Objectification of Tooth-Marked Tongue in Chinese Medicine Based on Morphological Feature Extraction

Jiaxin Yang; Dong Han; Xinming Dong; Ruixian Tang; Xiangyang Xu; Jingjing Luo; Peng Zhou

doi:10.3788/LOP202259.1117001

Journals >Laser & Optoelectronics Progress >Volume 59 >Issue 11 >Page 1117001 > Article

Laser & Optoelectronics Progress
Vol. 59, Issue 11, 1117001 (2022)

Objectification of Tooth-Marked Tongue in Chinese Medicine Based on Morphological Feature Extraction

Jiaxin Yang^1、**, Dong Han², Xinming Dong³, Ruixian Tang⁴, Xiangyang Xu⁴, Jingjing Luo^5、6, and Peng Zhou^{1、4、6、*}

Author Affiliations

¹Academy of Medical Engineering and Translational Medicine, Tianjin University, Tianjin 300072, China

²Linyi Third People’s Hospital, Linyi 276004, Shandong , China

³Tianjin Rehabilitation Center, Tianjin 300191, China

⁴School of Precision Instrument and Optoelectronics Engineering, Tianjin University, Tianjin 300072, China

⁵Academy for Engineering and Technology, Fudan University, Shanghai 200043, China

⁶Ji Hua Laboratory, Foshan 528200, Guangdong , China

show less

DOI: 10.3788/LOP202259.1117001 Cite this Article Set citation alerts

Jiaxin Yang, Dong Han, Xinming Dong, Ruixian Tang, Xiangyang Xu, Jingjing Luo, Peng Zhou. Objectification of Tooth-Marked Tongue in Chinese Medicine Based on Morphological Feature Extraction[J]. Laser & Optoelectronics Progress, 2022, 59(11): 1117001 Copy Citation Text

show less

Fig. 1. Function flow chart of tooth-marked tongue analysis algorithm

Download full size

Preprocessing process of tooth-marked tongue image. (a) Tongue image after segmentation; (b) smallest rectangular containing tongue; (c) tooth mark part extraction image; (d) tooth mark region binarization; (e) image after opening and closing operation; (f) image after contour filling

Fig. 2. Preprocessing process of tooth-marked tongue image. (a) Tongue image after segmentation; (b) smallest rectangular containing tongue; (c) tooth mark part extraction image; (d) tooth mark region binarization; (e) image after opening and closing operation; (f) image after contour filling

Download full size

Fig. 3. Simplified schematic diagram of DP algorithm implementation process

Download full size

Fig. 4. Establishment process of tooth mark thickness array of scallop tongue edge

Download full size

Fig. 5. Compression effect of DP algorithm and VDLS-DP algorithm on original data. (a)(b) Severe tooth-marked tongue; (c)(d) moderate tooth-marked tongue; (e)(f) mild tooth-marked tongue

Download full size

Item	Tooth-marked tongue	Non tooth- marked tongue	Total
Correct	135	89	224
Incorrect	28	25	53
Accuracy /%	82.82	78.07	80.86

Table 1. Performance of tooth marks recognition

Algorithm type	$R_{s}$ /%
DP	11.07
VDLS-DP	8.65

Table 2. Relative area deviation of DP algorithm and VDLS-DP algorithm

Algorithm type	Incorrect	Total	Accuracy /%
Ref.［6］	71	135	45.19
DP	41		69.63
VDLS-DP	27		80.00

Table 3. Performance of tooth marks number detection

Tooth mark degree	Amount	Incorrect			Accuracy /%
Tooth mark degree	Amount	Error type	Amount		Accuracy /%
Mild	93	Mild→moderate	2	5	94.62
Mild	93	Mild→severe	3	5	94.62
Moderate	27	Moderate→mild	4	7	74.07
Moderate	27	Moderate→severe	3	7	74.07
Severe	15	Severe→mild	0	2	86.67
Severe	15	Severe→moderate	2	2	86.67
Total	135	14			89.63

Table 4. Test results of tooth marks

Download Citation

Set citation alerts for the article

Tools

Set citation alerts for the article

Save the article for my favorites

Paper Information