Build a document scanner into your Android app

1
// 1. Convert a captured page image into a PDF.
2
val imageSize = Size(image.width.toFloat(), image.height.toFloat())
3
val pageImage = PageImage(image, PagePosition.CENTER).apply { setJpegQuality(70) }
4
val newPage = NewPage.emptyPage(imageSize).withPageItem(pageImage).build()
5
val creationTask = PdfProcessorTask.newPage(newPage)
6

7
PdfProcessor.processDocumentAsync(creationTask, outputFile).subscribe({ }, { }, {
8
    // 2. Run OCR so the scan becomes selectable and searchable.
9
    val document = PdfDocumentLoader.openDocument(context, Uri.parse(outputFile.absolutePath))
10
    val ocrTask = PdfProcessorTask
11
        .fromDocument(document)
12
        .performOcrOnPages((0 until document.pageCount).toSet(), OcrLanguage.ENGLISH)
13

14
    PdfProcessor.processDocumentAsync(ocrTask, ocrFile).subscribe({ }, { }, {
15
        // 3. Extract the recognized text.
16
        val ocrDocument = PdfDocumentLoader.openDocument(context, Uri.parse(ocrFile.absolutePath))
17
        Log.d("Nutrient OCR", ocrDocument.getPageText(0))
18
    })
19
})

1
// 1. Convert a captured page image into a PDF.
2
final Size imageSize = new Size(image.getWidth(), image.getHeight());
3
final PageImage pageImage = new PageImage(image, PagePosition.CENTER);
4
pageImage.setJpegQuality(70);
5
final NewPage newPage = NewPage.emptyPage(imageSize).withPageItem(pageImage).build();
6
final PdfProcessorTask creationTask = PdfProcessorTask.newPage(newPage);
7

8
PdfProcessor.processDocumentAsync(creationTask, outputFile).subscribe(progress -> { }, throwable -> { }, () -> {
9
    // 2. Run OCR so the scan becomes selectable and searchable.
10
    final PdfDocument document = PdfDocumentLoader
11
        .openDocument(context, Uri.parse(outputFile.getAbsolutePath()));
12
    final Set<Integer> pages = new HashSet<>();
13
    for (int i = 0; i < document.getPageCount(); i++) pages.add(i);
14

15
    final PdfProcessorTask ocrTask = PdfProcessorTask
16
        .fromDocument(document)
17
        .performOcrOnPages(pages, OcrLanguage.ENGLISH);
18

19
    PdfProcessor.processDocumentAsync(ocrTask, ocrFile).subscribe(progress -> { }, throwable -> { }, () -> {
20
        // 3. Extract the recognized text.
21
        final PdfDocument ocrDocument = PdfDocumentLoader
22
            .openDocument(context, Uri.parse(ocrFile.getAbsolutePath()));
23
        Log.d("Nutrient OCR", ocrDocument.getPageText(0));
24
    });
25
});