In addition to Tesseract, all the big boys: Google, Microsoft and IBM now have their own OCR API offer. These APIs provide easy OCR scanning with text images with varying degrees of accuracy. I believe that Google Vision is the most accurate for photo receipts. You still have to extract data from the text with half the garbage.
If you need an API that returns field metadata, such as: total amount, tax amount, date and seller information, where applications can consume directly. Check out https://www.taggun.io . I specifically developed an API for this purpose.
source share