Class DocumentScanResult

java.lang.Object
com.codename1.ai.vision.DocumentScanResult
public final class DocumentScanResult extends Object
Corrected document pages returned as encoded image data. Pages are ordered as detected and both construction and access make defensive copies, so the result can safely cross asynchronous boundaries. The still-image scanner is currently Apple-only; Android's ML Kit document API owns an interactive camera flow and does not implement this analyzer contract.

  • Constructor Details

    • DocumentScanResult

      public DocumentScanResult(byte[][] pages)
      Creates a corrected document scan without backend metadata.
      Parameters:
      pages - encoded corrected page images, deeply defensively copied

    • DocumentScanResult

      public DocumentScanResult(byte[][] pages, VisionMetadata metadata)
      Creates a corrected document scan with backend diagnostics.
      Parameters:
      pages - encoded corrected page images, deeply defensively copied
      metadata - backend details, or null

  • Method Details

    • getPageCount

      public int getPageCount()
      Returns:
      number of corrected pages

    • getPage

      public byte[] getPage(int index)
      Parameters:
      index - zero-based page index
      Returns:
      defensive copy of that page's encoded image

    • getMetadata

      public VisionMetadata getMetadata()
      Returns:
      backend metadata, or null for manually created results