고정 카메라 영상의 히스토그램을 이용한 주차 공간 검출

다국어 초록 (Multilingual Abstract)

As technology and industry continue to be developed, demand of vehicles increases together. Currently, there are vehicles that are close to half of population in Korea on the road now from statistics of Ministry of Land, Infrastructure and Transport. We spend a lot of time in the car and it means also for parking. In addition, much research is now underway on ’Auto Valet parking’ which means cars park by themselves. For these reasons we propose a detecting system using image histogram through fixed and non-movable sensor. This system is to detect empty several parking spaces with only low cost monocular cameras. This eliminates the need of thousands of sensor equipped in parking lots due to it detect several spaces at one scene. The composition of this paper is as follows. We explain a low-value, monocular camera system to extract parking spaces using ROI(Region of Interest). With this image, we adjust the threshold values and ranges of the video processing algorithm that uses histogram to detect a number of parking spaces.