Building IoT sensors to estimate PM2.5 concentrations
Ting‐Yu Chen, Li-Wei Lo, Yan-Ting Lin, Jiun‐Jian Liaw · 2022 International Conference on Electrical, Computer and Energy Technologies (ICECET) · 2022
Air pollution is getting worse now, and suspended particles are also increasing. With the progress of technology and sensing technology, small and low-cost sensing platforms appear one after another. However, if small sensors require a large number of arrangements, it is still a big burden. In this study, this research use Camera Module V2 image sensor to capture images, and the image features are found by image processing. Based on the Raspberry Pi v4 test platform, the image sensor set up outdoors to shots the estimated area in fixed time, and stored the image in MySQL for use of data, using crawler to capture the feature of the weather in the monitoring station, extracting feature values from the collected images, and input the feature values into support vector regression (SVR) to estimate PM2.5concentration. The result of experiment shows that the image cannot be characterized by the visibility because the distance between the monitor and building is not enough.