Sledování pohybu objektů v obrazovém signálu

Abstract

Tato diplomova prace se zameruje na sledovani pohybu vice objektu. Prace popisuje dve implementace filtru, ktere jsou v podstate zalozeny na principu Kalmanova filtru. Obe implementace jsou zalozeny na principu sledovani vice objektu, na zaklade znalosti pozic vsech objektu v kazdem snimku. Prvni implementace je smisena verze Globalniho a Standardniho filtru nejblizsich sousedu. Druha implementace je postavena na pravde- podobnostnim pristupu k procesu sdruzeni. Posledni kapitola poskytuje srovnani mezi temito filtry a Zakladnim filtrem. Algoritmy byly realizovany v jave.This thesis focuses on multiple object tracking in clutter environment. The paper describes two implementations of filters in detail, that are essentially based on the Kalman Filter. Both implementations are based on tracking multiple objects with the knowledge of their positions in every frame. The first implementation is a hybrid of the Global and the Standard Nearest Neighbour Filters. The second implementation takes on a probabilistic approach to the data association process. The last chapter provides a comparison between these trackers and also between the Basic filter. The trackers were implemented in java.