Nomad changes

docs/snippets/php/ocr/image_preprocessing.md

@@ -0,0 +1,36 @@
+```php title="PHP"
+<?php
+declare(strict_types=1);
+
+require_once __DIR__ . '/vendor/autoload.php';
+
+use Kreuzberg\Kreuzberg;
+use Kreuzberg\Config\ExtractionConfig;
+use Kreuzberg\Config\OcrConfig;
+use Kreuzberg\Config\ImagePreprocessingConfig;
+
+// Enhance OCR accuracy with image preprocessing
+$config = new ExtractionConfig(
+    ocr: new OcrConfig(
+        backend: 'tesseract',
+        language: 'eng',
+        imagePreprocessing: new ImagePreprocessingConfig(
+            targetDpi: 300,
+            autoRotate: true,
+            deskew: true,
+            denoise: true,
+            contrastEnhance: true,
+            binarizationMethod: 'otsu',
+            invertColors: false
+        )
+    )
+);
+
+$kreuzberg = new Kreuzberg($config);
+$result = $kreuzberg->extractFile('scanned_document.pdf');
+
+echo "Preprocessed OCR Results:\n";
+echo "Characters extracted: " . strlen($result->content) . "\n";
+echo "Preview: " . substr($result->content, 0, 300) . "...\n";
+?>
+```