Moving object detection in real-time visual surveillance using background subtraction technique

In current scenario, the need of surveillance applications, technological improvements, and the numbers of active real-time surveillance systems are increasing continuously throughout the world. In this research work, we have presented a background subtraction based scheme for moving object detection in video frames. The proposed scheme has a strong potential for applications in consumer electronics and real time surveillance systems. This work is presented in three different sections, first is to model the background using initial few frames. Second is to extract moving object with a threshold and update the background. The threshold value is used to avoid unusual false detection. Third is enhancement where, we have used morphological operators in order to improve the detection quality. The employed Pixel Intensity Based Background Subtraction (PIBBS) scheme demonstrates how a system can be improved by means of a mean value controlling scheme and the incorporation of feedback based background updation scheme, showing an outranking performance in comparison with considered state-of-the-art models.