Algorithm based on machine learning will analyze the amount of harmful impurities.

Air pollution is a major environmental problem. About 90% of substances polluting the atmosphere have anthropogenic origin. These are emissions associated with burning fuel, leaking volatile organic compounds, errors in the storage of production waste. There are also non-anthropogenic sources of pollution, such as forest fires.

According to the World Health Organization, annually harmful substances in the air cause premature death of about 7 million people around the world. Atmospheric pollution is associated with cancer and cardiovascular diseases.

The level of air pollution is measured by automatic monitoring stations. The disadvantage of these facilities is their high cost, since expensive high-precision tools are needed for the operation of the station.

Scientists at the University of California at Los Angeles have developed a technology that will track pollution levels with small portable devices. For measurements, only a lensless holographic microscope and a smartphone are needed, naked-science.ru  reports. The microscope processes 6.5 liters of air in 30 seconds, making photos of suspended particles of contaminants. Using a smartphone, images are transferred to a remote server, and the self-learning algorithm analyzes the size and shape of the particles. The gadget is called c-Air.

According to the developers, the accuracy of c-Air measurements is comparable to the accuracy of modern monitoring stations. The relatively low price of devices will allow them to be placed even in hard-to-reach locations, which is important for sparsely populated areas. In large cities, such systems can be combined into a network and receive a huge array of data on atmospheric pollution.

© All rights reserved. Citing to www.ict.az is necessary upon using news