Inconsistent scanning api v website

Hi there!

Recently I have sent out an email about this issue but perhaps the forum would have been better to start on as there might be others with the same issue. I have been using the api for quite some time and in general I am really positive about it. However, one issue that keeps popping up which I cannot fix myself is really annoying. When uploading the image below to the website it is in fact able to read the “0” (zero) just fine. The API on the other hand refuses to detect it on several images.

Website settings:

  • Detect orientation and auto-rotate image if needed, disabled
  • Do receipt scanning and/or table recognition, enabled
  • Auto-enlarge content (recommended for low DPI), enabled
  • Create Searchable PDF, Just extract text and show overlay
  • Select OCR Engine to use, OCR Engine2

API settings:

  • ‘isOverlayRequired’: False,
  • ‘isTable’: True,
  • ‘apikey’: api_key,
  • ‘base64Image’: ‘data:image/jpeg;base64,’ + something,
  • ‘OCREngine’: 2,
  • ‘scale’: True

Thanks in advance!

The only difference I see is that for the online ocr website you upload the image, but for the OCR API you use base64. Maybe something goes wrong during the base64 to/from image conversion (=> lower image quality).

As a test, can you upload the image during the api call as well?