how can i read text from scanned document in c#?

Because scanning always results in an image, what you need is Optical Character Recognition (OCR) in order to extract the text. Note that the difficulty of this depends on the characters you want to read and how clear and sharp-looking they are. In general, it is more likely to succeed if the image is clear, high resolution and high quality (in focus). To achieve this, try to use a high DPI and scan images with text in grayscale or black and white.

There are free OCR SDKs you can try like MODI, GOCR, and Tesseract. Other answers here have links to Tesseract and MODI. For GOCR, see here: http://jocr.sourceforge.net/index.html

Another option is to use a professional library like LEADTOOLS (https://www.leadtools.com/sdk/recognition-imaging) that has OCR features. (Disclaimer: I work for the company that created this library).
The code to load a PDF (eg: a scanned PDF) as an image and OCR it looks like this:

IOcrEngine _ocrEngine;

_ocrEngine = OcrEngineManager.CreateEngine(OcrEngineType.Advantage, false);

_ocrEngine.Startup(null, null, null, null);

IOcrDocument ocrDoc = _ocrEngine.DocumentManager.CreateDocument();

ocrDoc.Pages.AddPage(pdfFileName, null);

ocrDoc.Pages[0].Recognize(null);

// Save the result to a memory stream as plain text

System.IO.MemoryStream OcrResultStream = new System.IO.MemoryStream();

ocrDoc.Save(OcrResultStream, Leadtools.Forms.DocumentWriters.DocumentFormat.Text, null);

string txtResult = Encoding.ASCII.GetString(OcrResultStream.ToArray());

_ocrEngine.Shutdown();

The toolkit also has features for processing and cleaning document images to improve recognition results.