A New Head Location Method in TMS Treatment
Qingpei Zhang, Xin Wang, Zecai Lin · 2018
In the intelligent and automatic TMS treatment system, head location is a difficult problem. In this paper, we present a new method for head location based on Kinect2. First, we used the HOG feature descriptor and SVM algorithm to detect the face in the Kinect2 RGB image, and we determined the pixel position of the facial landmarks using the cascade regression algorithm. Then we used the depth image of Kinect2 to obtain the 3D point cloud of the feature points, we determined the head position by calculating the spatial position of the feature point cloud. And finally, we verified the proposed method through experiments. These experimental results showed the method is reliable and the accuracy is high and the method can be used in the TMS treatment.