Fig 2 - available via license: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International
Content may be subject to copyright.
The motion features of two imageries. (a) is 30 April 2018 at 02:00(UTC). (b) is 30 April 2018 at 03:00(UTC). (c) is the motion features of the imagery (b). (d) is the area A in (c), this sea fog area has a smaller displacement. (e) is the area B in (c), this cloud area has a greater displacement. The length and direction of the arrow represent the moving distance and direction of the pixel block. When the displacement is too small, arrows exist as dots in the figure.
Source publication
Sea fog detection is a significant and challenging issue in meteorological satellite imagery. Distinguishing between sea fog and low clouds is challenging due to the similar morphology and brightness characteristics of these two phenomena on the imageries. Most of the existing deep learning methods are based on a single imagery feature extraction w...
Contexts in source publication
Context 1
... we calculate the motion features of the current imagery and the previous N imageries. The motion feature between two consecutive satellite imageries is shown in Figure 2. Next generate an attention map of the motion features and feed it into the neural network along with the raw multi-channel satellite data. ...
Context 2
... each region block with the optical flow field E(u, v), a two-dimensional coordinate system is established with the center point as the origin, and the displacements in the x-axis and y-axis directions are respectively represented by u and v. In Figure 2, arrows with different lengths and directions indicate the movement direction and distance of the region blocks between adjacent frames. When the displacement is too small, the arrows exist as dots in the figure. ...
