Advanced Ear Detection Algorithm Using Faster R-CNN, Refocus Filters, and the Gradient Map

Generally, the ear is a distinctive feature in face images. Many researchers carrying out the studies on face image processing attempt to extract the features of ears to improve the performances of face detection and recognition. In this paper, we propose a novel and efficient algorithm to detect ears from profile images automatically. The proposed algorithm is based on the techniques of the convolutional neural network (CNN), the object refocus filter, and the gradient map to avoid illumination variation and make the features of ears more prominent. We perform several simulations on well-known datasets and show that the proposed algorithm methods can achieve even higher detection rates than state-of-the-art methods.