...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>International Journal of Innovative Computing Information and Control >VIDEO RESOLUTION ENHANCEMENT USING DEEP NEURAL NETWORKS AND INTENSITY BASED REGISTRATIONS
【24h】

VIDEO RESOLUTION ENHANCEMENT USING DEEP NEURAL NETWORKS AND INTENSITY BASED REGISTRATIONS

机译：使用深层神经网络和基于强度的注册的视频分辨率增强

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Thanks to the recent rapid improvements made to the maximum possibleresolution of display devices, higher qualities of experience have been made possible, whichnecessitates either producing and transmitting considerably higher volumes of data orsuper-resolving lower-resolution contents at the display side, where the former mightnot be practically feasible. Therefore, aiming at the latter, this paper proposes a novelsuper-resolution technique, which takes advantage of convolutional neural networ'ks. Eachimage is registered into a window consisting of two frames, the second one standing forthe reference image, using various intensity-based techniques, which have been testedand compared throughout the paper. According to the experimental results, the proposedmethod leads to substantial enhancements in the quality of the super-resolved images,compared with the state-of-the-art techniques introduced within the existing literature. Onthe Akiyo video sequence, on average, the result possesses 5.38dB higher PS NR valuesthan those of the Vandewalle registration technique, with structure adaptive normalisedconvolution being utilized as the reconstruction approach.
机译：由于最近对显示设备的最大可能分辨率进行了快速改进，因此可以实现更高质量的体验，这需要在显示侧生成和传输大量数据或超分辨率的低分辨率内容，而前者可能不会切实可行。因此，针对后者，本文提出了一种利用卷积神经网络的新型超分辨率技术。使用各种基于强度的技术，将每个图像记录到一个由两帧组成的窗口中，第二帧代表参考图像，该技术已在整个论文中进行了测试和比较。根据实验结果，与现有文献中介绍的最新技术相比，该方法可显着提高超分辨图像的质量。在Akiyo视频序列上，结果平均比Vandewalle配准技术具有5.38dB的PS NR值，并使用结构自适应归一化卷积作为重建方法。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号