earticle

영어

Various algorithms based on image processing techniques have been used to detect and track objects in video surveillance systems. In this paper, followed by color RGB for multiple objects is present. To do this, real-time tracking algorithm is proposed for objects navigate and identify current positions and movements, because each object has a color with label on top. RGB color space is used to calculate the threshold values of each color. It is for example, video surveillance applications where cameras and other sources of information are used to monitor the activities of a sensitive site. Threshold values can be defined according to the object to track. Linear model is applied to the color discrimination. To remove unwanted objects to executives, morphological transformation is applied on the image binary and technical Blob analysis is then used to facilitate the detection of object. Our proposed algorithm is capable of tracking objects in the format full-frame (640 x 480 pixels) at rate (50 images per second). High performance robustness and in real time is confirmed by experiments.