{"title":"一种基于ICA的视频盲水印方案","authors":"Jiande Sun, Ju Liu, M. Lagunas","doi":"10.1109/ICNNSP.2003.1281067","DOIUrl":null,"url":null,"abstract":"A novel blind video watermarking scheme based on motion estimation of original video is proposed in this paper. The motion component is extracted by independent component analysis (ICA) from the sequent frames. The watermark is embedded into the motion component according to the mean of the neighboring wavelet coefficients because the motion component is essential to the video and the compression affects little on it. The watermark can be extracted blindly and there is little probability of false error. Computer simulations show the good performance of the proposed scheme in robustness and invisibility.","PeriodicalId":336216,"journal":{"name":"International Conference on Neural Networks and Signal Processing, 2003. Proceedings of the 2003","volume":"2 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":"{\"title\":\"An blind video watermarking scheme based on ICA\",\"authors\":\"Jiande Sun, Ju Liu, M. Lagunas\",\"doi\":\"10.1109/ICNNSP.2003.1281067\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"A novel blind video watermarking scheme based on motion estimation of original video is proposed in this paper. The motion component is extracted by independent component analysis (ICA) from the sequent frames. The watermark is embedded into the motion component according to the mean of the neighboring wavelet coefficients because the motion component is essential to the video and the compression affects little on it. The watermark can be extracted blindly and there is little probability of false error. Computer simulations show the good performance of the proposed scheme in robustness and invisibility.\",\"PeriodicalId\":336216,\"journal\":{\"name\":\"International Conference on Neural Networks and Signal Processing, 2003. Proceedings of the 2003\",\"volume\":\"2 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"1900-01-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"3\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"International Conference on Neural Networks and Signal Processing, 2003. Proceedings of the 2003\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICNNSP.2003.1281067\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Conference on Neural Networks and Signal Processing, 2003. Proceedings of the 2003","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICNNSP.2003.1281067","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
A novel blind video watermarking scheme based on motion estimation of original video is proposed in this paper. The motion component is extracted by independent component analysis (ICA) from the sequent frames. The watermark is embedded into the motion component according to the mean of the neighboring wavelet coefficients because the motion component is essential to the video and the compression affects little on it. The watermark can be extracted blindly and there is little probability of false error. Computer simulations show the good performance of the proposed scheme in robustness and invisibility.