IoT-Based Aquatic Pollution Monitoring using Convolutional Neural Networks

Darian Daji
2019 International Journal for Research in Applied Science and Engineering Technology  
The continuously decreasing quality of drinking water is of great concern across the globe. Due to the incessant mixing of unprocessed industrial and domestic wastes, the contamination rate of water is rapidly increasing.The contaminations are of physical, organic, and chemical nature. Owing to this rise in pollution, the idea proposed is to build a model equipped with a visual device such as a camera, and other sensors, that by means of floating on water, can constantly monitorthe water body.
more » ... sing techniques of computer vision, floating or other debris is detected on the surface of water. The use of sensors can help to keep in check the physical as well as the chemical parameters of water. Conditions like temperature, pH, turbidity, conductivity etcetera can be used either directly or indirectly to judge the overall quality of water.The optical device captures real-time images and a debris detection algorithm using Convolutional Neural Network (CNN) sends vital measurements to the user. The processed dataafter converting intoa user-understandable format can be used to get a general idea of whether the water is safe for consumption, that is, to determine whether it is potable or could be made potable by some treatment. Secondly, it could be used to monitor the balance in marine ecosystems.
doi:10.22214/ijraset.2019.10081 fatcat:ahen3qojanflhny4t65klcgyai