Visual tracking system and method thereof

摘要

The present invention provides a visual tracking system and its method comprising: a sensor unit, for capturing monitored scenes continuously; an image processor unit, for detecting when a target enters into a monitored scene, and extracting its characteristics to establish at least one model, and calculating the matching scores of the models; a hybrid tracking algorithm unit, for combining the matching scores to produce optimal matching results; a visual probability data association filter, for receiving the optimal matching results to eliminate the interference and output a tracking signal; an active moving platform, for driving the platform according to the tracking signal to situate the target at the center of the image. Therefore, the visual tracking system of the present invention can help a security camera system to record the target in details and maximize the visual information of the intruding target.

主权项

1. A visual tracking system, for generating a tracking result signal of a target, allowing an active moving platform to move according to the tracking result signal, and thereby positioning the target at an image center, the visual tracking system comprising:
a sensor unit for capturing images of a monitored scene consecutively; an image processor unit, coupled to the sensor unit, for detecting whether a target is entering the image, extracting characteristics from the image to formulate at least two models of the target, and calculating a matching score of the models; a hybrid tracking algorithm unit, coupled to the image processor unit, for using a fuzzy theory to create a membership function for the matching score of the models, and using the membership function for the fuzzy computation to determine the levels of excellent, above average, average, below average or poor, and performing a defuzzification for the fuzzy computation to obtain the final matching scores, and computing the probability of the target position for each pixel in the monitored scene to produce an optimal matching result; and a visual probability data association filter, coupled to the hybrid tracking algorithm unit, for receiving the optimal matching results to remove an interference of tracking from a complicated background and output the a tracking result signal, wherein the membership function is a triangular membership function (trimf), and the fuzziness degrees are referred to as excellent, above average, average, below average, and poor, respectively.