{"title":"基于集成卷积神经网络的场景识别实例研究","authors":"B. Oh, Junhyeok Lee","doi":"10.23919/ICACT.2018.8323751","DOIUrl":null,"url":null,"abstract":"This paper proposes architecture to recognize scene images based on an ensemble of two convolution neural networks. A convolution neural network is used to train massive scene images, and the other convolution neural network is used to extract objects from the scene images. The object lists are stored according to scene classes, and used as a clue to decide the top-1 and top-5 classes during scene image recognition stage.","PeriodicalId":228625,"journal":{"name":"2018 20th International Conference on Advanced Communication Technology (ICACT)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2018-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"5","resultStr":"{\"title\":\"A case study on scene recognition using an ensemble convolution neural network\",\"authors\":\"B. Oh, Junhyeok Lee\",\"doi\":\"10.23919/ICACT.2018.8323751\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This paper proposes architecture to recognize scene images based on an ensemble of two convolution neural networks. A convolution neural network is used to train massive scene images, and the other convolution neural network is used to extract objects from the scene images. The object lists are stored according to scene classes, and used as a clue to decide the top-1 and top-5 classes during scene image recognition stage.\",\"PeriodicalId\":228625,\"journal\":{\"name\":\"2018 20th International Conference on Advanced Communication Technology (ICACT)\",\"volume\":\"48 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2018-02-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"5\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2018 20th International Conference on Advanced Communication Technology (ICACT)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.23919/ICACT.2018.8323751\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2018 20th International Conference on Advanced Communication Technology (ICACT)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.23919/ICACT.2018.8323751","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
A case study on scene recognition using an ensemble convolution neural network
This paper proposes architecture to recognize scene images based on an ensemble of two convolution neural networks. A convolution neural network is used to train massive scene images, and the other convolution neural network is used to extract objects from the scene images. The object lists are stored according to scene classes, and used as a clue to decide the top-1 and top-5 classes during scene image recognition stage.