Global challenges resolved with Intel AIAug 20, 2019 | 106 views
3 Major Global Challenges Solved cA Intel launched its first generation Intel Neural Compute Stick (Intel NCS) in July 2017, encouraging tens of thousands of developers to create AI solutions that can solve some of the world's greatest challenges. Check out three of the challenges that can already be solved with technology.
Analyzing deforestation and urbanization
Innovative: Rose Day
Project: According to World Bank information, between 1990 and 2016 the world lost about 1.3 million square kilometers of forest areas, a larger area than South Africa. To help combat this challenge for future generations, Rose Day developed a unique approach to satellite imagery of the Earth's surface and divided by site, crop plants and deforested areas.
Rose used Intel's optimized TensorFlow technology to classify images, the Intel Distribution of OpenVINO kit to enhance its model, and the Intel Movidius Neural Compute Stick for real-time monitoring of the Earth's surface. The data collected show the significant impact of human behavior on plant life over time, including changes in vegetation types, deforestation effects and how plant life is forced to adapt. In addition, Day won the Intel AI Interplanetary Challenge, which challenged students to apply IA to solve space-related problems.
Identification of pathologies in plants
Innovative: Risab Biswasom the Intel AI
Project: At a time when more than 821 million people are malnourished, the Food and Agriculture Organization of the United Nations estimates that almost 20 percent of the fruit and vegetable supply is lost during production.
To improve harvest rates, Risab Biswas developed a computer vision application to help farmers more easily detect diseases in their plants. Using the Intel Distribution of OpenVINO kit and Intel RealSense cameras, the innovative methodology presented a 95% accuracy rate in identifying pathogen attacks on infected plants.
Detection of harmful bacteria in water in real time
Innovative: Peter Ma
Project: According to the World Health Organization (WHO), every minute a newborn baby dies because of infections caused by lack of clean water and hygiene. To improve this scenario, developer Peter Ma has created an application that uses AI to detect harmful bacteria in the water in real time without the need to connect to the cloud. The IA Clean Water prototype is powered by Intel's Movidius Neural Compute Stick and has an accuracy rate of 95%.