I work for the company that created the LEADTOOLS SDK. If you haven't tried it, there's a free evaluation edition here.
The evaluation comes with free technical support through email and chat.
The OCR works either by automatically creating zones, or by accepting pre-defined zones you specify. There are native libraries for different platforms, but if you want to use the same source code on both Android and iOS, you can use our Xamarin SDK (part of the main Evaluation for Windows).
The following code shows how you can define a zone and recognize text using LEADTOOLS classes:
using (IOcrPage ocrPage = ocrEngine.CreatePage(rasterImage, OcrImageSharingMode.AutoDispose))
LogicalRectangle bounds = LogicalRectangle.FromLTRB(left, top, right, bottom, LogicalUnit.Pixel);
OcrZone zone = new OcrZone();
zone.ZoneType = OcrZoneType.Text;
zone.Bounds = bounds;
string text = ocrPage.GetText(-1);