Likelihood-based threshold selection for imaging target trackers

Details Likelihood-based threshold selection for imaging target trackers Likelihood-based threshold selection for imaging target trackers

1995-10-11

1999-03-02

Couso, Yon J.

Image analysis

Histogram processing

For setting a threshold

382270, G06K 946

Patent

active

058781633

ABSTRACT:
An imaging target tracker and a method for determining thresholds that are used to optimally distinguish a target from background. The present invention is an imaging target tracker that comprises a likelihood-based threshold selection method that uses knowledge of the location of a target to build up temporally-smoothed grey-level distribution or likelihood maps. Inner and outer histograms are formed that comprise inner and outer regions of an image, respectively, using inner, buffer and outer gates that segment the image. The inner and outer histograms are then recursively smoothed. A lookup table of actually observed pixel values is formed. A target likelihood value for actually observed pixel values is then computed. Legal start and end points of histogram segments are identified. Valid histogram segments are identified. The longest histogram segment is identified. Finally, two optimal thresholds are selected 8 and are set at respective ends of histogram segments. The likelihood maps adapt over time to the signature of the target. The grey-level distribution of the target is used to select thresholds that pass a band of grey levels whose likelihood that they belong to the target is high.

REFERENCES:
patent: 4364089 (1982-12-01), Woolfson
patent: 4958224 (1990-09-01), Lepore et al.
patent: 5138671 (1992-08-01), Yokoyama
patent: 5341142 (1994-08-01), Reis et al.
patent: 5444556 (1995-08-01), Ito et al.
C. J. Samwell and G. A. Cain, "The Bae (Bracknell) Automatic Detection, Tracking and Classification Systems", in Conf. Proc. Advances In Guidance And Control Systems and Technology, London, 7-10 Oct. 1986, pp. 17/1-10.

Affiliated with

Also associated with

Rating

Likelihood-based threshold selection for imaging target trackers does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Likelihood-based threshold selection for imaging target trackers, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Likelihood-based threshold selection for imaging target trackers will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-430134