Classification Method of Ethnic Minority Patterns Based on Faster R-CNN

The pattern on the costumes and brocades of ethnic minorities are of profound significance, but they are various and often composed of many kinds of patterns. It is not accurate and time-consuming to classify them only by artificial methods. Taking Yao’s pattern symbols as an example, this paper collects and arranges the pattern pictures on clothing and brocade, preprocesses the pictures and labels the patterns on the processed pictures according to the preliminary classification. After the data set is made, the data set is trained and tested by Faster R-CNN algorithm. The results show that this method can effectively identify and classify the patterns of Yao nationality while reducing the time-consuming, and the average accuracy can reach 88.71%. It provides a useful exploration to help more ethnic minorities realize the intelligent classification of patterns.

[1]  Kaiming He,et al.  Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[3]  Ross B. Girshick,et al.  Fast R-CNN , 2015, 1504.08083.

[4]  Trevor Darrell,et al.  Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation , 2013, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[5]  Liu Zihao,et al.  Elements classification of vein patterns using convolutional neural networks for blue calico , 2020 .