Novel image analysis and computer vision techniques for multimedia, communication, and other applications

用于多媒体、通信和其他应用的新颖图像分析和计算机视觉技术

• 批准号: 312297-2010
• 负责人: Wu, Jonathan
• 金额: $ 1.89万
• 依托单位: University of Windsor
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2014
• 资助国家: 加拿大
• 起止时间: 2014-01-01 至 2015-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=548693
• 关键词: Novel; image; analysis; computer; vision

Information science and electronics are among the most important areas of science and technology in the 21st century. The key networked and digital living technologies will have a major impact to our daily life and various industrial sectors. How to employ both audio and visual information for interactive multimedia, communication has recently received much attention in both academia and industry. This research will deal with a broad range of topics concerning the development of novel image processing and computer vision algorithms and systems for security, communication, and other applications.The objectives of the proposed research are: (a). To develop an improved and efficient intra prediction algorithm for video coding with high compression efficiency. (b). To further develop new theory and new methods on camera calibration and three dimensional reconstruction from image sequences of rigid and nonrigid scenarios. (c). To develop novel biometric watermarking scheme to communicate and transmit biometrics with full protection. Biometrics is gaining more and more interest in public as well as in private sector since these (fingerprints, DNA, or irises) are the unique identifiers and permanent characteristic of the individuals. These properties make biometrics not only a very powerful tool for matching different pieces of information of individuals across multiple databases but also attempt to increase the security.

信息科学和电子学是21世纪最重要的科学技术领域之一。关键的网络化和数字化生活技术将对我们的日常生活和各个工业部门产生重大影响。如何利用音频和视频信息进行交互式多媒体通信最近受到学术界和工业界的广泛关注。这项研究将涉及广泛的主题，涉及开发新的图像处理和计算机视觉算法和系统，用于安全，通信和其他应用。提出了一种改进的高效帧内预测算法，用于视频编码，提高了压缩效率。（B）。进一步发展刚性和非刚性场景下摄像机标定和图像序列三维重建的新理论和新方法。（c）.研究一种新颖的生物特征水印方案，以实现生物特征信息的完整保护。生物识别技术在公共和私营部门越来越受到关注，因为这些（指纹，DNA或虹膜）是个人的唯一标识符和永久特征。这些特性使生物识别技术不仅成为一种非常强大的工具，用于在多个数据库中匹配个人的不同信息，而且还试图提高安全性。