Mesure de débit en rivière : analyse et optimisation du projet riverApp

Details

Supervisors

Soares Frazao, Sandra ; Gousenbourger, Pierre-Yves

Faculty

Ecole polytechnique de Louvain
Ecole polytechnique de Louvain

Degree label

Master [120] : ingénieur civil en mathématiques appliquées, à finalité spécialisée
Master [120] : ingénieur civil en informatique, à finalité spécialisée

Abstract

enThe measurement of river flow rate allows estimating the amount of water that flows through that river during a specific time interval. In this work, the riverApp application is used as a tool to calculate the flow rate from a video of the river. This application was developed by students at UCLouvain. The objective of this work is to make modifications to the application to improve its performances. This report begins with an introduction to flow rate measurement in rivers using Particle Image Velocimetry (PIV) analysis on a video. Then, a comparison is made between riverApp and pyOrc, another application with a similar purpose. This comparison concludes that riverApp is less efficient than pyOrc. A new version of riverApp is implemented based on the pyOrc project, and it is presented with its new pipeline structure. With the new implementation of riverApp, an additional feature is introduced. When launching the application, the user needs to mark 4 reference points on the video. With the new functionality, these 4 reference points are automatically detected. The algorithm relies on image processing techniques such as mathematical morphology, including erosion and dilation. This report also provides a comprehensive description of the final state of the riverApp application and potential future improvements.