In October we published a paper on a new low cost method for measuring SO2 using Raspberry Pi Cameras and the Raspberry Pi platform, where we demonstrated its function at a power station, this was also the subject of my previous blog post. However, in this previous paper we had yet to test our camera against a commericial (and much more expensive!) UV (ultraviolet) camera to compare and contrast camera performance and error. This is an essential step which will give us confidence in the ability of our UV camera. We therefore took a commercial JAI camera system and our cameras to the flanks of Mt. Etna, one of the most reliable volcanoes to measure at, and spent a week measuring SO2 fluxes. This was during my "Back to Etna" blog posts during July '16.
We conducted a number of tests including side-by-side measurements of degassing from the North-East Crater of Mt. Etna. This demonstrated that, following full processing (e.g., camera calibration using gas cells), both cameras produced almost identical flux traces. In addition, we tested and investigated a range of other parameters including signal-to-noise ratios and detection limits. We conclude that while the commerical camera may be slightly more sensitive in certain circumstances, our lower cost camera based on the Pi Camera should perform admirably for the overwhelming majority of volcanic applications. This is particularly significant given the large reduction in cost of our system (approximately 10 times cheaper) compared to a commerical grade one. The video below shows SO2 concentrations (higher concentrations are red colours) as imaged using the Pi Camera.
The full published paper, which is Open Access was published in the journal Remote Sensing today "A Low-Cost Smartphone Sensor-Based UV Camera for Volcanic SO2 Emission Measurements".
Finally, happy new year!
We conducted a number of tests including side-by-side measurements of degassing from the North-East Crater of Mt. Etna. This demonstrated that, following full processing (e.g., camera calibration using gas cells), both cameras produced almost identical flux traces. In addition, we tested and investigated a range of other parameters including signal-to-noise ratios and detection limits. We conclude that while the commerical camera may be slightly more sensitive in certain circumstances, our lower cost camera based on the Pi Camera should perform admirably for the overwhelming majority of volcanic applications. This is particularly significant given the large reduction in cost of our system (approximately 10 times cheaper) compared to a commerical grade one. The video below shows SO2 concentrations (higher concentrations are red colours) as imaged using the Pi Camera.
The full published paper, which is Open Access was published in the journal Remote Sensing today "A Low-Cost Smartphone Sensor-Based UV Camera for Volcanic SO2 Emission Measurements".
Finally, happy new year!
RSS Feed
