由於線型陣列的電荷藕合感測器，各像素的反應靈敏度不均勻，以及輻射模型參數的誤差，使根據海洋水色儀輻射資料所重建的影像往往會出現向軌條紋雜訊，雖然精確的飛前參數校正可以改進雜訊的影響，但是參數量測的不確定性卻無法避免。本論文研究與比較線性匹配法、統計匹配法、功率頻譜濾波法等影像的除紋法則後，提出用感測器匹配法來校正海洋水色儀的輻射模型參數。感測器匹配法是線性匹配法的延伸應用，特點是必須以均勻目標區的影像資料作為依據，主要的目的不但是要消除條紋雜訊，更重要的是要修正輻射模型參數，使去除條紋以後的影像，仍然能夠保持輻射數據的正確性，本文將感測器匹配法應用於處理海洋水色儀的實際影像樣本，並且完成輻射模型參數的校正。

Image of radiance data obtained by the ROCSAT-1 Ocean Color Imager with 896 pixels in the pushbroom scanning may appear in-track stripes resulted by parameter errors in the model for data reconstruction. Although more accurate parameter calibration before launch may be able to improve the striping, uncertainty in the measurement of the parameters is unavoidable. This research investigates conventional methods of image destriping such as linear matching, power spectral filtering and histogram matching and proposes a detector matching method to correct the OCI parameters. The objective is destriping while maintains the correctness of the radiance data instead of just polishing the image. The detector matching is basically a linear matching with constraint on the scene data. The formulation of the detector matching is shown and the corrected gain-offset tables of OCI are obtained. The method is applied to images acquired by OCI and obtains destriping successfully. The results are compared with those obtained by simply using pre-flight parameters and using destriping based on empirical correction.

目 錄
誌謝 …………………………………………………………………… I
中文摘要 ……………………………………………………………… II
英文摘要 …………………………………………………………… III
目錄 ………………………………………………………………… IV
圖目錄 ……………………………………………………………… VI
表目錄 ……………………………………………………………… IX
第一章 緒論 ………………………………………………………… 1
1.1 OCI的校正與資料處理模式 ……………………………………… 1
1.2 條紋雜訊效應與輻射校正的歷史背景 …………………………3
1.3 問題定義與動機 ………………………………………………… 6
1.4 論文架構 ………………………………………………………… 7
第二章 影像除紋法則 …………………………………………… 11
2.1 推掃式多頻譜瞄………………………………………………… 11
2.2 儀器誤差與影像條紋 …………………………………………… 15
2.3 中央濾波法法…………………………………………………… 17
2.4 統計匹配法 ……………………………………………………… 18
2.5 功率頻譜濾波法 ………………………………………………… 22
2.6 線性匹配法 ……………………………………………………… 23
2.7 影像除紋法則的選取與比較 …………………………………… 25
第三章 利用感測器匹配法校正OCI模型參數 …………………… 27
3.1 輻射資料的不確定度 …………………………………………… 27
3.2 OCI的線性輻射模型 …………………………………………… 31
3.3 感測器匹配法 …………………………………………………… 34
第四章 實驗結果與討論 …………………………………………… 39
4.1 OCI實驗準備─資料蒐集與處理步驟 ………………………… 39
4.2實驗結果 ………………………………………………………… 45
4.2.1 暗夜觀測資料 …………………………………………… 45
4.2.2 線性匹配法與感測器匹配法 …………………………… 48
4.2.3 OCI線性輻射模型校正實驗總評 ………………………… 84
4.3 未來的工作 ……………………………………………………… 85
第五章 結論 ………………………………………………………… 87
參考文獻 …………………………………………………………… 88
附錄 ………………………………………………………………… 91

[Buttner, 1990]
G. Buttner, A. Kapovits, 1990: Characterization and removal of horizontal striping from SPOT panchromatic imagery, International Journal of Remote Sensing, Vol.11, No.2, pp.359-366
[Cannon, 1983]
M. Cannon, A. Lehar and F. Preton, 1983: Background Pattern Removal by Power Spectral Filtering, Applied Optics, vol.22, pp.777-779
[Colwell, 1983]
R. N. Colwell, 1983: Manual of Remote Sensing 2nd, American Society of Photogrammetry
[George, 1990]
B. George, K. Adam, 1990: Characterization and removal of horizontal striping from SPOT panchromatic imagery, International Journal of Remote Sensing, vol.11, no.2, pp.359-366
[Horn, 1979]
K. P. Horn, R. J. Woodham, 1979: Destriping LANDSAT MSS Images by Histogram Modifications, Computer Graphics and Image Processing, vol.10, pp. 69-83
[Johnson, 1997]
B. C. Johnson, F. Sakuma, J. J. Butler, S. F. Biggar, J. W. Cooper, J. Ishida and K. Suzuki, 1997: Radiometric Measurement Comparison Using the Ocean Color Temperature Scanner (OCTS) Visible and Near Infrared Integrating Sphere, Journal of Research of the National Institute of Standards and Technology, vol.102, no.6, pp.627-646
[Lee, 1999]
L. S. Lee, Y. J. Chiang, 1999: Instrument Performance of the Ocean Color Imager, TAO, Supplementary Issue, pp. 43-61
[Lin, 1999]
W. S. Lin, J. Y. Wu, H. J. Chiu, C. S. Chen and Y. J. Chang, 1999: Radiometric characterization of the ROCSAT Ocean Color Imager, Internal Journal of Remote Sensing, vol. 20, no. 17, pp. 3247-3263
[Lillesand, 1994]
T. M. Lillesand, R. W. Kiefer, 1994: Remote Sensing and Image Interpretation, pp. 536-540
[Murphy, 1981]
J. Murphy, 1981: A Refined Destriping Procedure for LANDSAT MSS Data Products, Proceedings of the Canadian Symposium on Remote Sensing, pp.454-470
[Pratt, 2000]
W. K. Pratt, 2000: Digital Image Processing 3rd, pp.271-273
[Quarmby, 1987]
N. A. Quarmby, 1987: Noise removal for SPOT HRV imagery, International Journal of Remote Sensing, vol.8, no.8, pp.1229-1234
[Richards, 1985]
M. E. Richards, 1985: An Evaluation of a New Statistical Approach to Traditional Linear Destriping, Technical papers of the American Society of Photogrammetry, vol.2, pp.557-563
[Richards, 1999]
J. A. Richards, X. Jia, 1999: Remote Sensing Digital Image Analysis, pp.43-44
[Sakuma, 1989]
F. Sakuma, A. Ono, 1989: Prelaunch Calibration System for the Short Wavelength Infrared Radiometer of the Earth Resources Satellite-1, Geoscience and Remote Sensing Symposium, vol.15, pp.2942-2945
[Sakuma, 1993]
F. Sakuma, A. Ono, 1993: Spectral Radiancemeters for the Perlaunch Calibration of JERS-1 Ops, Geoscience and Remote Sensing Symposium, vol.13, pp.1309-1311
[Schowengerdt, 1997]
R. A. Schowengerdt, 1997: Remote Sensing: Models and Methods for Image Processing, pp.299-301
[Slater, 1982]
P. N. Slater, 1982: Atmospheric effects on radiation reflected from soil and vegetation as measured by orbital sensor using various scanning direction, Applied Optics, vol.21, pp. 3923
[Srinivasan, 1987]
R. Srinivasan, M. Cannon, and J. White, 1987: Destriping of LANDSAT data using power spectral filtering, Methods of Handing and Processing Imagery, vol.757, pp.34-39
[Srinivasan, 1986]
R. Srinivasan, 1986: Noise removal by the Karhunen-Loeve transform, Proc. Intl. Society for Photogrammetry and Remote Sensing Symposium, vol.26, pp.263-273
[Teillet, 1986]
P. M. Teillet, 1986: Image correction for radiometric effects in remote sensing, International Journal of Remote Sensing, vol.7, no.12, pp.1637-1651
[Wegener, 1990]
M. Wegener, 1990: Destriping multiple sensor imagery by improved histogram matching, International Journal of Remote Sensing, vol.11, no.5, pp.859-875