Minority Costume Connective Learning System Based On Image Recognition

Mr.Lu Xun once said that "national is the world",national culture as a carrier of national history,carrying the wisdom and characteristics of the minority.Yunnan Province as a large minority provinces,actively respond to the call of the state and the government,vigorously inheritance and protection of national culture.Minority costumes as the ancient art of the minority is the main symbol of the nation different from other nationalities,and reflects the historical and cultural imprint of various nationalities.However,due to the impact of modern culture,minority young people no longer put the minority costumes as the main clothing,minority clothing inheritance crisis,with the rapid development of computer technology,the majority of scholars began to digital protection of minority costumes launched a variety of studies based on the characteristics of minority costumes.In this paper,according to the characteristics of minority costumes to carry out clothing recognition research,and clothing recognition and clothing culture learning effectively combined for the transmission of national culture to find a new method,trying to contribute to the national cultural heritage,the main research as follows:(1)Based on the image recognition technology and the theoretical basis of connective learning view,Web2.0 and student input,a connective learning model is constructed.In the model,the paper analyzes the role of image recognition technology in the study of connective learning,constructs the learners,learning resources,learning emotion and learning support tools as the main constituent elements,and elaborates the learning process of learners based on image recognition technology.(2)The HOG + SVM algorithm is well applied to the classification and recognition of minority costumes.First of all,compared with different classifiers,the support vector machine(SVM)algorithm training classifier in the minority costumes on the classification of the best results.In order to improve the classification effect,the classical algorithm of color,texture and shape feature extraction is experimentedwith SVM in the database collected from various minority groups in Yunnan.Finally,the HOG feature combined with SVM classification algorithm is used in the minority costumes classification on the very good results.(3)The development of a minority costume connective learning system based on image recognition technology.According to the learning model proposed in this paper,the costume image classification algorithm and the design of national costume resources under connective learning,the use of Web technology for system development,ultimately to achieve the design of the learning system.This paper designs the national costume communication system based on image recognition technology,uses the collective wisdom to share the learning resources together,can keep the knowledge growth,avoid the knowledge aging,thus makes the knowledge more and more close to the people's needs.The learning system breaks traditional learning methods,builds a learning model in line with the information society.