Exam AI-102 topic 2 question 32 discussion

To me the correct answer is B. With the new Image Analysis API 4.0 (in preview) you could use OCR feature too, but as i said, it is in preview. And i don't think it is considered in the exam. https://learn.microsoft.com/en-us/azure/ai-services/computer-vision/overview-image-analysis?tabs=4-0 Instead the Read API is particularly adapted for text-heavy documents and it seems this is the case https://learn.microsoft.com/en-us/rest/api/computervision/read/read?view=rest-computervision-v3.1&tabs=HTTP

All answers should have a reference to prove the answer is true

The Read API in Computer Vision is CORRECT because it is specifically designed for extracting printed and handwritten text from images. This API is optimized for handling a large number of images and is suitable for scenarios where text needs to be extracted from scanned documents or magazine articles. B is the answer

Computer Vision is now Azure AI Vision. You can use the Read API of it's OCR service. The Optical Character Recognition (OCR) service extracts text from images. You can use the Read API to extract printed and handwritten text from photos and documents. https://learn.microsoft.com/en-us/azure/ai-services/computer-vision/overview

Answer is C Form Recognizer There is note in below microsoft doc mentioning about document images for text extraction OCR for images (version 4.0) https://learn.microsoft.com/en-us/azure/ai-services/computer-vision/concept-ocr Please refer important section to Select the Read edition that best fits your requirements. https://learn.microsoft.com/en-us/azure/ai-services/computer-vision/how-to/call-read-api

Azure AI Document Intelligence is a more sophisticated solution. For example, it can identify key/value pairs, tables, and context-specific fields. If you want to deploy a complete document analysis solution for both extracting AND understanding text, Azure AI Document Intelligence is a good solution. In this case, Document Intelligence is a too advanced solution as the question doesn't provide any information about what to do with the extracted text, the focus here is on text extraction alone. So the answer is B

A. Computer vision does not have read model. Read model is document intelligence (formregnize) model

Selected Answer: B

see if you can spot this one B.

This is a good explanation from chat GPT why B is the correct fit and A is not: A - Computer Vision Image Analysis: This is a service provided by Azure that can extract a wide variety of visual features from your images. It can determine whether an image contains adult content, find specific brands or objects, or find human faces. However, while it does have OCR capabilities, it is not specifically designed for large-scale text extraction from images. B - Read API in Computer Vision: This is a part of the Azure Computer Vision service that is designed to extract printed and handwritten text from images. It uses state-of-the-art Optical Character Recognition (OCR) algorithms optimized for text-heavy documents2. This could be a good fit for your needs as it is designed to handle large amounts of text in images. the key here is option A it is not specifically designed for large-scale text extraction from images.

the Read API in Computer Vision is right.

Read API is particularly adapted for text-heavy documents and it seems this is the case

Document Intelligence(Previously Form Recognizer) reads small to large volume of text from images and PDF documents. For example: receipts, articles, and invoices

i think it should be B based on ChatGpt using Azure Read API if: Your primary focus is on text extraction from documents, forms, or images with printed text. You have a batch processing requirement, and you need to process a large number of documents or images at once. You need highly accurate text extraction with structured output.

i also think its B Use this interface to get the result of a Read operation, employing the state-of-the-art Optical Character Recognition (OCR) algorithms optimized for text-heavy documents. https://learn.microsoft.com/en-us/rest/api/computervision/3.2preview2/read/read?tabs=HTTP

Why not B?