Object detection in urban environment

Abstract

This paper deals with problems of surface moving object detection in urban environmental conditions in video sequences acquired in visible spectrum using various techniques. The objects feature is changed in dependence on time and space condition such as weather condition, urban environment, etc. In this article, we proposed a combined method for object detection in video sequences using three techniques that are a template matching technique, a threshold technique and a background subtraction. The video sequences were captured using Samsung HMX - F90 with a resolution 1280× 720. However, to test the algorithm, we have resized to the resolution 160×120. The designed algorithms were tested in program Matlab.