Describe features of computer vision workloads on Azure

The correct answer is Optical Character Recognition (OCR).

OCR is the standard technical term in the field of computer vision and document processing that specifically refers to the technology and process of converting images containing text (whether from scanned documents, photos, or digital images) into machine-encoded text that can be edited, searched, and processed by computers.

This is a fundamental concept in Azure AI services, particularly in Azure Computer Vision and Azure AI Document Intelligence (formerly Form Recognizer), which provide OCR capabilities to extract text from images and documents.

Why the other options are incorrect:

Document layout analysis and segmentation refers to a different process that involves identifying and separating different structural elements of a document (like headers, paragraphs, tables, and columns). While this may be part of a complete document processing pipeline, it is not the term for converting text images into machine-readable text.

Visual text pattern identification and encoding is not a recognized technical term in the field. While it describes aspects of what happens during text recognition, it is not the established terminology used in computer vision or AI documentation.

Automated glyph detection and interpretation describes a component process that might be involved in character recognition (glyphs are individual character shapes), but this is not the standard technical term for the overall process. OCR is the universally accepted term used across the industry and in Azure AI services.

The correct answer is that JPEG, PNG, GIF, and BMP formats are supported for image analysis operations.

Azure AI Vision's Image Analysis API officially supports these four standard image formats: JPEG, PNG, GIF, and BMP. These are the most commonly used raster image formats and provide broad compatibility for various image analysis scenarios.

Why the other options are incorrect:

The option mentioning TIFF and WebP is incorrect because while these are valid image formats, TIFF is not listed among the supported formats for the standard Image Analysis API, and WebP support is limited.

The option that includes SVG and RAW formats alongside vector graphics is incorrect because SVG is a vector format (not a raster format), and the Image Analysis API is designed to work with raster images, not vector graphics. RAW formats are also not supported by the standard API.

The option listing HEIF and JPEG2000 is incorrect because these newer or specialized formats are not among the standard supported formats for Azure AI Vision's Image Analysis API. The API focuses on the most widely-used and compatible image formats.

The correct answer is Azure AI Vision.

Azure AI Vision is specifically designed to provide pre-built models for analyzing images with capabilities that include:

• Object Detection: Identifying and locating common objects within images
• Scene Recognition: Detecting activities and understanding the context of scenes
• Caption Generation: Automatically generating descriptive captions that explain the visual elements present in an image
• Tag Generation: Applying relevant tags based on visual content
• Face Detection: Identifying human faces and their attributes

This service is part of Azure AI Services and comes with pre-trained models ready to use out-of-the-box, making it ideal for general-purpose image analysis tasks without requiring custom model training.

Why the other options are incorrect:

The second option describes a service focused on training custom models for specific domains using your own labeled datasets. While useful, this requires you to build and train models rather than using pre-built ones for general image analysis tasks.

The third option refers to a search service that indexes content and can process images as part of search operations. However, its primary purpose is information retrieval and search functionality, not dedicated image analysis with pre-built computer vision models.

The fourth option is designed specifically for document processing, extracting structured information from forms and documents. While it processes visual layouts, it's specialized for document understanding rather than general image analysis, object detection, and caption generation.