Image Deblurring and Decomposition: Texture and Color Image Analysis

图像去模糊和分解：纹理和颜色图像分析

• 批准号: 0908517
• 负责人: Sung Kang
• 金额: $ 8.23万
• 依托单位: Georgia Tech Research Corporation
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2008
• 资助国家: 美国
• 起止时间: 2008-09-01 至 2010-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0908517&HistoricalAwards=false
• 关键词: Image; Deblurring; Decomposition; Texture; Color

Kang0707184 One of the recent developments of mathematical imageprocessing is the introduction of the G-norm by Meyer, whichopened the door for handling texture and oscillations. Theinvestigator further develops these techniques for color imagesand blurred texture analysis. The goals of this project are toexplore more sophisticated color models for image decomposition,as well as to correctly capture the texture in blurry images. The main challenge is to tackle image blind deconvolution incombination with image decomposition techniques. The investigator studies the classical problem of imagedeblurring in a modern setting of texture and color imageanalysis. Recognition of detail has become more crucial forbetter diagnosis of medical images and for more accurate recoveryof satellite images. This project provides theoreticalbackground for methodologies and can be applied to solve realimaging problems arising from digital photo restorations, medicalimaging, communications, information science, and many otherapplications.

KANG0707184数学图像过程的最新发展之一是Meyer的G-Norm引入，它为处理质地和振荡的门开了大门。 The Investigator进一步开发了这些技术，以用于颜色图像和模糊纹理分析。 该项目的目标是探索图像分解的更复杂的颜色模型，以及正确捕获模糊图像中的纹理。主要的挑战是通过图像分解技术来解决图像盲目反卷积结合。 研究人员研究了在现代质地和颜色成像分析的现代环境中成像的经典问题。 对细节的识别已成为对医学图像的诊断变得更加关键，并为卫星图像更准确地恢复。 该项目为方法论提供了理论背景，并可以应用于解决由数字照片修复，医学影像，通信，信息科学和许多其他应用程序引起的实现问题。