通过Emgu实现对图片上的数字进行识别。
前期步骤:
1.下载Emgu安装文件,我的版本是2.4.2.1777。3.0版本则实现对中文的支持。
2.安装后需填写环境变量,环境变量Path值后加入Emgu安装路径到bin下。如C:\Emgu\emgucv-windows-x86-gpu 2.4.2.1777\bin;
3.在bin下查找需要的dll如Emgu.CV.dll与Emgu.CV.OCR.dll等。
4.将C:\Emgu\emgucv-windows-x86-gpu 2.4.2.1777\bin下的文件夹tessdata赋值到程序运行目录下。
注:安装后的Emgu路径下有C#版本的demo可供参考
关键代码:
将需要的dll导入到项目中。
[C#] 纯文本查看 复制代码 private static Tesseract _ocr;//创建识别对象
[C#] 纯文本查看 复制代码 //传入图片进行识别
public static string ORC_(Bitmap img)
{
//""标示OCR识别调用失败
string re = "";
if (img == null)
return re;
else
{
Bgr drawColor = new Bgr(Color.Blue);
try
{
Image<Bgr, Byte> image = new Image<Bgr, byte>(img);
using (Image<Gray, byte> gray = image.Convert<Gray, Byte>())
{
_ocr.Recognize(gray);
Tesseract.Charactor[] charactors = _ocr.GetCharactors();
foreach (Tesseract.Charactor c in charactors)
{
image.Draw(c.Region, drawColor, 1);
}
re = _ocr.GetText();
}
return re;
}
catch (Exception ex)
{
return re;
}
}
}
//识别方法如点击按钮识别
private void btnXIdentification_Click(object sender, EventArgs e)
{
try
{
_ocr = new Tesseract(@"C:\Emgu\emgucv-windows-x86-gpu 2.4.2.1777\bin\tessdata", "eng", Tesseract.OcrEngineMode.OEM_TESSERACT_CUBE_COMBINED);//方法第一个参数可为""表示通过环境变量调用字库,第二个参数表示字库的文件,第三个表示识别方式,可看文档与资料查找。
_ocr.SetVariable("tessedit_char_whitelist", "0123456789X");//此方法表示只识别1234567890与x字母
string result = "";
Bitmap bitmap = new Bitmap(_emguImage.ToBitmap());
bitmap = BrightnessP(bitmap, Convert.ToInt32(this.textBoxX3.Text));//图片加亮处理
bitmap = KiContrast(bitmap, Convert.ToInt32(this.textBoxX2.Text));//调整对比对
this.pictureBox3.Image = bitmap;
result = ORC_(bitmap);
this.textBoxX1.Text = result;
_ocr.Dispose();
}
catch (Exception exception)
{
MessageBox.Show(exception.Message);
}
}
/// <summary>
/// 增加图像亮度
/// </summary>
/// <param name="a"></param>
/// <param name="v"></param>
/// <returns></returns>
public static Bitmap BrightnessP(Bitmap a, int v)
{
System.Drawing.Imaging.BitmapData bmpData = a.LockBits(new Rectangle(0, 0, a.Width, a.Height), System.Drawing.Imaging.ImageLockMode.ReadWrite, System.Drawing.Imaging.PixelFormat.Format24bppRgb);
int bytes = a.Width * a.Height * 3;
IntPtr ptr = bmpData.Scan0;
int stride = bmpData.Stride;
unsafe
{
byte* p = (byte*)ptr;
int temp;
for (int j = 0; j < a.Height; j++)
{
for (int i = 0; i < a.Width * 3; i++, p++)
{
temp = (int)(p[0] + v);
temp = (temp > 255) ? 255 : temp < 0 ? 0 : temp;
p[0] = (byte)temp;
}
p += stride - a.Width * 3;
}
}
a.UnlockBits(bmpData);
return a;
}
///<summary>
///图像对比度调整
///</summary>
///<param name="b">原始图</param>
///<param name="degree">对比度[-100, 100]</param>
///<returns></returns>
public static Bitmap KiContrast(Bitmap b, int degree)
{
if (b == null)
{
return null;
}
if (degree < -100) degree = -100;
if (degree > 100) degree = 100;
try
{
double pixel = 0;
double contrast = (100.0 + degree) / 100.0;
contrast *= contrast;
int width = b.Width;
int height = b.Height;
BitmapData data = b.LockBits(new Rectangle(0, 0, width, height), ImageLockMode.ReadWrite, PixelFormat.Format24bppRgb);
unsafe
{
byte* p = (byte*)data.Scan0;
int offset = data.Stride - width * 3;
for (int y = 0; y < height; y++)
{
for (int x = 0; x < width; x++)
{
// 处理指定位置像素的对比度
for (int i = 0; i < 3; i++)
{
pixel = ((p / 255.0 - 0.5) * contrast + 0.5) * 255;
if (pixel < 0) pixel = 0;
if (pixel > 255) pixel = 255;
p = (byte)pixel;
} // i
p += 3;
} // x
p += offset;
} // y
}
b.UnlockBits(data);
return b;
}
catch (Exception ex)
{
return null;
}
}
目前我只是识别文字与字幕,3.0版本虽然可以识别中文,但误读率实在不敢恭维。
本文如有错误之处请提出,灰常感谢!
|