Automatic Tracking System of Pedestrians Using Image Processing Method

This study is developing an automatic tracking system of pedestrians using an image processing method. First, I extract mobile objects ( pedestrians ) using motion imformation which image sequences have and label them to recognize each object. Second, I segment them form clusters using the K-mean method which is one of the cluster methods in five dimensions ( R,G,B color spaces and x,y position spaces) . Last, I track each object and get a tracking data of pedestrians. I inspect effects of this algorithm using images which are recorded on surveys.