• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Journal of visual communication & image representation >Blind quality assessment for multiply distorted stereoscopic images towards IoT-based 3D capture systems
【24h】

Blind quality assessment for multiply distorted stereoscopic images towards IoT-based 3D capture systems

机译：基于物联网3D捕获系统的致盲质量评估对IOT的三维捕获系统的差异扭曲立体图像

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

Empowered by 5G mobile communication networks, multimedia processing has been considered as a very promising application of Internet-of-Things (IoT). Stereoscopic image quality assessment (SIQA), as an important part of 3D capture system, can be embedded in the cloud or fog servers to automatically monitor the perceptual quality of the collected stereoscopic images. In this paper, a novel blind image quality assessment method towards IoT-based 3D capture systems is developed for multiply-distorted stereoscopic images (MDSIs), in which five complementary channels, including left view, right view, cyclopean map, summation map and difference map, are jointly considered in dictionary learning for characterizing the monocular receptive field (MRF) and binocular receptive field (BRF) properties of the visual cortex in response to MDSIs. Additionally, the high order statistics scheme is adopted by utilizing the statistical differences between the codebook and images to ensure the stable and robust quality prediction performance for MDSIs. The proposed method shows competitive prediction performances on four benchmark databases compared with the existing SIQA metrics.
机译：由5G移动通信网络赋予授权，多媒体处理被认为是对互联网（物联网）的有前途的应用。立体图像质量评估（SIQA）作为3D捕获系统的重要组成部分，可以嵌入在云或雾服务器中，以自动监测收集的立体图像的感知质量。在本文中，开发了一种新的基于IOT的3D捕获系统的盲目图像质量评估方法，用于乘法失真立体图像（MDSIS），其中五个互补信道，包括左视图，右视图，Cyperopean地图，求和图和差异地图，在响应于MDSIS时，在分类学习中共同考虑在字典学习中，用于表征视觉皮质的单眼接收领域（MRF）和双目接收领域（BRF）性质。另外，通过利用码本和图像之间的统计差异来采用高阶统计方案，以确保MDSIS的稳定且稳健的质量预测性能。该方法与现有的SIQA指标相比，在四个基准数据库上显示了竞争预测性能。

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