Detecting small, moving objects in image sequences using sequential hypothesis testing

Abstract

A new algorithm is proposed for the solution of an important class of multidimensional detection problems: the detection of small, barely discernible, moving objects of unknown position and velocity in a sequence of digital images. A large number of candidate trajectories, orgavized into a tree structure, are hypothesized at each pixel io the sequence and tested sequentially for a shift in mean intensity. The practicality of the algorithm is facilitated by the use of multistage bypothesis testing (MHT) for simultaneous inference, as well as the existence of exact expressions for MHT test performance in Gaussian white noise (GWN). These expressions predict the algorithm's computation and memory requirements, where it is shown theoretically that several orders of magnitude of processing are saved over a brute-force approach based on fixed sample-size tests. The algorithm is applied to real data by using a robust preprocessing procedure to eliminate background structure and transform the image sequence into a residual representation, modeled as GWN. Results are verified experimentally on a variety of video image sequences.

Into the woods: visual surveillance of noncooperative and camouflaged targets in complex outdoor settings

“…Other related research has worked on developing target motion estimators, e.g. [22,23,24,25]. These, however, are focused more on estimating models (usually smooth, periodic or planar models) of the moving targets.…”

Analgorithmic Framework for Automatic Detection and Tracking Moving Point Targets in IR Image Sequences

“…Temporal filtering schemes 8,9 use information from a set of previous frames for estimating the background information at PUT. The temporal filtering schemes can provide the better estimate for the background due to high correlation between successively sampled images.…”

A Survey of Motion-Based Multitarget Tracking Methods

“…(2) Multistage Hypothesis Testing Blostein and Huang [163] proposed a multistage hypothesis testing algorithm (MSHT) for target tacking and detection. The MSHT algorithm uses a truncated sequential probability ratio test (TSPRT), to efficiently evaluate a dense tree of linear, constant velocity candidate trajectory segments [164].…”