Mosquitoes, ticks, flies, spiders and other insects can be far more than a nuisance. The diseases they carry-malaria, dengue, yellow fever, West Nile virus, Lyme disease and many others-can cause serious illness and in some cases death. In the Americas, one out of every two people lives in an area at risk of one or more of these vector-borne diseases.

About half the population of the Western Hemisphere-some 500 million people-live in areas at risk of one or more vector-borne diseases. They include long-established diseases such as malaria, dengue, yellow fever, Chagas disease, onchocerciasis (also known as "river blindness") and lymphatic filariasis (also known as "elephantiasis"). More recent arrivals include West Nile virus, which arrived in the United States in 1999, and chikungunya virus, first reported in the Caribbean last year.

Bites spiders – such as the false widow spider – are capable of giving a nasty bite. Spider bites leave small puncture marks on the skin, which can be painful and cause redness and swelling. Some spiders bites can cause you to feel or be sick, sweating and dizziness. Bites can also become infected or cause a severe allergic reaction in rare cases.

This project has two focuses: 1) It can help protect people against possible bites from dangerous insects such as spiders; and 2) Protect insects beneficial for ecology like bees which pollinate flowers and produce honey.

This is a smart system that makes use of machine learning (artificial intelligence), and in it's first version can distinguish between spiders and bees. When it detects a spider it emits a spray and red LED flash, and when it detects a bee it doesn't spray and green LED flash. Apparently, any liquid with citrus fruits is repellent of these insects.

HOW DOES IT WORK?

1. First, I collected a enough database of images of bees, and spiders. Some I GOT from database websites and others using the Arduino Pro Nicla Vision board.
2. Then I created a project in EDGE IMPULSE which is a development platform for machine learning on edge devices, free for developers.
3. I followed the documentation regarding the Arduino Nicla Vision at: https://docs.edgeimpulse.com/docs/development-platforms/officially-supported-mcu-targets/arduino-nicla-vision
4. I created an AI model with EDGE IMPULSE to distinguish between bees, and spiders; and I deployed this model to the Arduino Pro Nicla Vision Board.

PUBLICATIONS:

1. In Hackaday:
   https://hackaday.io/project/190823-smart-insect-repellent-system
2. In element14:
   https://community.element14.com/challenges-projects/design-challenges/save-the-bees-design-challenge/b/blog/posts/bees-monitor-with-predators-repellent-1---introduction