Analysis of iron ore phases and microstructures used to be done manually at the world’s largest iron ore producers. We collaborated with LKAB to automate analysis.

In brief


Automate the inspection process of 12 mm iron ore pellets to identify mineral content, distribution of additives, and porosity levels.


Microscopes equipped with high-resolution digital cameras photo the pellets. The images are automatically analyzed, and the results are visualized in graph form.

How we did it

We trained a machine learning pixel Classifier to classify the images and built a system to quantify relevant pellet image data and visualize the results.


The R&D team can now dedicate their time to analyzing results rather than carrying out manual pellet inspection. Additionally, pellet analysis is now standardized.

About the client

LKAB is Europe’s largest iron ore producer, with 4,200 employees in 13 countries. The company, which is wholly owned by the Swedish state, is headquarted in Luleå, northern Sweden.  

Machine learning: Automating iron pellet analysis for the LKAB mining and processing company   

LKAB’s core business is the mining and processing of iron ore for the steel industry. Over the years, LKAB has developed a unique market offering for its customers, namely blast furnace pellets. These have now become LKAB’s most vital product – with an annual production capacity at its plants in northern Sweden, of 28 million tons per annum. The main benefit of using pellets over standard iron ore products at steel mills, is lower furnace energy requirements. An addition benefit is that the pellets contain extra minerals, such as olivine, which provide improved high-temperature properties.  

Pellet inspection  

The R&D team at LKAB are continuously searching for improved ways to produce more effective as well as customized furnace pellets. Part of this process includes manually inspecting the formation of different iron oxide phases and microstructures in sample pellets. This is a time-consuming process that must be done by experts, who examine sample pellets embedded in epoxy, through an optical microscope. Some minerals can be identified by color and intensity, such as magnetite and hematite, while others are based on pellet surface texture. This process lends itself perfectly to machine learning.  

Automating inspection 

Our goal was to automate this process. The first step was to create a dataset of high-quality microscopy images of representative iron ore pellets. Experts at LKAB provided us with sample pellets embedded in epoxy and access to automated microscopes – each equipped with high-resolution digital cameras – at their facility. 

Creating an image library 

High-magnification images of the sample pellets were acquired, and multiple images were pieced together into large mosaic images to create a complete view of each pellet. Depending on the size of the pellet, the images ranged between 500 and 900 mega pixels.   

Training a pixel classifier 

Once the image library was created, the images were annotated in close collaboration with analysis experts at LKAB. Regions from each class were marked in a set of images and used to train the pixel Classifier. The Classifier was subsequently used to classify the images and the annotation was improved by correcting the classification. This was repeated until a satisfactory classification was achieved. The Classifier was then evaluated on a set of images that were not included in the training process. 

Intuitive pellet analysis 

The last step was to extract and quantify relevant information from the classified images and create a visual format that could be easily interpreted by all members of the pellet analysis team at LKAB. The result was a series of graphs depicting a mineral map, the microstructure and the mineral content of a pellet. 

Thanks to the automated mapping of sample pellets, experts can now dedicate their time to analyzing the results, rather than inspecting the pellets. This has speeded up the analysis process considerably. Furthermore, analysis is no longer based on one person’s viewpoint, it is now standardized across all pellet samples – based on the collective experience and advise of the R&D team.  

Optimizing pellet properties 

The automated and quantitative characterization of microstructures can be used to generate the necessary data to better understand the effects of pellet additives and the different process parameters. Additionally, it is now possible to automate a quantitative study of the reaction mechanisms in the blast furnace based on pellet microstructures. With this knowledge, pellet properties can be optimized for different customer applications.  

“Thanks to a close collaboration with Data Ductus, we were able to define the machine learning project and move through development to implementation quickly and smoothly,” says Johan Sandberg, Section Manager, Process & Product Development at LKAB. Automating the inspection of pellets has freed up time for the R&D team to focus on more valuable and interesting work.”