索贝尔算子(Sobel operator)是图像处理中的算子之一，主要用作边缘检测。在技术上，它是一离散性差分算子，用来运算图像亮度函数的梯度之近似值。在图像的任何一点使用此算子，将会产生对应的梯度矢量或是其法矢量

Sobel算子有两个，一个是检测水平边沿的；另一个是检测垂直平边沿的。Sobel算子另一种形式是各向同性Sobel (Isotropic Sobel)算子，也有两个，一个是检测水平边沿的，另一个是检测垂直边沿的各向同性Sobel算子和普通Sobel算子相比，它的位置加权系数更为准确，在检测不同方向的边沿时梯度的幅度一致。由于建筑物图像的特殊性，我们可以发现，处理该类型图像轮廓时，并不需要对梯度方向进行运算，所以程序并没有给出各向同性Sobel算子的处理方法。由于Sobel算子是滤波算子的形式，用于提取边缘，可以利用快速卷积函数，简单有效，因此应用广泛。

-Sobel operator (Sobel operator) operator is one of image processing, mainly for edge detection. Technically, it is a discrete difference operator for the operation gradient approximation image brightness function. At any point using this image operator, will have its corresponding gradient vector or vector method Sobel operator has two, one is to detect the horizontal edge　the other is to detect vertical flat edge. Sobel operator is another form of isotropic Sobel (Isotropic Sobel) operator, there are two, one is to detect the horizontal edge, and the other vertical edge detection isotropic Sobel operator Sobel operator and ordinary compared to its location more accurate weighting factor, consistent detect edges in different directions when the magnitude of the gradient. Due to the special nature of the building of the image, we can see that, when dealing with this type of image contour, gradient direction does not need to be operational, so the program does not give an isotropic Sobe