| Summary: | 碩士 === 國立臺北科技大學 === 機械工程系機電整合碩士班 === 107 === The purpose of this study was to detect eutrophication of lake water by integrating the mobility of UAVs with multi-spectrometers. In order to obtain the accuracy of the spectral image, the vignetting correction and relative radiometric correction must be done before analyzing the spectral image. According to the Carlson trophic state index (CTSI), the Chlorophyll-a, total Phosphate and transparency were the three main parts in eutrophication. Moreover, dissolved oxygen would be affected by the different level of the eutrophication. Thus, this study also considered and included the detection of dissolved oxygen. The regression analysis was used to analyze the different reflectance of different matter in each band and the sampling data. In this case, the correlation between reflectance and the sampling data would be found and further built the regression models. The results of the experiment showed that the coefficient of determination of the four water quality parameters has a significant correlation, and the transparency has a high correlation in the red、green and NIR bands (R^2= 0.9598). This study may contribute on finding a rapid and non-contact method to detect the water quality in the future.
|
|---|
