( 您好!臺灣時間:2021/05/08 16:23
字體大小: 字級放大   字級縮小   預設字形  


研究生(外文):Guang-Ray Tsai
論文名稱(外文):Developing a portable hyperspectral camera to monitor air quality index
指導教授(外文):Jao-Jia Horng
外文關鍵詞:Multivariate Linear RegressionRemote sensingSupport Vector MachineHyperspectral cameraPollutants Standard Index
  • 被引用被引用:0
  • 點閱點閱:237
  • 評分評分:系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
傳統周遭空氣品質監測站係依人口數比例架設,僅能單點採樣,儀器精密且維修亦較複雜。目前遙測技術已廣泛應用於環境污染之監測,因此本研究擬開發一套可攜式高光譜影像儀(HyCAM-I)藉以更機動性地遙測空氣污染指標(Pollutants standard index, PSI),藉由建立之空氣品質光譜預測模式,用以即時監測任一未知區域之空氣品質。
本研究建模方法採用支撐向量機迴歸(SVR)模式與多變量線性迴歸(MLR)模式,並以500nm, 550nm及600nm波段反射率做為模式輸入變量以預測PSIhr,兩模式於學習樣本之R2(決定係數)分別為0.79與0.46;RMSE(均方根誤差)分別為6與10;RMSE/StDev(均方根誤差佔觀測值標準偏差比例)分別為0.46與0.73;MAPE(平均絕對誤差百分比)分別為8%與14%。如進一步將模式以多次隨機交叉驗證作不確定分析,兩模式在驗證樣本之平均R2分別為0.28與0.47;平均RMSE分別為12與10;平均RMSE/StDev分別為0.93與0.77;平均MAPE分別為20%與17%。顯示多變量線性迴歸模式雖未有良好學習結果但較支撐向量機迴歸模式穩定,亦說明支撐向量機迴歸模式會產生過度學習地現象。綜合而言,在分析目標物方面,以綜合性之PSIhr指標結果較其他單項污染物指標佳,其最佳波段值約位於500nm及600nm附近;MLR模式穩定性優於SVR模式。依此結果顯示以可攜式高光譜影像儀用於監測空氣污染指標相當具有其可行性,可機動作為即時監控各區域空氣品質之工具,惟仍有空間未來再改善實驗設備以及更多數據驗證,建立更具代表性之空氣污染物光譜資料。
Ground-based air quality observations were set up traditionally in accordance-populated area. It could only sample at single location with operation and maintenance. Nowadays, remote sensing technology has been widely applied to monitor air environment, this research is intended to develop a portable hyperspectral camera (HyCAM-I) to monitor air Pollutant Standard Index (PSI) remothly and flexibly. With the establishment of air quality prediction model from the hyperspectral data and PSIhr, we can measure a regional air quality promptly by mobile HyCAM-I rathen than the traditional site-specific monitoring data.
For building up the air quality prediction model, this study adopted Support Vector Regression (SVR) and Multivariate Linear Regression (MLR) models to analysis the measured hyperspectral data and air quality index. Three bands of 500nm, 550nm and 600nm were used as the input variables to estimate output of the PSIhr. The learning results have shown that the RMSE (Root Mean Square Error) the samples were 6 and 10 respectively for SVR and MLR, and the MAPE (Mean Absolute Percentage Error) were 8% and 14%. The uncertainty analysis with random cross- validation (CV) indicated that the estimation of the average RMSE of validations were 12 and 10, and the average MAPE of validations were 20% and 17%, respectively. In general, the results have shown that the SVR model have better estimation than that of over-learning MLR. However, the validation results did not prove the whole advantage of SVR model whereas the MLR model has better and more stable validations. Conclusionly, The prediction errors of the MLR model were acceptable but need to be verified by improving hyperspectral device and by expanding sample numbers to enhance the reliability.
摘要 i
致謝 v
目錄 vi
圖目錄 ix
表目錄 xii
一、 前言 1
1.1. 研究動機及目的 1
1.2. 文獻回顧 2
1.2.1. 衛星及高光譜遙測應用於空氣污染相關研究 2
1.2.2. 機器學習(支撐向量機)相關研究 4
1.2.3. 相關性分析相關研究 6
1.2.4. 其他相關研究 6
1.3. 研究流程及架構 7
二、 研究區域資料分析 9
2.1. 研究區域 9
2.2. 空氣品質監測站 9
2.3. 空氣品質標準 12
三、 研究方法及理論 16
3.1. 光譜反射及遙感探測 16
3.1.1. 遙感探測 16
3.1.2. 光譜反射特性 16
3.2. 實驗儀器 16
3.2.1. 可攜式高光譜影像儀 17
3.2.2. 儀器特性及輻射校正 18
3.3. 實驗方法 24
3.3.1. 實驗步驟 24
3.3.2. 資料前處理 25
3.3.3. PSI修正方式 25
3.4. 支撐向量機(Support Vector Machines, SVM) 37
3.4.1. SVM基礎理論 37
3.4.2. 支撐向量機迴歸(Support Vector Regression, SVR) 42
3.4.3. 參數最佳化 44
3.5. 多變量線性迴歸(Multivariate Linear Regression, MLR) 45
3.6. 相關性分析(Correlation Analysis) 45
3.6.1. 皮爾森相關係數(Pearson Correlation Coefficient) 45
3.6.2. 決定係數(Coefficeint of Determination) 46
3.7. 交叉驗證(Cross Validation) 47
3.8. 評估指標 48
3.9. 模式建立 49
四、 結果與討論 53
4.1. 變量篩選 55
4.2. 兩模式學習結果 65
4.2.1. 支撐向量機迴歸(SVR)學習結果 65
4.2.2. 多變量線性迴歸(MLR)學習結果 66
4.2.3. 兩模式學習結果(平均PSIhr) 74
4.3. 交叉驗證(CV)分析結果及綜合比較 76
4.3.1. SVR之CV分析結果 76
4.3.2. MLR之CV分析結果 77
4.3.3. 兩模式之CV分析結果(平均PSIhr) 78
4.4. SVR及MLR之分析比較 81
五、 結論與建議 92
5.1. 結論 92
5.2. 建議 93
參考文獻 95
附錄 100
1.Camps-Valls, G., L. Gomez-Chova, J. Munoz-Mari, J. Vila-Frances, J. Amoros-Lopez and J. Calpe-Maravilla et al, 2006, “Retrival of oceanic chlorophyll concentration with relevance vector machines”, Remote Sensing of Environment, 105, 23-33.
2.Chen, X., Y. S. Li, Z Liu, K. Yin, Z. Lid, O. W. Wai and B. King, 2004, “Integration of multi-source data for water quality classification in the Pearl River estuary and its adjacent coastal water of Hong Kong”, Continental Shelf Research, 23, 1827-1843.
3.Christopher, S. A., D. V. Kliche, J. Chou, and R. M. Welch, 1996, “First estimates of the radiative forcing of aerosols generated from biomass burning using satellite data”, Journal of Geophysical Research, 101, 21265-21273.
4.Cortes, C., V. Vapnik, 1995. “Support vector networks”, Mechine Learning, 20, 273-297, 1995.
5.Durbha S. S., R. L. King, N. H. Younan, 2007 ”Support vector machines regression for retrieval of leaf area index from multiangle imaging spectroradiometer”, Remote Sensing of Environment, 107, 348-361.
6.Fraser, R. S., 1976, “Satellite measurement of mass of Sahaara dust in the atmosphere”. Applied Optics, 15, 2471-2479.
7.Griggs, M., 1975, “Measurement of atmospheric aerosol optical thickness over water using ERTS-1 date”, Journal of the Air Pollution Control Association, 25, 622-626.
8.Gupta, P., S. A. Christopher, J. Wang, R. Gehrig, Y. Lee and N. Kumar, 2006, “Satellite remote sensing of particulate matter and air quality assessment over global cities”, Atmospheric Environment, 40, 5880-5892.
9.Houghton, J. T., L. G. Meira Filho, B. A. Callander, N. Harris, A. Kattenberg, and K. Maskell, Eds., 1996, “Climate change 1995: The science of climate change”, Intergovernmental Panel on Climate Change, Cambridge University Press, 572pp.
10.Husar, R. B., J. M. Prospero, and L. L. Stowe, 1997, “Characterization of tropospheric aerosols over the oceans with the NOAA advanced very high resolution radiometer optical thickness operational product”, Journal of Geophysical Research, 102(D14), 16889-16909.
11.Kent, G. S., P. H. Wang, M. P. McCormick, and K. M. Skeens, 1995, “Multiyear stratospheric aerosol and gas experiment II measurements of upper-tropospheric aerosol characteristics”, Journal of Geophysical Research, 100, 13875-13899.
12.McCormick, M. P., L. W. Thomason, and C. R. Trepte, 1995, “Atmospheric effects of the Mt. Pinatubo eruption”, Nature, 373, 399-404.
13.Mekler, Y., H. Quenzel, G. Ohring, and I. Marcus, 1977, ”Relative atmospheric aerosol content from ERTS observation”, Journal of Geophysical Research 82, 967-972.
14.Quenzel, H., and P. Koepke, 1984, “Tropospheric aerosol optical depth inverted from upwelling radiances”, Aerosols and Their Climatic Impact, H. E.Gerber and A. Deepak, Eds., A. Deepak Publ., 227-240.
15.Schwartz, S. E., F. Arnold, J. P. Blanchet, P. A. Durkee, D. J. Hofmann, W. A. Hoppel, MD. King, A. A. Lacis, T. Nakajima, J. A. Ogren, O. B. Toon and M. Wendisch, 1995, “Group report: Connections between aerosol properties and forcing of climate”, Aerosol Forcing of Climate, R. J. Charlson and J. Heintzenberg, Eds., John Wiley and Sons, 251-280.
16.Sohrabinia, M. and A. M. Khorshiddoust, 2007, “Application of satellite data and GIS in studying air pollutants in Tehran”, Habitat International, 31, 268-275.
17.Takayama, Y., and T. Takashima, 1986, “Aerosol thickness of yellow sand over the Yellow Sea derived from NOAA satellite data”, Atmospheric Environment, 20, 631-638.
18.Tegen, I., A. A. Lacis, and I. Fung, 1996, “The influence on climate forcing of mineral aerosols from disturbed soils” Nature, 380, 419-422.
19.Twomey, S. A., M. Piepgrass, and T. L. Wolfe, 1984, “An assessment of the impact of pollution of global cloud albedo”, Tellus, 36, 356-366.
20.Veefkind, J. P., G. de Leeuw, P. A. Durkee, P. B. Russell, P. V. Hobbs, and J. M. Livingston, 1999, “Aerosol optical depth retrieval using ATSR-2 and AVHRR data during TARFOX”, Journal of Geophysical Research, 104(D2), 2253-2260.
21.Wu, J., A. M. Winer and R. J. Delfino, 2006, “Exposure assessment of particulate matter air pollution before, during, and after the 2003 Southern California wildfires”, Atmospheric Environment, 40, 3333-3348.
32.行政院環境保護署,空氣品質監測網,http://taqm.epa.gov.tw/taqm/ zh-tw/DataHourly.aspx,2009年。
33.Google 地圖,http://Maps.google.com.tw/,2009年。
第一頁 上一頁 下一頁 最後一頁 top
系統版面圖檔 系統版面圖檔