Face recognition: using neural spatial-temporal segmentation to extract face features and perform feature geometry comparison

This research uses boundary cutting, spatial-temporal segmentation, block based searching, and gray scale histogram techniques to extract the eyes, nose, and mouth images. In this research, the gray scale histogram is used to extract the eye, nose, and mouth features. A spatial-temporal template is designed to segment the eye, nose, and mouth images to extract eye, nose, and mouth images. In order to obtain better results, the generic algorithm and spatial region partition technique are used to remove noise and precisely bind the object region to obtain a more accurate extraction result.