This commit is contained in:
Henrik Jess Nielsen
195
docs/snippets/php/ocr/advanced_ocr.php
Normal file
195
docs/snippets/php/ocr/advanced_ocr.php
Normal file
@@ -0,0 +1,195 @@
|
||||
```php title="advanced_ocr.php"
|
||||
<?php
|
||||
|
||||
declare(strict_types=1);
|
||||
|
||||
/**
|
||||
* Advanced OCR Configuration
|
||||
*
|
||||
* Fine-tune OCR performance with Tesseract configuration, image preprocessing,
|
||||
* and page segmentation modes.
|
||||
*/
|
||||
|
||||
require_once __DIR__ . '/vendor/autoload.php';
|
||||
|
||||
use Kreuzberg\Kreuzberg;
|
||||
use Kreuzberg\Config\ExtractionConfig;
|
||||
use Kreuzberg\Config\OcrConfig;
|
||||
use Kreuzberg\Config\TesseractConfig;
|
||||
use Kreuzberg\Config\ImagePreprocessingConfig;
|
||||
|
||||
$config = new ExtractionConfig(
|
||||
ocr: new OcrConfig(
|
||||
backend: 'tesseract',
|
||||
language: 'eng',
|
||||
tesseractConfig: new TesseractConfig(
|
||||
psm: 6,
|
||||
enableTableDetection: true
|
||||
)
|
||||
)
|
||||
);
|
||||
|
||||
$kreuzberg = new Kreuzberg($config);
|
||||
$result = $kreuzberg->extractFile('financial_report_scan.pdf');
|
||||
|
||||
echo "OCR with Table Detection:\n";
|
||||
echo str_repeat('=', 60) . "\n";
|
||||
echo "Tables found: " . count($result->tables) . "\n\n";
|
||||
|
||||
foreach ($result->tables as $index => $table) {
|
||||
echo "Table " . ($index + 1) . ":\n";
|
||||
echo $table->markdown . "\n\n";
|
||||
}
|
||||
|
||||
$invoiceConfig = new ExtractionConfig(
|
||||
ocr: new OcrConfig(
|
||||
backend: 'tesseract',
|
||||
language: 'eng',
|
||||
tesseractConfig: new TesseractConfig(
|
||||
psm: 6,
|
||||
tesseditCharWhitelist: '0123456789.,€$£¥-/'
|
||||
)
|
||||
)
|
||||
);
|
||||
|
||||
$kreuzberg = new Kreuzberg($invoiceConfig);
|
||||
$result = $kreuzberg->extractFile('invoice_scan.pdf');
|
||||
|
||||
echo "Invoice OCR (numbers only):\n";
|
||||
echo str_repeat('=', 60) . "\n";
|
||||
echo $result->content . "\n\n";
|
||||
|
||||
$preprocessedConfig = new ExtractionConfig(
|
||||
ocr: new OcrConfig(
|
||||
backend: 'tesseract',
|
||||
language: 'eng',
|
||||
imagePreprocessing: new ImagePreprocessingConfig(
|
||||
targetDpi: 300,
|
||||
denoise: true,
|
||||
sharpen: true,
|
||||
autoRotate: true,
|
||||
deskew: true
|
||||
),
|
||||
tesseractConfig: new TesseractConfig(
|
||||
psm: 3
|
||||
)
|
||||
)
|
||||
);
|
||||
|
||||
$kreuzberg = new Kreuzberg($preprocessedConfig);
|
||||
$result = $kreuzberg->extractFile('poor_quality_scan.pdf');
|
||||
|
||||
echo "OCR with Image Preprocessing:\n";
|
||||
echo str_repeat('=', 60) . "\n";
|
||||
echo "Extracted " . strlen($result->content) . " characters\n";
|
||||
echo "Preview: " . substr($result->content, 0, 200) . "...\n\n";
|
||||
|
||||
$psmModes = [
|
||||
0 => 'Orientation and script detection (OSD) only',
|
||||
1 => 'Automatic page segmentation with OSD',
|
||||
3 => 'Fully automatic page segmentation (default)',
|
||||
4 => 'Assume a single column of text',
|
||||
5 => 'Assume a single uniform block of vertically aligned text',
|
||||
6 => 'Assume a single uniform block of text',
|
||||
7 => 'Treat the image as a single text line',
|
||||
8 => 'Treat the image as a single word',
|
||||
9 => 'Treat the image as a single word in a circle',
|
||||
10 => 'Treat the image as a single character',
|
||||
11 => 'Sparse text - find as much text as possible',
|
||||
13 => 'Raw line - treat as a single text line',
|
||||
];
|
||||
|
||||
$testFile = 'various_layouts.pdf';
|
||||
if (file_exists($testFile)) {
|
||||
echo "Testing different PSM modes:\n";
|
||||
echo str_repeat('=', 60) . "\n";
|
||||
|
||||
foreach ([3, 4, 6, 11] as $psm) {
|
||||
$config = new ExtractionConfig(
|
||||
ocr: new OcrConfig(
|
||||
backend: 'tesseract',
|
||||
language: 'eng',
|
||||
tesseractConfig: new TesseractConfig(psm: $psm)
|
||||
)
|
||||
);
|
||||
|
||||
$kreuzberg = new Kreuzberg($config);
|
||||
$start = microtime(true);
|
||||
$result = $kreuzberg->extractFile($testFile);
|
||||
$elapsed = microtime(true) - $start;
|
||||
|
||||
echo "PSM $psm - {$psmModes[$psm]}:\n";
|
||||
echo " Time: " . number_format($elapsed, 3) . "s\n";
|
||||
echo " Characters: " . strlen($result->content) . "\n";
|
||||
echo " Preview: " . substr($result->content, 0, 80) . "...\n\n";
|
||||
}
|
||||
}
|
||||
|
||||
$singleColumnConfig = new ExtractionConfig(
|
||||
ocr: new OcrConfig(
|
||||
backend: 'tesseract',
|
||||
language: 'eng',
|
||||
tesseractConfig: new TesseractConfig(
|
||||
psm: 4
|
||||
),
|
||||
imagePreprocessing: new ImagePreprocessingConfig(
|
||||
targetDpi: 300,
|
||||
denoise: true
|
||||
)
|
||||
)
|
||||
);
|
||||
|
||||
$kreuzberg = new Kreuzberg($singleColumnConfig);
|
||||
$result = $kreuzberg->extractFile('book_scan.pdf');
|
||||
|
||||
echo "Single-column OCR:\n";
|
||||
echo $result->content . "\n\n";
|
||||
|
||||
$sparseConfig = new ExtractionConfig(
|
||||
ocr: new OcrConfig(
|
||||
backend: 'tesseract',
|
||||
language: 'eng',
|
||||
tesseractConfig: new TesseractConfig(
|
||||
psm: 11
|
||||
),
|
||||
imagePreprocessing: new ImagePreprocessingConfig(
|
||||
targetDpi: 300,
|
||||
denoise: true,
|
||||
sharpen: true
|
||||
)
|
||||
)
|
||||
);
|
||||
|
||||
$kreuzberg = new Kreuzberg($sparseConfig);
|
||||
$result = $kreuzberg->extractFile('receipt.jpg');
|
||||
|
||||
echo "Sparse text OCR (receipt):\n";
|
||||
echo str_repeat('=', 60) . "\n";
|
||||
echo $result->content . "\n\n";
|
||||
|
||||
$highAccuracyConfig = new ExtractionConfig(
|
||||
ocr: new OcrConfig(
|
||||
backend: 'tesseract',
|
||||
language: 'eng',
|
||||
tesseractConfig: new TesseractConfig(
|
||||
psm: 3,
|
||||
enableTableDetection: true
|
||||
),
|
||||
imagePreprocessing: new ImagePreprocessingConfig(
|
||||
targetDpi: 400,
|
||||
denoise: true,
|
||||
sharpen: true,
|
||||
autoRotate: true,
|
||||
deskew: true,
|
||||
removeBackground: true
|
||||
)
|
||||
)
|
||||
);
|
||||
|
||||
$kreuzberg = new Kreuzberg($highAccuracyConfig);
|
||||
$result = $kreuzberg->extractFile('legal_document_scan.pdf');
|
||||
|
||||
echo "High-accuracy OCR:\n";
|
||||
echo "Characters: " . strlen($result->content) . "\n";
|
||||
echo "Tables: " . count($result->tables) . "\n";
|
||||
```
|
||||
Reference in New Issue
Block a user