Insights from Dániel Elek: A student's journey in Applied Data Science & AI

Can you describe the project that you have been working on? 

Dániel: ‘Absolutely! Our project for block C centred around the development of a computer vision application. The core idea behind my project was to create an application capable of distinguishing between images of soldiers and civilians within war zones. This application was designed for use with drones and featured a tablet-like interface. I trained it using approximately 500 images, achieving an 85% accuracy rate, that’s quite impressive, considering the relatively small dataset. Finding suitable images proved challenging; I scoured sources like Google and various websites, ensuring the right size and aspect ratios.’ 

‘Market research played a crucial role in the project. I began by studying competitors, examining their approaches and target markets. I also conducted market segmentation to identify potential customer groups in different countries.’ 

‘Furthermore, I dedicated significant attention to enhancing the application's interpretability. This involved the integration of libraries that provided users with the capability to delve into the neural network's decision-making processes when distinguishing between soldiers and civilians.’ 

What was your role in this project? 

Dániel: ‘In this individual project, I took on multiple roles. I served as the market analyst, AI engineer, and computer vision specialist. Besides the technical aspects, I was involved in designing the application to ensure user-friendliness. It was a comprehensive experience that required a multi-faceted approach to deliver a successful result.’ 

‘While I was generally satisfied with my idea, there are a few areas where I would make improvements. I'd fine-tune certain aspects of the project, perhaps by adding an extra explainable AI method for interpretation to enhance its functionality and user experience.​’  

What are your future plans? 

Dániel: ‘At this stage, I'm still in the early stages of my academic journey, but I've developed a keen interest in computer vision. This year, the study programme's focus has shifted to group assignments, and my next project will revolve around robotics and object identification.’  

‘If the opportunity arises, I would like to continue working on improving my computer vision application. Ideally, I'd like to advance it to the point where it not only identifies the presence of soldiers or civilians in an image but also pinpoints their exact locations within the image, adding an extra layer of sophistication to its capabilities.’ 

You can follow Dániel’s professional journey and stay updated with his work via his LinkedIn profile.