口腔医学研究

口腔医学研究 ›› 2019, Vol. 35 ›› Issue (2): 163-167.DOI: 10.13701/j.cnki.kqyxyj.2019.02.015

• 口腔影像技术研究 • 上一篇    下一篇

腭皱法医学同一认定中倾斜图像矫正方法的研究

贾宇1,吉新新2,张雄2,李冰1*,上官宏2*   

  1. 1. 山西医科大学口腔医学院·口腔医院 山西 太原 030001;
    2. 太原科技大学电子信息工程学院 山西 太原 030024
  • 收稿日期:2018-08-06 出版日期:2019-02-18 发布日期:2019-02-25
  • 通讯作者: 李冰,E-mail: libing-1975@163.com 上官宏,E-mail:shangguan_hong01@163.com
  • 作者简介:贾宇(1990~ ),男,山西忻州人,硕士在读,医师,主要从事口腔修复学及口腔法医学的研究。
  • 基金资助:
    山西省科技厅青年基金(编号:201701D221106) 山西省国际科技合作项目(编号:201803D421062) 太原科技大学博士启动基金(编号:20162044)

Research on the Correction of Tilted Image in Forensic Identification of Palatal Rugae

JIA Yu1,JI Xin-xin2,ZHANG Xiong2,LI Bing1*,SHANGGUAN-hong2*   

  1. 1. Stomatology Hospital, Affiliated to Shanxi Medical University,Taiyuan 030001, China;
    2. School of Electronic Information Engineering, Taiyuan University of Science and Technology,Taiyuan 030024, China.
  • Received:2018-08-06 Online:2019-02-18 Published:2019-02-25

PDF (PC)

288

摘要： 目的: 对腭皱法医学同一认定数字化系统中的倾斜图像进行矫正并对其进行效果评价。方法: 设定腭皱标准图像的判断准则及特征点的选取规则,依据规则选择特征点进行标定并拟合出两条直线,求其方向向量;最后由两方向向量的夹角确定目标图像的旋转角度并进行旋转,以获得角度矫正图像,并对其从不同角度采集的腭皱图像进行仿真测试。结果: 对旋转不同角度的图像进行多次角度矫正后,从视觉效果上来看,矫正结果比较理想,与标准图像的角度差别基本可以忽略,进一步表明了该算法的稳定性。结论: 该算法能够很好的对目标图像进行角度矫正以获得标准图像,具有较强的鲁棒性。

关键词: 法医齿科学, 腭皱, 同一认定, 方向向量, 图像矫正

Abstract: To correct the tilted image in the digital system for forensic identification of the palatal rugae and to evaluate its effect. Methods: Firstly, the criteria for standard palatal rugae image and the selection rules for feature points were set. Secondly, feature points according to the rules were characterized, and two lines were fitted and found their direction vector. Finally, to obtain the corrected images, the tilted images were rotated by the angle determined by the two direction vectors. Results: From the visual effect, the correction results were ideal, and the difference from the angle of the standard image could be ignored, and the stability of the algorithm was further demonstrated. Conclusion: The algorithm can correct the target image very well to get standard images and has strong robustness.

Key words: Forensic odontology, Palatal rugae, Forensic identification, Direction vector, Image correction