Hi,
I would look through each letter in your word to get the word's confidence. So something like this:
ETEXT_DESC* results = tess->Recognize_all_Words();
for (int iter = 0; iter < results->count; iter++) {
const EANYCODE_CHAR* ch = &results->text[iter];
if (255 > ch->confidence) {
// TODO: reject the letter and thus the whole word associated with it
}
}
-
Albert