隨影像式照度量測技術在水下光學領域日益重要，本文擬結合數位影像處理、主成分分析及回歸分析，發展視覺化三維水下照度估測技術，以利水下照度量測。首先，吾人定義水下為實驗環境，使用商用發光二極體燈具為量測光源，以感光耦合元件擷取受到光源所照射之平面影像。第二，經數位影像處理，可得三個影像特徵值，以主成分分析方法降低變數維度，得到主成分值。同時，使用照度感測器進行受照面上的陣列式照度量測。最後，透過回歸分析，獲得主成分值與照度之間的關係。實驗結果顯示，本文所設計之視覺化三維水下照度估測技術，不僅估測照度值與量測照度值進行正規化交互相關性比較，相似性可達97 %，且可藉此推廣至不同光型、不同距光源之長度，讓水下量測照度的時間由數十分鐘縮短到數十秒，提升照度量測的效率，改善傳統量測耗時的缺點。而估測過程中，只需要感光耦合元件，即可獲得水下環境的照度分布，使照度量測技術可普及且不拘受照面環境。

Image illuminance measuring technology increasingly applies in scientific research on underwater optics. Based on the digital image processing, principal component analysis (PCA) and regression analysis, a novel 3-D visualization illuminance estimation method could improve underwater illuminance measurement technology examined in this study. First, the study defines the underwater experimental environment and uses commercial underwater LED, then captures an image of the illuminated area by a CCD. Second, digital images are processed, and three image key factors obtained. The study substitutes these three key factors into principal component analysis to reduce the variable dimension, and the principal component value obtained. Moreover, the study positions an illuminance meter manually to complete the array measurement of the illuminated area. Finally, the study conducts regression analysis to determine the relationship between principal component and illuminance. The experimental results indicate that the similarity between the estimated and measured values can reach 97% which calculated by NCC, and there has the ability to analyze different light types and distances by this system; therefore, compared with traditional methods, adopting the new method reduces the measurement time to a few seconds and enhanced measurement efficiency. Thus, using the 3-D visualization illuminance estimation method enables users to easily measure and capture the environmental illuminance, thereby enhancing the availability of such information.

[1] Andersen, M., De Boer, J.(2006), “Goniophotometry and assessment of bidirectional photometric properties of complex fenestration systems,” Energy and Buildings, Vol.38, No.7, pp.836~848.
[2] Aydin, T. O., Mantiuk, R., Seidel, H. P.(2008), “Extending Quality Metrics to Full Luminance Range Images,” Proceedings of SPIE - The International Society for Optical Engineering, Vol.6806, No.68060B, pp.1~10.
[3] Bellia, L., Cesarano, A., Minichiello, F., Sibilio, S.(2002), “Setting up a CCD photometer for lighting research and design,” Building and Environment, Vol.37, No.11, pp.1099~1106.
[4] Chang, K. C., Tian, Y. P., Shih, H. C.(2012), “Using Multi-temporal and PCA+NDVI to Improve the Accuracy and Integrity of Land Cover Classification,” Journal of Geographical Research, Vol.57, No.3, pp.41~60.
[5] Chen, W. X.(2012), The pyramid image segmentation method according to metric measurement technology in vision research, Master Thesis, Department of Systems and Naval Mechatronic Engineering, National Cheng Kung University.
[6] Degro, J.(2007), “School Experiments with a Lux Meter,” Physics Education, Vol.42, No.3, pp.275~280.
[7] Ekrias, A. , Eloholma, M., Halonen, L., Song, X. J., Zhang, X., Wen, Y.(2007), “Road lighting and headlights: Luminance measurements and automobile lighting simulations,” Building and Environment, Vol.43, No.4, pp.530~536.
[8] Gonzalez, R. C., Woods, R. E. (2007), Digital Image Processing(3rd ed.), Pearson Education.
[9] Gonzalez, R. C., Woods, R. E., Eddins, S. L. (2009), Digital Image Processing Using MATLAB(2nd ed.), Gatesmark Publishing.
[10] Huang, H. M., Su, K. H.(2008), “Based on Principal Component Analysis and Support Vector Regression to Prediction of House Pricing,” Journal of Tung Fang Institute of Technology, Vol.29, pp.155~162.
[11] Huang, H. J.(2012), A novel LED lighting module using patternable design concept for underwater illumination, Master Thesis, Department of Systems and Naval Mechatronic Engineering, National Cheng Kung University.
[12] İsmail, K., Vedat, T., Bülent, O.(2011), “Modeling of dimmable High Power LED illumination distribution using ANFIS on the isolated area,” Expert Systems with Applications, Vol.38, No.9, pp.11843~11848.
[13] Jen, C. H., Chen, Y. Y., Whang Allen, J. W., Lu, M. J.(2011), “Non- axisymmetrical freeform design for short LED street lamp,” Proceedings of SPIE - The International Society for Optical Engineering, Vol.8123, No.812307, pp.1~7.
[14] Kutner, M. H., Neter, J., Nachtsheim, C. J.(2004), Applied Linear Regression Models(5th ed.), McGraw Hill.
[15] KONICA MINOLTA SENSING, INC.(2012), Illuminance Meter T-10A / T-10MA Instruction manual, http://sensing.konicaminolta.asia/products/.
[16] Lewis, J. P.(1995), “Fast Template Matching,” Vision Interface, Canadian Image Processing and Pattern Recognition Society, Quebec City, Canada, Vol.10, No.1, pp.120~123.
[17] Liu, C. M., Fairchild, M. D.(2014), “Measuring the Relationship between Perceived Image contrast and Surround Illumination,” Color and Imaging Conference, 12th Color and Imaging Conference Final Program and Proceedings, Vol.12, No.7, pp.282~288.
[18] Pearson, K.(1901), “On lines and planes of closest fit to systems of points in space,” Philosophical Magazine, Vol.2, No.6, pp.559~572.
[19] Rossini, E. G., Krenzinger, A.(2007), “Maps of sky relative radiance and luminance distributions acquired with a monochromatic CCD camera,” Solar Energy, Vol.81, No.11, pp.1323~1332.
[20] Shlens, J.(2005), A Tutorial on Principal Component Analysis, Systems Neurobiology Laboratory, Salk Institute for Biological Studies La Jolla, CA92037 and Institute for Nonlinear Science, University of California, San Diego La Jolla, CA 92093-0402.
[21] Sun, C. C., Lee, T. X., Ma, S. H., Lee, Y. L., Huang, S. M.(2006), “Precise optical modeling for LED lighting verified by cross correlation in the midfield region,” Optics Letters, Vol.31, No.14, pp.2193~2195.
[22] Son, D. K., Cho, E. B., Moon, I., Park, Y., and Lee, C. G.(2011), “Development of an Illumination Measurement Device for Color Distribution Based on a CIE 1931 XYZ Sensor,” Journal of the Optical Society of Korea, Vol.15, No.1, pp.44~51.
[23] Tsai, D. M., Lin, C. T.(2003), “Fast normalized cross correlation for defect detection,” Pattern Recognition Letters, Vol.24, No.15, pp.2625~2631.
[24] Wang, L.(2006), Software calibration method of light-match screen for detection of photometric characteristics of motorcycle headlights using photogrammetry, Master Thesis, Institute of Computer Application Technology, Tianjin University.
[25] Wüllera, D., Gabeleb, H.(2007), “The usage of digital cameras as luminance meters,” Proceedings of SPIE - Digital Photography III, Vol.6502, No.65020U, pp.1~11.
[26] Wang, K., Luo, X., Liu, Z., Zhou, B., Gan, Z., Liu, S.(2008), “Optical analysis of an 80-W light-emitting-diode street lamp,” Optical Engineering, Vol.47, No.013002, pp.1~7.
[27] Wu, Z. Y.(2009), Using the principal component analysis (PCA) for the study of three-dimensional face recognition, Master Thesis, Department of Environment Information Technology Master Program, Feng Chia University.
[28] Wang, K., Liu, S., Chen, F., Qin, Z., Liu, Z. Y., Luo, X. B.(2009), “Freeform LED lens for rectangularly prescribed illumination,” Journal of Optics A: Pure and Applied Optics, Vol.11, No.105501, pp.1~10.
[29] Wang, J., Zhu, J. H., Feng, H., Xiao, J.(2013), “Based on CCD and MATLAB Isolux line drawing method,” China Illuminating Engineering Journal, Vol.24, No.4, pp.40~43.
[30] Zhang, Y. T., Gong, L.(2007), The principles and techniques of data mining, Wunan Book Co., Ltd.
[31] Zhou, Y. J., Zhao, B. Q., Wang, P.(2010), “CCD Imaging Study Type Luminance Meter Measurement Methods,” Electronics Optics & Control, Vol.17, No.20, pp.49~52.