Autofocus window selection algorithm based on saliency detection

Focus window selection is a very important step in the process of Auto-Focusing(AF). This paper proposes a new method for the selection of focus window, where a fast AF window selection algorithm based on image saliency region extraction is used to cut down the computation time and overcome the disturbance of the background in the automatic focusing system. Firstly, the salient object detection method based on the Minimum Barrier Plus(MB+) Transform algorithm is utilized to calculate the salient regions of the image in order to obtain a feature map. Secondly, a threshold method is used to de-noise the feature map. Then, correlation treatment method and boundary expansion method are used to build the focus window, of which the size and position are self-adaptive with the target. To the end, in this study, a comparison is made between the commonly used algorithm and the introduced window selection algorithm based on the improved MB + saliency detection in terms of accuracy and computation time. The result obtained indicates that our algorithm has better performance in highlighting the potential focus targets. And its better accuracy and less computation time make it suitable for tasks in general scenes and complex backgrounds.