Firebase MLKit iOS Text recognition seems to work pretty well if text is formatted in a paragraph or long phrase. However, it fails to work if the numbers are just scattered around which is our case and if there is some line geometry going on around. Some digits are recognized correctly while other exact same digits are not.
I would like to know:
- Can MLKit team improve these cases?
- If we can set MLKit to look only for numbers - this could improve performance and maybe detectability?
- If describing what kind of images we expect as input, in case it can help to improve detection e.g. white background, black numbers only, ignore geometry etc.
- If someone can recommend a better tool for ios. Please note that I am not a machine learning expert, i am young to the field. )
- Any kind of image processing that can be done to improve results?
For this test, i have used the official MLKit quick-start project and in particular the MLVisionExample. Here are some cases:
Thank you very much!