Remote sensing inversion modeling of chlorophyll-a concentration in Wuliangsuhai Lake based on BP neural network. Issue 1 (June 2021)

• Record Type: Journal Article
• Authors: Fu, Xueliang; Liu, Wenyao; Hu, Hua
• Is Part Of: Journal of physics. Volume 1955:Issue 1(2021)
• Publication Date: 2021-06
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/1955/1/012103
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: Monitoring chlorophyll-a concentration in Wuliangsuhai Lake can effectively grasp the primary production of Wuliangsuhai Lake. The traditional manual sampling monitoring method is time-consuming and laborious, compared with the monitoring method, the monitoring range of chlorophyll-a in lake water by remote sensing is not only wide, long time sequence, but also lower than that of the original. Based on the Landsat 8 remote sensing image data of Wuliangsuhai Lake and the measured data of chlorophyll-a concentration sampling points in Wuliangsuhai Lake, the spectral reflectance of each sampling point was extracted through the longitude and latitude coordinates of each sampling point, the first to fifth band spectral reflectance combination of Landsat 8 remote sensing image data is used as input, and the measured chlorophyll-a concentration is used as output, 26 BP neural network models were constructed to Retrieve Chlorophyll-a Concentration in Wuliangsuhai Lake. Through the verification and analysis of all the models, it is found that the inversion model with spectral reflectance of band 1, 2, 3, 4 and 5 as input has the best performance.