20200806 Package: azcv.
Can computers really see? What does it actually mean to see? They can certainly capture a likeness of images from the real world, but how much do they understand about it? Some deep questions for another time, or maybe for another chapter. See the Computer Vision chapter in the Data Science Desktop Survival Guide. In this chapter we explore an MLHub package that provides computer vision capabilities.
Microsoft’s proprietary Azure Computer Vision service can perform a variety of vision related tasks. AI can indeed see some things, though maybe it doesn’t actually understand what it is seeing. For free we can use the Azure service to get an idea of what is possible.
The azcv package utilises the Azure Computer Vision cloud API to access closed source pre-built models for computer vision tasks. The package provides a demonstration, a graphical user interface, and command line tools that utilise pre-built, generally deep learning, computer vision models.
Individual command line tools are packaged for common computer vision tasks including image analysis to extract descriptions of the images, word recognition from images, landmark identification, thumbnail generation, and more. The command line tools can be used within command pipelines for tasks including the tagging of personal photos folder, analysis of images from a cameras monitoring a bird feeder, reading street signs to support a driver, and reading handwritten texts.
Most of the commands provided by the package take an image as a parameter which may be a URL or a path to a local file. For brevity through this chapter the URLs we use are short URLs generated through bitly.
To install, configure, and demonstrate the package:
ml install azcv ml configure azcv ml readme azcv ml commands azcv ml demo azcv ml gui azcv
In addition to the demo and gui commands the package supports many computer vision operations including adult, brands, category, celebrities, color, describe, faces, landmarks, objects, ocr, tags, thumbnail, and type.
Azure-based models, unlike the MLHub models in general, use closed source services which have no guarantee of ongoing availability and do not come with the freedom to modify and share. This cloud based service also sends your image files to the Azure cloud for analysis.
Your donation will support ongoing availability and give you access to the PDF version of this book. Desktop Survival Guides include Data Science, GNU/Linux, and MLHub. Books available on Amazon include Data Mining with Rattle and Essentials of Data Science. Popular open source software includes rattle, wajig, and mlhub. Hosted by Togaware, a pioneer of free and open source software since 1984. Copyright © 1995-2022 Graham.Williams@togaware.com Creative Commons Attribution-ShareAlike 4.0