Intel unveiled the first-generation Neural Compute Stick (NCS) in July 2017, fueling a community of tens of thousands of developers to create artificial intelligence (AI) solutions that have the capability to solve some of the world’s most challenging problems.
Rose Day is one of those innovators. According to the World Bank, between 1990 and 2016, the world lost 502,000 square miles of forest, an area larger than South Africa. To help combat this challenge for future generations, Day developed a unique approach to classify images of the earth’s surface taken by satellite and labelled by location, plants grown and deforested areas.
Day used Intel-optimised TensorFlow for image classification, the Intel Distribution of OpenVINO toolkit to optimise her model, and the Intel Movidius Neural Compute Stick to conduct real-time monitoring of the earth’s surface.
Collected data shows the significant impact of human behaviour on plant life over time, including changes in the types of plants that grow, deforestation effects, and how plant life is forced to adapt. In addition, Day is the winner of the Intel AI Interplanetary Challenge, which tasked students with applying AI to space-related problems.
Another example of an AI innovator is Risab Biswas. At a time when over 821 million people are undernourished, the Food and Agriculture Organisation estimates that nearly 20% of fruit and vegetable supply is lost during production.
To improve crop harvest rates, Biswas developed a computer vision application to help farmers more easily detect pathological disease in their plants. Using Intel’s distribution of the OpenVINO toolkit and Intel RealSense cameras, his methodology yielded a 95% accuracy rate in identifying pathogen attacks in infected plants.
Peter Ma used AI to detect harmful bacteria in water. Every minute, according to the World Health Organisation, a newborn dies from an infection caused by a lack of safe water and an unclean environment.
To alleviate this, developer Ma built an AI application to detect harmful bacteria in water in real-time, without needing to connect to the cloud. His Clean Water AI prototype is powered by the Intel Movidius Neural Compute Stick and has a 95% accuracy rate.