Edge-color-histogram and Kalman filter-based real-time object tracking

An object tracking algorithm based on Kalman filter and edge orientation histogram(EOH) was used to improve the tracking performance with scene illumination variations or complicated backgrounds.Since the EOH is not sensitive to light or color variation,this method works well when tracking object in such cases.The Kalman filter is used to estimate the new position and angle and then a new estimate of the EOH is calculated to quickly and efficiently match the target.Tests show that the color histogram method easily losts objects while the EOH follows them very well.The tests show that the EOH and the Kalman filter can track objects in real-time when the illumination varies or background clutter is present.