&lt;title&gt;Real-time hardware implementation for area correlation tracker&lt;/title&gt;

Venkateswarlu, Ronda; Rao, B. Venkateswara; Raju, P. V. S.

doi:10.1117/12.138173

Cited by 4 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The reference area designated from the initial distance and look-angle completely differs as the image sensor approaches or recedes from the target area. Hence the reference is updated with the new image reference for the next frame and the old reference is discarded (Venkateswarlu et al 1992).…”

Section: Area Correlation Algorithm (Aca)mentioning

confidence: 99%

“…Area correlation consists of overlapping a reference image over a search image and computing a similarity function. The window within the search image having maximum similarity with the reference is taken to be the target's location (Kumar et al 1987;Venkateswarlu et al 1992). A number of image registration algorithms and schemes are reported in the literature (Anuta 1970;Barnea & Silverman 1972;Pratt 1974).…”

Section: Introductionmentioning

confidence: 99%

“…A periodic updating of the reference target-image circumvents these problems. Otherwise, the image-tracking algorithm may become divergent with time, thereby losing the track (Venkateswarlu et al 1992).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Robust correlation tracker

Kishore

Rao

2001

Sadhana

View full text Add to dashboard Cite

Correlation tracking plays an important role in the automation of weapon systems. Area correlation is an effective technique for tracking targets that have neither prominent features nor high contrast with the background and the 'target' can even be an area or a scene of interest. Even though this technique is robust under varying conditions of target background and light conditions, it has some problems like target drift and false registration. When the tracker or target is moving, the registration point drifts due to the discrete pixel size and aspect angle change. In this research work, an attempt has been made to improve the performance of a correlation tracker for tracking ground targets with very poor contrast. In the present work only the CCD visible images with very poor target to background contrast are considered. Applying novel linear and nonlinear filters, the problems present in the correlation tracker are overcome. Confidence and redundancy measures have been proposed to improve the performance by detecting misregistration. The proposed algorithm is tested on different sequences of images and its performance is satisfactory.

show abstract

Section: Area Correlation Algorithm (Aca)mentioning

confidence: 99%