图像识别学习—3

Robets 算子是一种微分算子

设输入图像为f(i,j) 算子输出图像为 g(i,j)

则Roberts算子表示为差分形式:

g(x) = f(i,j)-f(i+1,j)

g(y) = f(i+1,j) - f(i,j+1)

g(i,j) = abs(gx)+ abs(gy)

这是一个Winform的代码

private void button1_Click(object sender, EventArgs e)
{
Color c1
= new Color();
Color c2
= new Color();
Color c3
= new Color();
Color c4
= new Color();
int rr, gg, bb, r1, r2, r3, r4, fxr, fyr, i, j;
int g1, g2, g3, g4, fxg, fyg, b1, b2, b3, b4, fxb, fyb;

Bitmap box1
= new Bitmap(pictureBox1.Image);
Bitmap box2
= new Bitmap(pictureBox2.Image);

for (i = 0; i < pictureBox1.Image.Width - 2; i++)
{
for (j = 0; j < pictureBox1.Image.Height - 2; j++)
{
c1
= box1.GetPixel(i, j);
c2
= box1.GetPixel(i + 1, j + 1);
c3
= box1.GetPixel(i + 1, j);
c4
= box1.GetPixel(i, j + 1);
r1
= c1.R;
r2
= c2.R;
r3
= c3.R;
r4
= c4.R;
fxr
= r1 - r2;
fyr
= r3 - r4;
rr
= Math.Abs(fxr) + Math.Abs(fyr) + 128;
if (rr < 0)
rr
= 0;
if (rr > 255)
rr
= 255;
g1
= c1.G;
g2
= c2.G;
g3
= c3.G;
g4
= c4.G;
fxg
= g1 - g2;
fyg
= g3 - g4;
gg
= Math.Abs(fxg) + Math.Abs(fyg) + 128;
if (gg < 0)
gg
= 0;
if (gg > 255)
gg
= 255;
b1
= c1.B;
b2
= c2.B;
b3
= c3.B;
b4
= c4.B;
fxb
= b1 - b2;
fyb
= b3 - b4;
bb
= Math.Abs(fxb) + Math.Abs(fyb);
if (bb < 0)
bb
= 0;
if (bb > 255)
bb
= 255;

Color cc
= Color.FromArgb(rr, gg, bb);
box2.SetPixel(i, j, cc);
}

pictureBox2.Refresh();
pictureBox2.Image
= box2;
}

 

posted @ 2010-07-29 18:27  向恺然  阅读(1135)  评论(0编辑  收藏  举报

我必须说的是:我崇尚开源,但鄙视剽窃。本博客所有引用的图片,文章,和代码,均只作为研究学习使用,不作为商业应用。如果我无意中冒犯了您,请发消息留言,我将立即删除。