Title: Three-dimensional image reconstruction method for human maxillofacial defect images based on block sample feature matching
Authors: Zhaofeng Li; Ling-ling Cui
Addresses: School of Information Engineering, Henan Institute of Science and Technology, Xinxiang, Henan Province, 453003, China ' The First Hospital, and Affiliated to: Jinzhou Medical University, Jinzhou 121001, China
Abstract: In traditional methods, template matching of human face image sample and edge contour segmentation is ignored, and the application is susceptible to the robustness of skin colour features and pose changes, resulting in poor accurate three-dimensional reconstruction and recognition of human maxillofacial defect images. Therefore, a three-dimensional image reconstruction method for human maxillofacial defect images based on block sample feature matching is proposed. A feature matching template of human maxillofacial images is constructed, and template matching and three-dimensional edge contour segmentation are performed to the sample images through the blocking technique. The simulation results show that this method has good feature detection and detection ability in three-dimensional reconstruction of human maxillofacial defect images and can provide high accuracy in three-dimensional reconstruction.
Keywords: human maxillofacial defect; three-dimensional reconstruction; corner detection; image.
DOI: 10.1504/IJESMS.2019.103777
International Journal of Engineering Systems Modelling and Simulation, 2019 Vol.11 No.3, pp.112 - 118
Received: 03 Jan 2019
Accepted: 09 Aug 2019
Published online: 27 Nov 2019 *