[C#] OCR , tessnet2

참고

1. http://www.pixel-technology.com/freeware/tessnet2/

 

2. https://github.com/tesseract-ocr

 

 

 

 

1번 다운로드 후

 

실행

 

 

 

문자 입력은

 

tessdata 다운로드 해야함

 

 

public partial class Form1 : Form
    {
        public Form1()
        {
            InitializeComponent();
        }

        private void button1_Click(object sender, EventArgs e)
        {           

            var image = new Bitmap(@"d:\0.png");
            var ocr = new tessnet2.Tesseract();
            //ocr.SetVariable("tessedit_char_whitelist", "0123456789"); // If digit only
            //@"C:\OCRTest\tessdata" contains the language package, without this the method crash and app breaks
            ocr.Init(@"C:\tessdata", "eng", true);
            var result = ocr.DoOCR(image, Rectangle.Empty);
            foreach (tessnet2.Word word in result)
            {
                //Console.WriteLine("{0} : {1}", word.Confidence, word.Text);
                label1.Text += string.Format("{0} : {1}", word.Confidence, word.Text);
            }
        }

    }

    public class Ocr
    {
        public void DumpResult(List<tessnet2.Word> result)
        {
            foreach (tessnet2.Word word in result)
                Console.WriteLine("{0} : {1}", word.Confidence, word.Text);
        }

        public List<tessnet2.Word> DoOCRNormal(Bitmap image, string lang)
        {
            tessnet2.Tesseract ocr = new tessnet2.Tesseract();
            ocr.Init(null, lang, false);
            List<tessnet2.Word> result = ocr.DoOCR(image, Rectangle.Empty);
            DumpResult(result);
            return result;
        }

        ManualResetEvent m_event;

        public void DoOCRMultiThred(Bitmap image, string lang)
        {
            tessnet2.Tesseract ocr = new tessnet2.Tesseract();
            ocr.Init(null, lang, false);
            // If the OcrDone delegate is not null then this'll be the multithreaded version
            ocr.OcrDone = new tessnet2.Tesseract.OcrDoneHandler(Finished);
            // For event to work, must use the multithreaded version
            ocr.ProgressEvent += new tessnet2.Tesseract.ProgressHandler(ocr_ProgressEvent);
            m_event = new ManualResetEvent(false);
            ocr.DoOCR(image, Rectangle.Empty);
            // Wait here it's finished
            m_event.WaitOne();
        }

        public void Finished(List<tessnet2.Word> result)
        {
            DumpResult(result);
            m_event.Set();
        }

        void ocr_ProgressEvent(int percent)
        {
            Console.WriteLine("{0}% progression", percent);
        }
    }