- A Cloud Guru
Image Analysis with Azure Computer Vision
In this hands-on lab, you are working as an AI Engineer for Marketing Man, a marketing company, and you’ve been asked to analyze images and get information back for marketing purposes. You’ll use a pre-provided command-line application to utilize Azure Computer Vision to scan images and get back an image description, image tags, and detect faces in the picture.
Table of Contents
Review the Image to Be Analyzed and Connect to the Virtual Machine
View the image provided for the lab (https://raw.githubusercontent.com/ACloudGuru/content-AI-900/main/images/image-analysis/1-computervision-couple.jpg).
Make some guesses about what description will come back and what tags the Computer Vision service will provide.
Use a remote desktop client application to connect to the public IP address of the Windows virtual machine shown on the lab page.
If you are on Windows OS, see step 2 in the following article on how to use the Remote Desktop client. In place of the PC name, use the public IP address given in the lab page.
If you are on Mac OS, you can download the Microsoft Remote Desktop client from the Apple Store.
Run the Command-Line Application
Using File Explorer, browse to
C:\code\content-AI-900\and double-click on 1-computer_vision.py to start the image analysis!
Review the Results of the Image Analysis
Look at the description, tags, and facial detection features.
Recently, Microsoft removed the detection of age, gender, emotion, etc. from the facial recognition services, please see below for a statement from Microsoft detailing thier decision.
Bonus: Try Your Own Images!
As a bonus objective, right-click on 1-computervision.py and click Edit with IDLE. In the window that opens, look for the red comment that says, “change the URL between the quotes below to run your own image”, and change the URL to a picture you’d like to run through image analysis!
You can use the following link to the second image (https://raw.githubusercontent.com/ACloudGuru/content-AI-900/main/images/image-analysis/1-computervision-kids.jpg) I use in the solution video.
What's a lab?
Hands-on Labs are real environments created by industry experts to help you learn. These environments help you gain knowledge and experience, practice without compromising your system, test without risk, destroy without fear, and let you learn from your mistakes. Hands-on Labs: practice your skills before delivering in the real world.