Sea surface temperature inversion model for infrared remote sensing images based on deep neural network

Ai, Bo; Wen, Zhen; Jiang, Yingchao; Gao, Song; Lv, Guannan

doi:10.1016/j.infrared.2019.04.022

Cited by 26 publications

(15 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The statistical results show that the accuracy of water depth retrieval is significantly reduced when the depth is lower than 15 m; the accuracy of retrieval is better when the depth is greater than -10 m. The mean absolute error (MAE) of ZY3 image can reach 0.668; the MAE of WV2 image can reach 0.447. In this article, Numpy is applied to establish a linear regression model [51]; Keras is applied to establish a single-layer neural network model and a CNN model [52]; and then the above three models are applied to perform retrieval experiments on WV2 data. A precision comparison analysis for these three models is shown in the Table V. The results show that the CNN has higher water depth retrieval accuracy than the statistical linear regression model and the single-layer neural network without considering adjacent pixels.…”

Section: Model Resultsmentioning

confidence: 99%

Convolutional Neural Network to Retrieve Water Depth in Marine Shallow Water Area From Remote Sensing Images

Wen

Wang

et al. 2020

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

Self Cite

View full text Add to dashboard Cite

The local connection characteristics of convolutional neural network (CNN) are linked with the local spatial correlation of image pixels for water depth retrieval in this article. The method has greater advantages and higher precision than traditional retrieval methods. Traditional remote sensing empirical models require manual extraction of retrieval factors and the process is complex. This article proposes a model based on CNN, which uses different remote sensing images in four spectral bands, red, green, blue, and near-infrared, to retrieve the water depth. In general, CNN is mostly used for image recognition and classification tasks, which can make full use of the local spatial correlation between pixels. The method in this article exploits this feature of CNN for water depth retrieval, taking into consideration the nonlinear relationship between the radiance value and water depth value from adjacent and central pixels. In this article, remote sensing image data, measured water depth data, and lidar sounding data are used as input data to build the model. Then, the retrieval error is analyzed and the parameters are adjusted to further optimize the model. Quantitative analysis and experimental results show that the accuracy of the CNN model in shallow sea areas retrieval is improved by more than 50%. The mean absolute error can reach within 0.8 m. Finally, the model is shown to be highly portable and capable of retrieving water depth data with resolution equal to the spatial resolution of the remote sensing image using only a small amount of input water depth data.

show abstract

Section: Model Resultsmentioning

confidence: 99%

Convolutional Neural Network to Retrieve Water Depth in Marine Shallow Water Area From Remote Sensing Images

Wen

Wang

et al. 2020

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

Self Cite

View full text Add to dashboard Cite

show abstract

“…Rozenstein et al [6] and Chen et al [7] revised the SWA to use data acquired by the Advanced Very-High-Resolution Radiometer instruments onboard the United States National Oceanic and Atmospheric Administration (NOAA) family of satellites in order to retrieve SSTs from Landsat satellite data. Ai et al [8] presented a new SWA-based SST retrieval model and validated its reliability by comparing SST data for the Bohai Sea with those extracted from a MODIS SST product.…”

Section: Introductionmentioning

confidence: 99%

Long-Term Changes and Factors That Influence Changes in Thermal Discharge from Nuclear Power Plants in Daya Bay, China

Zhang

Wang

Cheng³

et al. 2022

Remote Sensing

View full text Add to dashboard Cite

Thermal discharge (i.e., warm water) from nuclear power plants (NPPs) in Daya Bay, China, was analyzed in this study. To determine temporal and spatial patterns as well as factors affecting thermal discharge, data were acquired by the Landsat series of remote-sensing satellites for the period 1993–2020. First, sea surface temperature (SST) data for waters off NPPs were retrieved from Landsat imagery using a radiative transfer equation in conjunction with a split-window algorithm. Then, retrieved SST data were used to analyze seasonal and interannual changes in areas affected by NPP thermal discharge, as well as the effects of NPP installed capacity, tides, and wind field on the diffusion of thermal discharge. Analysis of interannual changes revealed an increase in SST with an increase in NPP installed capacity, with the area affected by increased drainage outlet temperature increasing to different degrees. Sea surface temperature and NPP installed capacity were significantly linearly related. Both flood tides (peak spring and neap) and ebb tides (peak spring and neap) affected areas of warming zones, with ebb tides having greater effects. The total area of all warming zones in summer was approximately twice that in spring, regardless of whether winds were favorable (i.e., westerly) or adverse (i.e., easterly). The effects of tides on areas of warming zones exceeded those of winds.

show abstract

“…In general, IR radiometers measure SSTs using the brightness temperatures (BTs) of thermal infrared (TIR) bands at 11 μm and 12 μm or using TIR bands combined with the mid-infrared band (e.g., at 3.7 μm for VIIRS) by applying split-window algorithms (e.g., the multichannel SST algorithm (MCSST) [10] and nonlinear SST algorithm (NLSST) [1]), physical models based on radiative transfer model simulations [12], [13], or neural network models [14]. SST retrieval algorithms can be developed for the global ocean [15]- [17] or for regional marine waters [18]- [20].…”

Section: Introductionmentioning

confidence: 99%

Evaluation of Sea Surface Temperatures Derived From the HY-1D Satellite

Liu

Lin

et al. 2022

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Global sea surface temperatures (SSTs) have been detecting by the Chinese ocean color and temperature scanner (COCTS) instruments aboard the HaiYang (HY)-1C and HY-1D satellites. In this study, the SSTs derived from the COCTS instrument on the HY-1D satellite (COCTS/HY-1D) and a nonlinear SST algorithm with corresponding coefficients were introduced. The COCTS/HY-1D SSTs recorded from April 26 to August 31, 2021, were evaluated against water temperature measurements taken at depths above one meter from the in situ Quality Monitor(iQuam) system; root mean square errors (RMSEs) of 0.65 °C and 0.71 °C and robust standard deviations (RSDs) of 0.51 °C and 0.47 °C were obtained for the daytime and nighttime SSTs, respectively, using a spatiotemporal matching window of 4 hours and 2.5 km. Daily gridded SSTs derived from COCTS/HY-1D were compared with those obtained from the visible infrared imaging radiometer (VIIRS) aboard the Suomi National Polar-orbiting Partnership (S-NPP) satellite in the same period, and RMSEs of 0.67±0.06 °C and 0.81±0.06 °C and RSDs of 0.49±0.04 °C and 0.58±0.05 °C were obtained for the daytime and nighttime SSTs, respectively. The COCTS/HY-1D-derived SSTs covering Gulf Stream waters were cross-validated against the VIIRS/S-NPP data as a case study, and RMSEs of 0.53 °C and 0.47 °C for daytime and nighttime, respectively, were obtained. Index Terms-Sea surface temperature (SST); Chinese ocean color and temperature scanner (COCTS); HY-1D satellite; Validation.

show abstract

Sea surface temperature inversion model for infrared remote sensing images based on deep neural network

Cited by 26 publications

References 12 publications

Convolutional Neural Network to Retrieve Water Depth in Marine Shallow Water Area From Remote Sensing Images

Convolutional Neural Network to Retrieve Water Depth in Marine Shallow Water Area From Remote Sensing Images

Long-Term Changes and Factors That Influence Changes in Thermal Discharge from Nuclear Power Plants in Daya Bay, China

Evaluation of Sea Surface Temperatures Derived From the HY-1D Satellite

Contact Info

Product

Resources

About