diff --git a/layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/classification/DocuMineClassificationService.java b/layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/classification/DocuMineClassificationService.java index 4e9459f..8a73452 100644 --- a/layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/classification/DocuMineClassificationService.java +++ b/layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/classification/DocuMineClassificationService.java @@ -111,10 +111,12 @@ public class DocuMineClassificationService { && (textBlock.getMostPopularWordStyle().contains("bold") && Character.isDigit(textBlock.toString().charAt(0)) && atLeast3Matcher.reset().find() - && !textBlock.toString().contains(":") || textBlock.toString().equals(textBlock.toString().toUpperCase(Locale.ROOT)) - && atLeast3Matcher.reset().find() - && !textBlock.toString().contains(":") || textBlock.toString().startsWith("APPENDIX") || textBlock.toString() - .startsWith("FIGURE") || textBlock.toString().startsWith("Continued TABLE") || textBlock.toString().startsWith("TABLE")) + && !textBlock.toString().contains(":") // + || textBlock.toString().equals(textBlock.toString().toUpperCase(Locale.ROOT)) && atLeast3Matcher.reset().find() && !textBlock.toString().contains(":") // + || textBlock.toString().startsWith("APPENDIX") // + || textBlock.toString().startsWith("FIGURE") // + || textBlock.toString().startsWith("Continued TABLE") // + || textBlock.toString().startsWith("TABLE")) && !textBlock.toString().endsWith(":") && atLeast3Matcher.reset().find()) { PageBlockType headlineType = PageBlockType.getHeadlineType(1);