Discover how Azure AI Vision’s Optical Character Recognition (OCR) can extract text from scanned handwritten letters to Santa Claus. Learn how this tool helps index and analyze data for historical trends and insights.
Table of Contents
Question
Xerigon Corporation is a toy company that uses the Azure AI Vision service. They are creating an app that will extract text from an archive of over 20,000 handwritten letters addressed to Santa Claus from decades in the past. The physical letters have been scanned to images.
The images must be indexed. Xerigon wants to query the images to find out the following:
- Which toys were the most requested in a given year?
- What was the average age of the letter writer?
- What was the gender of the letter writer?
You need to create an indexer that has a skill set.
Which skill set should you include?
A. Optical Character Recognition (OCR)
B. Name Entity Recognition (NER)
C. Image Analysis
D. Personally Identifiable Information (PII)
Answer
A. Optical Character Recognition (OCR)
Explanation
You would use the Optical Character Recognition (OCR) skill set. The Azure AI Vision service, mainly through its Read API, is designed for OCR tasks, including extracting text from both printed and handwritten documents such as the scanned letters in this scenario.
App developers can index the data to make it searchable. This searchable data can be in the form of text or vectors. An indexer can include a skill set for data enrichment capabilities such as:
- Name Entity Recognition (NER) from the Azure AI Language service
- Personally Identifiable Information (PII) from the Azure AI Language service
- Image Analysis from the Azure AI Vision service
- OCR from the Azure AI Vision service
In this scenario, you want to read and extract data from the scanned handwritten documents, which requires the OCR skill set.
You would not use the Image Analysis skill set. Image Analysis can extract information containing adult content, human faces, or specific brands. Since the scanned images are handwritten letters and not photographs, OCR would be a better skill set than Image Analysis.
You would not use Name Entity Recognition (NER). NER is a feature that finds and classifies entities from unstructured text such as person, job, job title, address, and phone. It is a feature of the Azure AI Language service, and you are using the Azure Vision service for this app. While NER could be used to find the letter writer’s name, it would not provide the other information that needed to be queried.
You would not use Personally Identifiable Information (PII). PII is a feature that finds information such as email addresses, phone numbers, credit card numbers, and other forms of identification. It is a feature of the Azure AI Language service, and you are using the Azure Vision service for this app. While PII could be used to find the letter writer’s name, it would not provide other information that needed to be queried.
Microsoft Azure AI Engineer Associate AI-102 certification exam practice question and answer (Q&A) dump with detail explanation and reference available free, helpful to pass the Microsoft Azure AI Engineer Associate AI-102 exam and earn Microsoft Azure AI Engineer Associate AI-102 certification.