The students of the École Polytechnique Fédérale de Lausanne in Switzerland have recently come up with a machine learning strategy that can easily identify the parasites which pose threat to the bees.
Bees are facing a deadly threat from a parasite called Varroa mite. It infects and harms the bees and sucks blood from them and their young ones. Although, it can’t kill them straight away but can make them weak right after their birth. This can be a serious risk to the entire territory.
The beekeepers naturally want to avoid such encounters immediately. They started pulling out the hive regularly for a few days in order to inspect the dirt for the tiny mites. It is a painful task to perform as the mites aren’t even visible because they are too small in size. And, even if one is left out, the situation can get even worse.
This problem could be solved by the use of Machine Learning as its algorithms can easily search and sort through a noisy dataset and identify a particular pattern which is the shape of these mites, in our case.
The project just requires a good smartphone picture by the beekeepers and have them uploaded to the EPFL system. The implementation of the idea started in 2017 when the model was trained with a large number of images. The accuracy rate of the model achieved till now is 90%.
Not only this, but the movement of the bees is also monitored by a BeeSmart device. The device has sensors to track the activity of the entire colony, measure the humidity and temperature of the hive, and even to trace various mating patterns. The collected data is then sent to the beekeepers using the BeeSmart who will analyze and process it.
Artificial Intelligence has literally worked upon every existing problem in the world - be it gadgets, security, environment-safety etc. It provides a really wide spectrum to be explored with each passing day. It is amazing how AI or ML is transforming our lives for the good but we still have a long way to go.
