Remote Sensing of Floating Macroalgae Blooms in the East China Sea Based on UNet Deep Learning Model

Xiaofan Li; Shengqiang Wang; Xuan Weng; Deyong Sun; Hailong Zhang; Hongbo Jiao; Hanwei Liang

doi:10.3788/AOS202141.0201002

Acta Optica Sinica, Volume. 41, Issue 2, 0201002(2021)

Remote Sensing of Floating Macroalgae Blooms in the East China Sea Based on UNet Deep Learning Model

Xiaofan Li¹, Shengqiang Wang^1、*, Xuan Weng¹, Deyong Sun¹, Hailong Zhang¹, Hongbo Jiao⁴, and Hanwei Liang^2,3

Author Affiliations

¹School of Marine Sciences, Nanjing University of Information Science & Technology, Nanjing, Jiangsu 210044, China;

²Key Laboratory of Coastal Zone Exploitation and Protection, Ministry of Natural Resources, Nanjing, Jiangsu 210024, China

³School of Geographic Sciences, Nanjing University of Information Science & Technology, Nanjing, Jiangsu 210044, China;

⁴National Marine Data and Information Service, Tianjin 300171, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(7)

Fig. 1. Specific process of establishing deep learning model for floating macroalgae bloom extraction

Download full size

Fig. 2. GOCI pseudo-color synthetic images for different regions, and extracted results of floating macroalgae blooms using UNet based deep learning model. (a)(c) GOCI pseudo-color synthetic images for different regions; (b)(d) extracted results of floating macroalgae blooms using UNet based deep learning model

Download full size

Fig. 3. GOCI pseudo-color synthetic images for different regions, and distributions of floating macroalgae blooms obtained by UNet based deep learning model, NDVI method, and AFAI method. (a)(e) GOCI pseudo-color synthetic images for different regions; (b)(f) distributions of floating macroalgae blooms obtained by UNet based deep learning model; (c)(g) distributions of floating macroalgae blooms obtained by NDVI method; (d)(h) distributions of floating macroalgae blooms obtained by AFAI method

Download full size

Fig. 4. GOCI pseudo-color synthetic image on 24 March, 2015 and distributions of floating macroalgae blooms obtained by UNet based deep learning model, NDVI method, and AFAI method. (a) GOCI pseudo-color synthetic image on 24 March, 2015; (b) distribution of floating macroalgae blooms obtained by UNet based deep learning model; (c) distribution of floating macroalgae blooms obtained by NDVI method; (d) distribution of floating macroalgae blooms obtained by AFAI method

Download full size

Fig. 5. Distributions of floating macroalgae blooms near the East China Sea obtained by UNet based deep learning model. (a) 3 April, 2017; (b) 19 April, 2017; (c) 18 May, 2017; (d) 28 May, 2017

Download full size

Table 1. Confusion matrix
View table
Table 1. Confusion matrix
Predicted value True value
True False
Positive value N_TP N_FP
Negative value N_TN N_FN

Table 2. Distribution results of floating macroalgae blooms obtained by UNet based deep learning model, NDVI method, and AFAI method

View table

Table 2. Distribution results of floating macroalgae blooms obtained by UNet based deep learning model, NDVI method, and AFAI method

Method	Pixel number of floatingmacroalgae blooms	A_sar /km²	Pixel number of floatingmacroalgae blooms in R1	A_sarin R1 /km²	F₁ /%
UNet based deeplearning model	19898	4974.50	5460	1365.00	86.09
NDVI	16526	4131.50	5224	1306.00	79.38
AFAI	15463	3865.75	4901	1225.25	79.78
Note: A_sar represents distribution area of floating macroalgae blooms.

Tools

Get Citation

Copy Citation Text

Xiaofan Li, Shengqiang Wang, Xuan Weng, Deyong Sun, Hailong Zhang, Hongbo Jiao, Hanwei Liang. Remote Sensing of Floating Macroalgae Blooms in the East China Sea Based on UNet Deep Learning Model[J]. Acta Optica Sinica, 2021, 41(2): 0201002

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites