Posts

This is the second post in a series of posts, where we’ll look into Microsoft Cognitive Services. More specifically we’re looking at the Face API, and what that can do for us in terms of identifying people. The end goal is to connect a web camera to a Raspberry Pi, grab an image of a person, identify that person and show the name on a screen. In the first post we described some basic terms one should know and we looked at how you could get API keys for the Face API. We also looked into detecting faces in an image, where we marked a face with a square. If this is the first post you’re reading in this series, I highly suggest you read the first post, as we’ll continue building on the app we started on there. ...

A few weeks back I had a problem with my car. The windscreen wiper wouldn’t work, or it was rather slow when it did work. Being an older model car, I figured I could probably fix it myself. Long story short, I did fix it myself. After I had fixed it, I started to think about the process I’d been through. Interestingly enough I discovered that there were at least a couple of things, which I could relate back to software development. ...

As the world gets increasingly more data power, we are enabled to do more things (in the lack of better words). One of those things is to make computers and applications more intelligent, and Microsoft have created a series of API’s to help doing so. Using Microsoft Cognitive Services you’re able to do use a whole bunch of API’s, ranging from translation, searching, language-based and facial recognition. I find a lot of them interesting, especially the Face API, which I’ll spend some time taking a closer look at. ...

I’ve spent this week in London, attending a course on best practices and design patterns. The course was aimed at .NET developers, where the following was the main targets in terms of what we were to learn: Solve complex programming problems using design patterns Deliver bug-free code using test-driven development in Visual Studio Create layered architectures for reusability and eliminating functional duplication Apply the S.O.L.I.D. principles for better class design Simplify and automate repetitive tasks ...

Day 2 of the course continued where we left of yesterday, namely with topics in regards to test-driven development. So far this day was the one with the most design patterns in it, as we went through a whole bunch in chapter 4. We also started on chapter 5, however I think it’s best to leave my notes for that until we finish it on day 3. I see that chapter 3 comes out weirdly (split over two posts), so I’ll write about the entire chapter 5 tomorrow. ...