When you upload photos to Facebook, have you noticed that the website already seems to know who’s in them? It’s remarkable, and you can give the credit to big data. Face recognition software, like fraud detection and ad matching algorithms, draws on deep libraries of content in order to deliver the correct results. And these data collections are hard at work across the web and in many of your favorite apps.
Faces And Ferns
It comes as no surprise that developers have been hard at work on face recognition software since it’s an integral part of security programs. Police departments use face recognition software to search for criminals based on surveillance videos and it’s certainly made photography fans’ lives easier. But what about more unusual subject matter?
Yes, there are data libraries full of interesting material helping us identify things in our day-to-day lives. When you see different types of ferns, can you tell if they’re glade ferns, autumn ferns, or Boston ferns? Unless you’re an avid gardener, you probably can’t tell. The same goes for many other plant types – ivy, chrysanthemums, or tulips, for example. Now, however, there are tools that can help.
With assistance from PlantNet, a French app, a quick photo can help you determine what you’re looking at. The app draws on 64,000 database entries to identify plants from many different parts of the world, though it’s most accurate in France and Western Europe. Similarly, LeafSnap, a tree-identification app developed by the Smithsonian, uses the power of big data to determine a tree’s species based on their leaves.
Seeing Through The Noise
Of course, there are certain challenges facing big data when it comes to the field of image recognition. What happens, for example, when a photo is out of focus and pixelated? As it turns out, the technology is becoming so advanced that even when your face is blurred out, some programs can still recognize you. The key is the growth of big data in the direction of neural networks – the more data the computer sees, the more easily the system can recognize even highly disguised photos.
See It Yourself
Face recognition programs have become so popular – and so mainstream – that people can easily develop their own programs if they’re willing to put time into the data end. Using OpenCV in Python, programmers can activate an easy computer vision program for use in recognition programs. Combining that program with a library of images, you can create identification programs not just for faces, but also for a range of other objects. Whatever you’re interested in, if you build the data foundations, your computer can tell you what it’s looking at.
Further developments in big data, including the ability to draw more easily on preexisting libraries, mean that in the next few years, we’re likely to encounter a proliferation of identification programs for anything and everything. In most cases, this is good – we all want to know what that plant, bug, or rock is – but there are privacy risks as well.
Going forward, then, we will need to be careful not to compromise people in sensitive situations in our quest to see and know it all.