空間混沌模型於合成孔徑雷達影像變遷偵測之研究 ; A Study on Change Detection in SAR Images using Spatial Chaotic Model

NCU Institutional Repository > 資訊電機學院 > 資訊工程研究所 > 博碩士論文 > Item 987654321/44660

jsp.display-item.identifier=請使用永久網址來引用或連結此文件: http://ir.lib.ncu.edu.tw/handle/987654321/44660

题名:	空間混沌模型於合成孔徑雷達影像變遷偵測之研究;A Study on Change Detection in SAR Images using Spatial Chaotic Model
作者:	周念湘;Nien-Shiang Chou
贡献者:	資訊工程研究所
关键词:	碎型;空間混沌模型;wavelet;fractal;change detection;fractional Brownian motion;SCM;SAR
日期:	2010-07-25
上传时间:	2010-12-09 13:51:56 (UTC+8)
出版者:	國立中央大學
摘要:	於本篇論文中，我們利用空間混沌模型(Spatial chaotic model)設計了一個新的變遷偵測演算法，用以辨認兩張同地點不同時間取得之合成孔徑雷達(SAR)影像的變異，以下簡稱SCM。此方法植基於具同調性的合成孔徑雷達影像可被塑模成空間混沌系統。在此研究中，偵測效能指標分成三類：辨識率(Detection rate)、誤判率(False detection rate)以及漏判率(Loss detection rate)；我們的方法都優於簡易影像相異法(Simple image difference)以及主成分分析法(Principle component analysis)。尤其在兩張前後期影像未完美疊合(mis-registration)，或者是變化細微的情況下，SCM的優勢更加明顯。在通常的SAR影像變遷偵測的方法中，去斑駁雜訊的前置程序是必要的，但也同時降低了影像的幾何特徵或者是空間解析度，進而降低辨識率；SCM不需要進行去斑駁的前置程序，其偵測率因而相對提高。In this thesis, we propose a new change detection algorithm for SAR images using the concept of spatial chaotic model. The new method was built on the fact that the coherent SAR images can be modeled by a spatial chaotic system. The proposed method was applied to multi-temporal polarimetric SAR images for change detections. As a reference, the simple image difference (DI) technique and the principal component analysis (PCA) were compared. Also, images mis-registration effects were also tested. The proposed method (hereafter called SCM) is capable of tolerating mis-registration effect even when the signal-to-noise ratio is relatively low, as compared to both DI and PCA methods. Comparison was made on the case when the radiometric changes are subtle. It is shown that the proposed method performs very well to detect such diminutive changes without being deteriorated by the presence of speckle for which both DI and PCA fail to carry out the detection.
显示于类别:	[資訊工程研究所] 博碩士論文

文件中的档案:

档案	描述	大小	格式	浏览次数
index.html		0Kb	HTML	614	检视/开启

在NCUIR中所有的数据项都受到原著作权保护.

社群 sharing

数据加载中.....