28 lines
664 B
Markdown
28 lines
664 B
Markdown
|
```csharp title="C#"
|
||
|
|
using Kreuzberg;
|
||
|
|
|
||
|
|
var config = new ExtractionConfig
|
||
|
|
{
|
||
|
|
Ocr = new OcrConfig
|
||
|
|
{
|
||
|
|
TesseractConfig = new TesseractConfig
|
||
|
|
{
|
||
|
|
Preprocessing = new ImagePreprocessingConfig
|
||
|
|
{
|
||
|
|
TargetDpi = 300,
|
||
|
|
Denoise = true,
|
||
|
|
Deskew = true,
|
||
|
|
ContrastEnhance = true,
|
||
|
|
BinarizationMethod = "otsu"
|
||
|
|
}
|
||
|
|
}
|
||
|
|
}
|
||
|
|
};
|
||
|
|
|
||
|
|
var result = KreuzbergLib.ExtractFileSync("scanned.pdf", null, config);
|
||
|
|
|
||
|
|
string content = result.Content;
|
||
|
|
string preview = content.Length > 100 ? content[..100] : content;
|
||
|
|
Console.WriteLine($"Content: {preview}");
|
||
|
|
```
|