An algorithm for detection of moving objects in video streams from the monitoring cameras is presented. A system composed of a standard video camera and a thermal camera, mounted in close proximity to each other, is used for object detection. First, a background subtraction is performed in both video streams separately, using the popular Gaussian Mixture Models method. For the next processing stage, the authors propose an algorithm which synchronizes the video streams and performs a projective transformation of the images so that they are properly aligned. Finally, the algorithm processes the partial background subtraction results from both cameras in order to obtain a combined result, from which connected components representing moving objects may be extracted. The tests of the proposed algorithm confirm that employing the dual camera system for moving object detection improves its accuracy in difficult lighting conditions.