Pearson's correlation and background subtraction (BGS) based approach for object's motion detection in infrared video frame sequences

The importance of infrared or thermal imaging can be seen in various applications based on Machine Vision. The real-time based application in surveillance systems like detection and tracking of moving objects has shown strong potential. According to available research, there exist some significant issues in colored video frames due to sudden environmental change or other external effects like varying backgrounds. In contrast, the thermal video frames are less affected by sudden light intensity and varying backgrounds. The suggested work utilizes a combined approach of Pearson's correlation and Background Subtraction (BGS) technique over thermal video frame sequences for detection of moving objects. The automatically generated threshold value boosts the efficacy of the algorithm in differentiating between background and foreground frames. The quantitative analysis shows the higher value of different performance parameters like Accuracy, F-measure, Recall, etc. Moreover, the qualitative analysis of obtained results strongly indicates that the proposed method outperforms when compared with the existing methods.