Virtual Stream Stage Sensor Using Projected Geometry and Augmented Reality for Crowdsourcing Citizen Science Applications

Accurately measuring the surface level of a river is a vital component of environmental monitoring and modeling efforts. Reliable data points are required for calibrating the statistical models that are used for, among other things, flood prediction and model validation. While current embedded monitoring systems provide accurate measurements, the cost to replicate this current system on a large scale is prohibitively expensive, limiting the quantity of data available. In this project, we describe a new method to accurately measure river levels using smartphone sensors. We take three pictures of the same point on the river's surface and perform calculations based on the GPS location and spatial orientation of the smartphone for each picture using projected geometry. Augmented reality is used to improve the accuracy of smartphone sensor readings. This proposed implementation is significantly cheaper than existing water measuring systems while offering similar accuracy. Additionally, since the measurements are taken by sensors that are commonly found in smartphones, crowdsourcing the collection of river measurements to citizen-scientists is possible. Thus, our proposed method leads to a much higher quantity of reliable data points than currently possible at a fraction of the cost. Sample runs and an analysis of the results are included. The presentation concludes with a discussion of future work, including applications to other fields and plans to implement a fully automated system using this method in tandem with image recognition and machine learning.