Finding problems using unsupervised image categorization
Needle in a Haystack

The most tedious part of supervised machine learning is providing sufficient supervision. However, if the samples come from a restricted sample space, unsupervised learning might be fine for the task.
In any classification project, it is certainly possible to get someone to review a certain number of images and build a classification list. However, when entering a new domain, it can be difficult to identify domain knowledge experts or to develop a ground truth for classification upon which all experts can agree. This is true regardless of whether you are looking at the backside of a silicon wafer for the first time, or if you are trying to identify the presence of volcanoes in radar images from the surface of Venus [1].
Alternatively, you can bypass all these problems and kick-start a classification project with unsupervised machine learning. Unsupervised machine learning is particularly applicable to environments where the typical images are largely identical, much like the pieces of hay in the haystack that you need to ignore when looking for needles.
In this article, I examine the potential for using unsupervised machine learning in Python (version 3.8.3 64-bit) to identify image categories for a restricted image space without resorting to training neural networks. This technique follows from the long tradition within engineering of finding the simplest solution to a problem. In this particular case, the solution relies upon the ability of the functions within the OpenCV and mahotas computer vision libraries to generate parameters for the texture and form within an image.
[...]
Buy this article as PDF
(incl. VAT)
Buy Linux Magazine
Subscribe to our Linux Newsletters
Find Linux and Open Source Jobs
Subscribe to our ADMIN Newsletters
Support Our Work
Linux Magazine content is made possible with support from readers like you. Please consider contributing when you’ve found an article to be beneficial.

News
-
Linux Kernel 6.17 is Available
Linus Torvalds has announced that the latest kernel has been released with plenty of core improvements and even more hardware support.
-
Kali Linux 2025.3 Released with New Hacking Tools
If you're a Kali Linux fan, you'll be glad to know that the third release of this famous pen-testing distribution is now available with updates for key components.
-
Zorin OS 18 Beta Available for Testing
The latest release from the team behind Zorin OS is ready for public testing, and it includes plenty of improvements to make it more powerful, user-friendly, and productive.
-
Fedora Linux 43 Beta Now Available for Testing
Fedora Linux 43 Beta ships with Gnome 49 and KDE Plasma 6.4 (and other goodies).
-
USB4 Maintainer Leaves Intel
Michael Jamet, one of the primary maintainers of USB4 and Thunderbolt drivers, has left Intel, leaving a gaping hole for the Linux community to deal with.
-
Budgie 10.9.3 Now Available
The latest version of this elegant and configurable Linux desktop aligns with changes in Gnome 49.
-
KDE Linux Alpha Available for Daring Users
It's official, KDE Linux has arrived, but it's not quite ready for prime time.
-
AMD Initiates Graphics Driver Updates for Linux Kernel 6.18
This new AMD update focuses on power management, display handling, and hardware support for Radeon GPUs.
-
AerynOS Alpha Release Available
With a choice of several desktop environments, AerynOS 2025.08 is almost ready to be your next operating system.
-
AUR Repository Still Under DDoS Attack
Arch User Repository continues to be under a DDoS attack that has been going on for more than two weeks.