public/docs
1
0
Fork 0
mirror of https://github.com/sismics/docs.git synced 2025-12-29 01:21:49 +00:00
Code Issues Releases Wiki Activity

Closes #373: high quality PDF to image conversion before OCR

Browse Source
This commit is contained in:
bgamard
2020-02-13 17:43:07 +01:00
parent a7423caeb1
commit 90a49efa4a
5 changed files with 41 additions and 21 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
docs-core/src/main/java/com/sismics/docs/core/util/format/PdfFormatHandler.java
View File

@@ -6,6 +6,7 @@ import com.sismics.util.mime.MimeType;
import org.apache.pdfbox.io.MemoryUsageSetting;
import org.apache.pdfbox.multipdf.PDFMergerUtility;
import org.apache.pdfbox.pdmodel.PDDocument;
import org.apache.pdfbox.rendering.ImageType;
import org.apache.pdfbox.rendering.PDFRenderer;
import org.apache.pdfbox.text.PDFTextStripper;
import org.slf4j.Logger;
@@ -60,7 +61,7 @@ public class PdfFormatHandler implements FormatHandler {
for (int pageIndex = 0; pageIndex < pdfDocument.getNumberOfPages(); pageIndex++) {
log.info("OCR page " + (pageIndex + 1) + "/" + pdfDocument.getNumberOfPages() + " of PDF file containing only images");
sb.append(" ");
sb.append(FileUtil.ocrFile(language, renderer.renderImage(pageIndex)));
sb.append(FileUtil.ocrFile(language, renderer.renderImageWithDPI(pageIndex, 300, ImageType.GRAY)));
}
return sb.toString();
} catch (Exception e) {