ocr / RecognizeCharacters / 0.3.0


Use https://algorithmia.com/algorithms/tesseractocr/OCR for a more customizable interface.

Uses Tesseract to perform OCR on a given image. As the trained model, it uses the latest model uploaded to Tesseract's Google Drive.

Suggestion: If your image has noise that you would like to get rid of, you may wish to try out the ImageBinarization algorithm first, before feeding it to this algorithm.


The input should be an image and can be from an arbitrary url, the data api, or binary. 


The output is the recognized text.

If you would like to write the output to the data api, you can provide a second input that contains the target data api url (with filename). The output is same as this second input.