使用C#版Tesseract库
使用C#版Tesseract库
上一篇介绍了Tesseract库的使用(OCR库Tesseract初探),文末提到了Tesseract是用c/c++开发的,也有C#的开源版本,本篇介绍一下如何使用C#版的Tesseract。
C#版本源码下载地址:https://github.com/charlesw/tesseract
其实在vs中可以直接用NuGet工具进行下载:
打开nuget,搜索tesseract,点安装即可。
源码是vs2015编译的,需要安装vs2015以上版本。
打开项目后如:
我们再添加一个winform项目,画界面如:
实现点击“选择需要识别的图片”,打开一张图片,调用算法并显示结果。比较简单。源码如下:
using System; using System.Collections.Generic; using System.ComponentModel; using System.Data; using System.Drawing; using System.Linq; using System.Text; using System.Threading.Tasks; using System.Windows.Forms; using Tesseract; namespace TesseractDemo { public partial class Form1 : Form { public Form1() { InitializeComponent(); } //选图片并调用ocr识别方法 private void btnRec_Click(object sender, EventArgs e) { //openFileDialog1.Filter = ""; if (openFileDialog1.ShowDialog() == DialogResult.OK) { var imgPath = openFileDialog1.FileName; pictureBox1.Image=Image.FromFile(imgPath); string strResult = ImageToText(imgPath); if (string.IsNullOrEmpty(strResult)) { txtResult.Text = "无法识别"; } else { txtResult.Text = strResult; } } } //调用tesseract实现OCR识别 public string ImageToText(string imgPath) { using (var engine = new TesseractEngine("tessdata", "eng", EngineMode.Default)) { using (var img = Pix.LoadFromFile(imgPath)) { using (var page = engine.Process(img)) { return page.GetText(); } } } } } }
有一点要注意的是,tesseract的识别语言包要自己下载后包含到项目里面,并设置为始终复制,或者直接把这个文件包放到运行程序目录(bin\debug)下:
eng是英文字符的意思,要识别其他语言字符,需要自己下载:
Tesseract has unicode (UTF-8) support, and can recognize more than 100 languages "out of the box".
这个库支持100种语言的识别
字库下载地址为:https://github.com/tesseract-ocr/tessdata
用OpencvSharp先降噪再调OCR识别:
//用opencv进行降噪处理再ocr识别 private void button3_Click(object sender, EventArgs e) { //从网上读取一张图片 string imgUrl = "https://service.cheshi.com/user/validate/validatev3.php"; MemoryStream ms = ReadImgFromWeb(imgUrl); Image img = Image.FromStream(ms); pictureBox1.Image = img; //降噪 Mat simg = Mat.FromStream(ms, ImreadModes.Grayscale); Cv2.ImShow("Input Image", simg); //阈值操作 阈值参数可以用一些可视化工具来调试得到 Mat ThresholdImg = simg.Threshold(29, 255, ThresholdTypes.Binary); Cv2.ImShow("Threshold", ThresholdImg); Cv2.ImWrite("d:\\img.png", ThresholdImg); textBox1.Text= ImageToText("d:\\img.png"); } /// <summary> /// 从网上读取一张图片 /// </summary> /// <param name="Url"></param> public MemoryStream ReadImgFromWeb(string Url) { HttpWebRequest request = (HttpWebRequest)WebRequest.Create(Url); request.Credentials = CredentialCache.DefaultCredentials; // 添加授权证书 request.UserAgent = "Microsoft Internet Explorer"; WebResponse response = request.GetResponse(); Stream s = response.GetResponseStream(); byte[] data = new byte[1024]; int length = 0; MemoryStream ms = new MemoryStream(); while ((length = s.Read(data, 0, data.Length)) > 0) { ms.Write(data, 0, length); } ms.Seek(0, SeekOrigin.Begin); //pictureBox1.Image = Image.FromStream(ms); return ms; }
请自行用NuGet程序下载opencvsharp3.0库,参考https://www.cnblogs.com/tuyile006/p/10819570.html
另外专门有篇文章介绍中文识别:Tesseract-OCR识别中文与训练字库实例