Multi-Measure Similarity Method for Interpreting Bronze Inscriptions Based on Hu Moment and TF-KSURF

Table 1. Average recognition rate of two feature measures
View table
Table 1. Average recognition rate of two feature measures
Measure Average recognition rate /%
SURF 50.82
TF-KSURF 63.04

Table 2. Retrieval performance comparison of different methods

View table

Table 2. Retrieval performance comparison of different methods

Category	Measure in Ref.[5]		SURF		Hu moment		Proposed method
Category	R_M /%	R_NP	R_M /%	R_NP	R_M /%	R_NP	R_M /%	R_NP
Class 1	60.9	0.31	58.8	0.32	61.7	0.42	73.5	0.57
Class 2	75.6	0.52	76.5	0.47	88.2	0.31	100	0.68
Class 3	84.6	0.40	92.6	0.32	85.2	0.39	96.2	0.60
Class 4	73.1	0.38	73.1	0.48	75.6	0.34	85.4	0.61
Class 5	95.5	0.51	95.5	0.45	95.5	0.40	100	0.69
Class 6	88.6	0.42	83.3	0.33	90.5	0.44	92.9	0.59
Class 7	92.9	0.31	94.7	0.31	94.7	0.42	94.7	0.63
Class 8	88.2	0.39	91.7	0.40	91.7	0.55	100	0.71
Class 9	91.4	0.38	85.7	0.39	88.6	0.35	91.4	0.53
Class 10	84.6	0.35	76.9	0.32	84.6	0.34	92.3	0.49

Table 3. Comparison of running time of different measures
View table
Table 3. Comparison of running time of different measures
Measure Average running time /s
SURF 5.0230
TF-KSURF 3.5395
Hu moment 2.1013
Measure in Ref.[5] 7.8857
Proposed method 3.9321

Tools

Get Citation

Copy Citation Text

Lili Shang, Huiqin Wang, Ke Wang, Zhan Wang. Multi-Measure Similarity Method for Interpreting Bronze Inscriptions Based on Hu Moment and TF-KSURF[J]. Laser & Optoelectronics Progress, 2021, 58(8): 0810009

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites