Fast template match algorithm for spatial object detection using a stereo vision system for autonomous navigation

The spatial object detection for autonomous navigation is an active research topic with many works published about it, many of them use object detection and stereo vision systems to fulfill this task. One of the main tasks of the stereo vision system is to find the corresponding object on the image pair in order to compute the object three-dimensional coordinates, the most used methods to perform this task is to use disparity maps. The disadvantages of these methods are that they are time consuming with a high computational cost which is not ideal for real-time applications. Other methods to perform this task are template match algorithms like SAD, SSD or NCC. These algorithms have a low computational cost, but the problem is that they are time consuming, making them not suitable for real-time applications. This is why this paper proposes a novel template match algorithm called SoRA with faster run-times and low computational cost. The performed experiments demonstrate that the proposed algorithm outperforms existing template match algorithms maintaining accuracy and a low computational cost which is ideal for the intended application.