臺灣博碩士論文加值系統

傳統的生物光學演算法使用迴歸方法，利用輻射量求取葉綠素濃度，如海洋廣角感測器（簡稱SeaWiFS）所使用的OC2演算法。本篇論文利用倒傳遞網路演算法，建立海水中葉綠素濃度與海水輻射量間的生物光學模式。使用類神經網路的優點為：1) 利用類神經網路的非線性解題能力能有效建立葉綠素濃度與輻射量間的轉換函數。2) 類神經網路使用所有可用的波段，而非傳統的方法僅使用兩個波段比值，如此可有效利用所有波段所包含的訊息。
本篇論文使用SeaWiFS生物光學演算會議（SeaWiFS Bio-optical Algorithm Mini-workshop, SeaBAM）所獲得的生物光學資料，發展並驗證類神經網路生物光學演算法。結果顯示使用類神經網路推算生物光學演算法時，其葉綠素濃度檢定係數為0.98，葉綠素濃度均方根誤差值為0.59 ug/l。此結果遠較傳統的OC2v2演算法準確，OC2v2的葉綠素濃度檢定係數為0.56，葉綠素濃度均方根誤差值為2.84 ug/l。

Traditional bio-optical algorithm uses regression method to derive chlorophyll concentration form radiance, such as the OC2 algorithm for the Sea-viewing Wide Field-of-view Sensor (SeaWiFS). In this study, a back propagation neural network is employed to set up the bio-optical algorithm between chlorophyll concentration in the oceans and radiance from the oceans. The advantages of the neural network algorithm are 1) the ability for solving the non-linear transfer function between chlorophyll concentration and radiance is better than that of regression method. 2) it uses the information of all bands instead of the two-band ratio of traditional algorithm.
Bio-optical data from SeaWiFS Bio-optical Algorithm Mini-workshop (SeaBAM) were used to develop and verify the neural network algorithm. The result show that the coefficient of determination ( R ) for the neural network algorithm is 0.98 and the root-mean-square (RMS) error of derived chlorophyll concentration is 0.59 ug/l. They are much better than the results of version 2 of OC2 (OC2v2) algorithm, while the R is 0.56 and the RMS error is 2.84 ug/l.

中文摘要 Ⅰ
英文摘要 Ⅱ
誌謝 Ⅲ
目錄 Ⅳ
表目錄 Ⅵ
圖目錄 Ⅶ
第一章 導論 1
1.1 研究動機 1
1.2 文獻回顧 3
1.2.1 海洋水色 3
1.2.2 類神經網路 7
1.3 論文大綱 8
第二章 理論與方法 9
2.1 類神經網路 9
2.1.1 類神經網路原理 9
2.1.2 生物神經學說 9
2.1.3 常用之非線性轉換函數 11
2.1.4 類神經網路之運算過程 13
2.1.5 倒傳遞網路演算摘理論推導 14
2.2 海洋水色生光模式演算法 17
2.1.1 遙測返照率 17
2.1.2 離水輻射量 19
2.3 光譜值資料來源與規範 21
2.3.1 SeaBAM篩選資料原則 21
2.3.2 SeaBAM最佳葉綠素濃度推算模式 23
第三章 神經網路應用於海洋生光模式 29
3.1 神經網路初試 31
3.2 迴歸公式的誤差 42
3.3 資料的選取 50
3.4 類神經網路生光模式 64
第四章 結論與建議 75
參考文獻 77

岸野元彰（1981）.海洋光合成に關連する海中光エネルギ一收支，太陽エ
ネルギ一の生物、化學的利用Ⅲ，柴田和雄‧池上明‧山田瑛，學會出版 ヤンタ一，278頁。
Adrian, E. D. (1946). “The physical background perception”, Oxford:Clarendon Press, New York.
Austin, R. W. ( 1974). “The remote sensing of spectral radiance from below the ocean surface”, in Optical Aspects of Oceanography (ed. By V. G. Jerlov and W. S. Nielson), Academic Press, pp. 317-344.
Baum, E. and D. Haussler (1989). “What size net gives valid generalization?,” In Touretzky, D., ED., Advances in Neural Information Processing Systems 1: Morgan Kaufmann, pp. 81-90.
Bukata, R. P., J. H. Bruton, and J. H. Jerome (1991). “The state of vegetation cover as an indicator of groundwater, based on observations from space”, Soviet Journal of remote sensing, 9 (2) : pp. 328-343.
Campbell, J. W. (1995). “The lognormal distribution as a model for bio-optical variability in the sea,” J. Gerphys. Res., vol. 100, pp. 13,237-13,254.
Carder, K. L. and R. G. Steward (1985). ”A remote-sensing reflectance model of red-tide dinoflagellate off West Florida,” Limnol. Oceanogr., 30 : pp. 286- 298.
Clark, D. K. (1981). ”Phytoplankton pigment algorithms for the Nimbus-7 CZCS,” in Oceanography from space (ed. By J.G.R. Gower), Pleum Press, New York, pp. 227-238.
Duntley, S. Q., R. W. Austin, W. H. Wilson, C. F. Edgerton and S. E. Moran (1974). “Ocean color analysis”, Scripts Institution of Oceanography, Uni- versity of California,San Diego. SIO Ref. 74-10.
John, E., O’Reilly, M. Stephane, B. G. Mitchell, D.A. Siegel, L. K Carder, S.A. Garver,M. Kahru, and C.McClain (1998). ”Ocean color chlorphyll algori- -thms for SeaWiFS,” J. Gerphys. Res.,vol. 103, pp. 24,937-24,953
Gibbs, C. F. (1979). “Chlorophyll b interference in the fluorometric determin- ation of chlorophyll a and phaeo-pigment”. Aust. J. Mar. Freshwater Res.,30: pp. 597-606.
Gordon, H. R., O. B. Brown, R. H. Evans, J. W. Brown, R. C. Smith, K. S. Baker, and D. K. Clark (1988). “A semianalytic radiance model of ocean color,” J.Geophys.Res.,93, pp. 10,909-10,924.
Gordon, H. R., D. K. Clark, J. W. Brown, O. B. Brown, R. H. Evans, W. W. Broenkow (1983). “Phyto-plankton pigment concentrations in the Middle Atlantic Bight: comparisons between ship determinations andcoastal zone color scanner estimates,” Appl. Optics 22, pp. 20—36.
Gordon, H. R., D. K. Clark, J. L. Mueller, W.A. Hovis (1980). “Phytoplankton pigments derived from Nimbus 7 CZCS: Initial comparisons with surface measurements.” Science 210, pp. 63—66.
Gordon, H. R. and W. R. McCluney (1975). “Estimation of depth of sunlight penetration in sea for remote sensing,” Appl. Opt.,18:pp. 1161-1166.
Grossberg, S. (1976). “Adaptive pattern classification and universal recoding:I. Parallel development and coding of neural detectors, ”Biological Cybernetics, vol. 23, pp. 121-134.
Hagan, M. T.,Menhaj, M. (1994). “Training feedforward networks with the Marquardt algorithm,” IEEE Transactions on Neural Networks,5(6):989.
Hecht-Nielson, R. (1987), “Counter-propagation networks,”ICNN-87,Ⅱ, pp. 19-32.
Hebb, Donald O. (1949). The Organization of Behavior:Wiley.
Holm-Hansen, O., C. J. Lorenzen ,R. W. Holmes and J. D. H. Strickland (1965). “Fluorometric determination of chlorophyll”. J. Cons. Perm. Int. Explore. Mer., 30:3-15.
Hopfield, J. J. (1982). “Neural networks and physical systems with emergent collective computational abilities,” Proceedings of the National Academy of Science,USA, vol. 79, pp. 2554-2558.
Hopfield, J. J. and T. W. Tank (1985). “Neural computation of decisions in optimization problems, “Psychological Review, vol. 52, pp. 141-152.
Jorlov, N. G. (1968). Optical Oceanography, Elsevier Publishing Company, Amsterdam,194.pp
Keiner, L.E., X. H. Yan (1998). “A neural network model for estimating sea surface chlorophyll and sediments from thematic mapper imagery”, Remote sensing of environment,66 (2):pp. 153-165.
Kohonen, T., G. Barna, and R. Chrisley (1988). “Statistical pattern recongnition with neural networks:benchmark studies, “IEEE International Conference on Neural Networks, vol. Ⅰ, pp. 61-68, San Diego,CA.
Kohonen, T. (1982). “Self-organized formation of topologically correct feature maps, “Biological Cybernetics, vol. 43, pp. 59-69.
Kosko, B. (1988). “Bidirectional associative memories,” IEEE Transactions on Ststems, Man, and Cybernetics, vol. 18, pp. 49-60.
Lorenzen, C. J., and S. W. Jeffrey (1980). Determination of chlorophyll in seawater, UNESCO Tech.Pap.,Mar. Sci., 35, 20.
Mackay, DJC. (1992). “A practical Bayesian framework for backprprogation networks”, Neural Compution,4 (3): 448-472 MAY 1992
Maritorena, S. A. Morel , B. Gentili (1994). “Diffuse-reflectance of oceanic shallow waters-influence of water depth and bottom albedo”, Limnology and Oceanography.39 (7): 1689-1703 NOV 1994
McCulloch, W. S. and W. Pitts (1943). ”A logical calculus of the ideas immanent in nervous activity,” Bulletin of Mathematical Biophysics, vol. 5, pp. 115-133.
Minsky, M., and S. Papert (1969). Perceptrons:MIT Press.
Morel, A. (1988). “Optical modeling of the upper ocean in relation to its biogenous matter content (Case 1 waters)”,J. Geophys. Res.,93,pp. 10748- 10768.
Morel, A. and L. Prieur (1977). ”Analysis of variations in ocean color ”.Limnol Oceanogr.,22(4):709-722.
Mueller, J. L., and R. W. Austin (1995). “Ocean optics protocols for SeaWiFS validation, Revision 1,” NASA Tech. Memo, 104566,vol. 25, 67 pp.
Murray, A. P.,C. F. Gibbs, A. R. Longmore and D. J. Flett (1986). “Deter- mination of chlorophyll in marine water:Intercomparision of a rapid HPLC method with full HPLC,spectrophotometric and fluorometric methods” . Marine Chem”.,19:pp. 211-227.
Powell, M. J. D. (1988). “Radial basis function approximations to polyn- omials, ”Numerical Analysis 1987 Proceedings, pp. 223-241.
Robinson, I. S. (1983). “Satellite observations of ocean colour ”, Phil. Trans. Roy. Soc. London., A309 : pp. 415-432.
Rosenblatt, F. (1958). “The Perceptron:A probabilistic model for information storage and organization in the brain, ”Psychological Review, vol.65, pp. 386-408
Rumelhart, D. E., G. E. Hinton, and R. J. Williams (1986). “Learning represent- ations of back-propagation errors,” Nature(London), vol. 32, pp. 533-536.
Sathyendranath, S. and A.Y.Morel (1983). Light emerging from the sea interpret- ation and uses in remote sensing.Remote sensing applications in marine science and technology. A. P. Cracknell and D. Deidel. eds., Dordrecht,pp. 323-358.
Smith, R. C. and K. S. Baker (1978). ”The bio-optical state of ocean waters and remote sensing”,Limnol.Oceanogr.,23:247:259.
Specht, D. F. (1990). “Probabilistic neural networks, “Neural Networks, vpl. 3, pp. 109-118.
Strickland, J. D. H. and T. R. Parsons (1972). “A practical handbook of sea water Analysis”. Fisheries research board of Canada,167,310 pp.
Tassan, S. (1994). “Local algorithms using SeaWiFS data for the retrieval of phytoplankton,pigments,suspended sediment and yellow substance in coastal water”.Appl. Opt.,33(12):2369-2378.
Tree, C. C., M. C. Kennicutt and J. M. Brooks (1985). “Errors associated with the standard fluorometric determination of chlorophylls and phaeopig- ments.” Marine Chem., 17:1-12.
Welschmeyer, N. A. (1994). “Fluorometric analysis of chlorophyll a in the presence of chlorophyll b and phaeopigments,” Limnol. Oceanogr., 39(8),1985-1992.
Werbos, P. (1974). “Beyond regression:New tools for prediction and analysis in the behavioral sciences: Ph.D. Dissertation,” Applied Math, Harvard University, Cambridge, MA.
Widrow, B., and M. Hoff (1960). “Adaptive switching circuits”: IRE WESCON Convention Record, pp. 96-104.
Van den Bout, D. E., and T. K. Miller (1988), “A traveling salesman objective functions that works, “ICNN-88, Ⅱ, 299-303.
Yentsch, C. S. and D. W. Menzel (1963). ”A method for determination of phyto- plankton chlorophyll and phaeophytin by fluorescence”.Deep Sea Res.,pp. 10:221-231.