Add benchmark code for the Chars74k dataset. Using the CNN character classifier reaches 75% and 84% accuracy for case-sensitive and case-insensitive recognition respectively.

Add benchmark for ICDAR2015 dataset using OCRTesseract and ERFilter classes. Gives word spotting f-score 0.642082 with strongly contextualized lexicon (100 words).

Report the total f1 score (0.37) and not the mean-f1. This is in accordance with the standard SVT evaluation protocol and allows for comparison with other published results.

Changes SVT recognition evaluation to be Case Insensitive (according to the standard evaluation protocol). This makes the benchmark obtained mean-f1 score increase from 0.23 to 0.27