[1]YU Miao,YAN Zifei,WANG Kuan-quan,et al.An algor ithm for extraction of near-infrared sublingual ve ins[J].CAAI Transactions on Intelligent Systems,2008,3(4):309-312.
Copy
CAAI Transactions on Intelligent Systems[ISSN 1673-4785/CN 23-1538/TP] Volume:
3
Number of periods:
2008 4
Page number:
309-312
Column:
学术论文—机器感知与模式识别
Public date:
2008-08-25
- Title:
-
An algor ithm for extraction of near-infrared sublingual ve ins
- Author(s):
-
YU Miao1 ; YAN Zi2fei1 ; WANG Kuan-quan1 ; L INai-min2
-
1. School of Computer Science and Technology, Harbin Institute of Technology, Harbin 150001, China;
2. The PLA No. 211 Hosp i2 tal, Harbin 150080, China
-
- Keywords:
-
tongue diagnosis; near infrared sublingual vein image; watershed; radial p rojection; regional growth
- CLC:
-
TP311
- DOI:
-
-
- Abstract:
-
The condition of the sublingual vein is one of the most important diagnostic criteria in tongue diagnosis. A segmentation method that allows better identification of sublingual veins from near2infrared sublingual images is p roposed in this paper. The radial p rojection method, based on the watershed algorithm, is first app lied to acquire the entire contour of the sublingual region, removing the interference of useless background surrounding the tongue body. Dynamic threshold is then utilized to comp lete the binarization of the newly defined sublingual region, and further confirm the candidate sublingual vein regions by considering intensity features of the near infrared sublingual vein image. Finally, region growing is iteratively performed, restricted by the original intensity feature of the near infrared sublingual image, to trace the contour of sublingual veins. Experimental results revealed that the p roposed method effectively isolates the sublingual veins and the surrounding tongue p roper, and p recisely traces the bounda2 ry of the sublingual veins.