Identify features of OCR solutions
The Computer Vision (CV) capabilities of Azure Machine Learning can be used as a solution for OCR.
The OCR solution can be used to extract “text” from an image. Letters and numbers are identified from shapes and then converted into machine-encoded text that can then be further utilized for processing by applications or users.
An example of OCR for an image can be seen in Figure 6.7:
Figure 6.7 – Extracting text from an image with OCR capability
The OCR model is trained to recognize elements of text, including punctuation, as well as numerals from individual shapes, and then produce an output as text. An example of a text output produced by an OCR model is shown in Figure 6.8:
Figure 6.8 – OCR model text extraction output
The following is an extract of the corresponding API’s JSON response for the object detection from Figure 6.8:
...