Text area analysis
Contents
[
Hide
]
This structure analysis algorithm works best with sparse irregular text and low-quality photos. It detects smaller text areas in an image, such as individual words, phrases, or lines, and then positions them relative to each other in recognition results.
It is optimal for invoices, screenshots, driver’s licenses, social security cards, government and work IDs, visas, math formulas, code snippets, and more. It can also detect small texts such as handwritten notes, signatures or stamps. In addition, it is well suited for reading smartphone photos and low-quality scans.
This structure analysis algorithm may be less efficient when dealing with large amounts of structured textual data, such as articles and books. Try Complex structure analysis instead.