detection of targets using hierarchical spatial/temporal processing

Eliezer Haber, Yehuda Sylman

Proceedings Volume 3110, 10th Meeting on Optical Engineering in Israel; (1997) https://doi.org/10.1117/12.281401
Event: 10th Meeting on Optical Engineering in Israel, 1997, Jerusalem, Israel

Abstract

An algorithm for automatic target detection has been developed.An image is processed using a pyramidal structure in a two stage process. Each frame is spatially filtered to identify potential target areas. The temporal behavior of blobs identified is analyzed; targets are declared where blobs exhibit motion consistent with targets of interest. Targets of different sizes are detected at different pyramid levels. A single target will usually be detected at several levels. Integration of information from the various levels is essential. Two data merging approaches were examined. The first projects blobs from each level to the pyramid base. Blobs exhibiting target-like behavior are extracted. The second innovative approach builds tracks on each level. Only blobs jointed to tracks are projected to the base. Tracks are formed on these projections. Targets are declared on tracks that achieve a given quality. This hierarchical approach permits the system to operate with great sensitivity. False detections are eliminated at two stages. Most spurious blobs will not form tracks at the pyramid levels and will not be projected to the base. Tracking at the base eliminates others. Experimental results demonstrate high detection probability and low false alarm rate under a wide variety of conditions.

Citation Download Citation

Eliezer Haber and Yehuda Sylman "Automatic detection of targets using hierarchical spatial/temporal processing", Proc. SPIE 3110, 10th Meeting on Optical Engineering in Israel, (22 September 1997); https://doi.org/10.1117/12.281401

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available