在本文中,我们提出了一个用于自动增强和识别活动海断层的框架,通常使用测深图像与地震活动相关联。 据我们所知,基于图像分析方法的海断层自动识别问题尚属首次。 首先,我们计算 Slope 和 Aspect 图像以及它们的导数,然后我们有效地结合旋转和尺度不变滤波器和像素标记方法,提供对海断层的稳健检测。 近海真实地形图像的实验结果及其与近期地震活动的相关性表明所提出方案的可靠性能。
% Computes the slope (Slope and Aspect (Asp) array according to arcGIS method
% Slope \in [0,90] degrees
% Aspect \in [0, 360] degrees
function [Slope] = getSlopeOfAspect(Data,x_cell_size,y_cell_size)
lines = size(Data,1);
cols = size(Data,2);
Slope = zeros(lines,cols);
for i1=2:lines-1,
for i2=2:cols-2,
n = Data(i1-1:i1+1,i2-1:i2+1);
a = n(1,1);
b = n(1,2);
c = n(1,3);
d = n(2,1);
%e = n(2,2);
f = n(2,3);
g = n(3,1);
h = n(3,2);
i = n(3,3);
dz_dx = ((c + 2*f + i) - (a + 2*d + g));
dz_dy = ((g + 2*h + i) - (a + 2*b + c));
dz_dx = abs(rem(dz_dx,360));
dz_dy = abs(rem(dz_dy,360));
if dz_dx > 180,
dz_dx = abs(dz_dx-360);
end
if dz_dy > 180,
dz_dy = abs(dz_dy-360);
end
dz_dx = dz_dx / (8*x_cell_size);
dz_dy = dz_dy / (8*y_cell_size);
rise_run = sqrt(dz_dx^2 + dz_dy^2);
slope_degrees = atan(rise_run)*57.29578;
Slope(i1,i2) = slope_degrees;
end
end
[1] Panagiotakis C , Kokinou E . Automatic Enhancement and Detection of Active Sea Faults from Bathymetry[C]// International Conference on Pattern Recognition. IEEE Computer Society, 2014.
部分理论引用网络文献,若有侵权联系博主删除。