From February till June, I worked at Stage Gate 11 with a lot of pleasure. As a Master student Applied Mathematics at Delft University of Technology, with a specialisation in Stochastics (i.e. statistics and probability), I have learned a lot of theoretical mathematics. At Stage Gate 11, I could finally apply everything I learned in my studies and gain some work experience.
The project
For my internship, I did a project on the comparison of spectra and probability distributions, helping with the classification algorithm of the Delta R shoe scanner. Every pixel from a scan contains a spectrum, and this spectrum must be compared to all known spectra of illegal substances, in order to classify the substance on the pixel as either dangerous or safe. The algorithm that was used thus far by Stage Gate 11 had to be updated, so I set out to find a function for the comparison of spectra that gave a better performance.
First, I did a lot of literature research into the Delta R method and into functions that are commonly used for this type of problem. Then, I implemented a few select functions. I wrote the algorithms that compare the spectra and visualized the results. Eventually, I developped the algorithm that indeed gave better results than the previous one, meaning we can now get a higher true positive rate for the same false positive rate, or a lower false positive rate for the same true positive rate!
I hope Stage Gate 11 will continue to use my findings. Anyway, I am grateful for the work experience I gained, newly found programming skills, and the great time I had at Stage Gate 11.
If you are interested in a similar internship, just reach out and SG provides the details. They are very open to discuss any of your preferences. Check the related article on the left, call (088-2501600) or send an email (info@sg11.nl).