derivate_gauss — Convolve an image with derivatives of the Gaussian.用高斯的导数卷积图像

高斯导数滤波用作 平滑滤波器（使用分水岭watershed得到contours）（用于很多小块的图像）

derivate_gauss(Image : DerivGauss : Sigma, Component : )

用一个图片Image和一个高斯函数的导数求卷积，从而计算出不同的特征值。sigma控制高斯函数，当sigma为一个值时候，行和列的方向上sigma相同，当sigma为两个值时候，第一个控制列的程度，第二个控制行的程度。‘none’这里指Smoothing only

* usage as smoothing filter
read_image (Meningg5, 'meningg5')
derivate_gauss (Meningg5, Smoothed, 2, 'none')
*转换图像类型
convert_image_type (Smoothed, SmoothedByte, 'byte')
watersheds (SmoothedByte, Basins, Watersheds)
dev_display (Meningg5)
dev_set_draw ('margin')
dev_set_colored (6)
dev_display (Basins)
stop ()
* usage as edge detector
read_image (Image, 'mreut')
derivate_gauss (Image, GradientAmpl, 1.5, 'gradient')
threshold (GradientAmpl, Gradient, 10, 255)
dev_display (Image)
dev_set_draw ('fill')
dev_set_color ('red')
dev_display (Gradient)
stop ()
* usage as corner detector
read_image (Image, 'fabrik')
derivate_gauss (Image, Det, 1.5, 'det')
threshold (Det, Corners, 20, 1000000)
dev_display (Image)
dev_set_draw ('fill')
dev_set_color ('red')
dev_display (Corners)
stop ()
* Usage as edge detector
read_image (Image, 'mreut')
derivate_gauss (Image, EdgesAreZero, 3, '2nd_ddg')
zero_crossing (EdgesAreZero, Edges)
dev_display (Image)
dev_set_draw ('fill')
dev_set_color ('red')
dev_display (Edges)