{"title":"LOCATION AND SEGMENTATION OF FACIAL FEATURES COMBINING PSO ALGORITHM AND SKIN COLOR","authors":"LI Zhi-jie","doi":"10.53555/eijse.v5i3.69","DOIUrl":null,"url":null,"abstract":"A new method using PSO algorithm and skin color for the location and segmentation of face and facial features is proposed. In the preprocessing stage, segmented face image is obtained from initial color image. To achieve this goal, PSO algorithm is applied to search for the best face region. Then, based on the edge density of face image, eye region is located with PSO. Then, lips region is located using color component of skin segmentation. Finally, nose region is segmented based on the result of eyes and lips. The Simulation results show that this hybrid method is accurate and effective. \n ","PeriodicalId":354866,"journal":{"name":"EPH - International Journal of Science And Engineering","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2019-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"EPH - International Journal of Science And Engineering","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.53555/eijse.v5i3.69","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
A new method using PSO algorithm and skin color for the location and segmentation of face and facial features is proposed. In the preprocessing stage, segmented face image is obtained from initial color image. To achieve this goal, PSO algorithm is applied to search for the best face region. Then, based on the edge density of face image, eye region is located with PSO. Then, lips region is located using color component of skin segmentation. Finally, nose region is segmented based on the result of eyes and lips. The Simulation results show that this hybrid method is accurate and effective.