Register now After registration you will be able to apply for this opportunity online.
Software development for automated study of freshwater organisms
Calibration and software development of an underwater camera-system, designed to observe small river organisms in flow. The goal is to optimize the image processing pipeline and extend the uninterrupted deployment time from 1.5 hrs to a minimum 5 days. Deployments in a river are possible to test the system and developed software improvements.
Keywords: Machine vision, aquatic ecology, software development
Drift is a crucial part of the life cycle of many freshwater organisms, in which juvenile fish and benthic invertebrates use the flow of water to move into new habitat. However, there is much that we still do not know about how organisms use drift to disperse, especially in the context of human exploitation of river and stream landscapes such as hydropower generation. We are developing the Riverine Organism Drift Imager (RODI) to streamline how we sample the drift of aquatic organisms in rivers and streams. RODI will allow us to study the drift of small riverine organisms in unprecedented detail by recording drift within rivers using an underwater camera system. In the current prototype, all recorded images are stored and analysed later during post-processing. As a result, the storage capacity of the hardware limits operation time to 1.5 hours. We are looking for a motivated mechanical/software engineering student with an interest in applications of machine vision to extend the operating time of RODI to several days or weeks by automating the selection of images for storage. The student will perform calibration of the newest iteration of RODI and develop a software pipeline in Python to operate the system while only saving frames that contain an object of interest. Extending the deployment time of RODI will open new possibilities for scientific and practical analysis, facilitating the development of more effective river management strategies.
de Schaetzen, F., Impiö, M., Wagner, B., Nienaltowski, P., Arnold, M., Huber, M., Meyer, M., Raitoharju, J., Silva, L. G. M., & Stocker, R. (2023). The Riverine Organism Drift Imager: A new technology to study organism drift in rivers and streams. Methods in Ecology and Evolution, 14, 2341–2353. https://doi.org/10.1111/2041-210X.14130
Drift is a crucial part of the life cycle of many freshwater organisms, in which juvenile fish and benthic invertebrates use the flow of water to move into new habitat. However, there is much that we still do not know about how organisms use drift to disperse, especially in the context of human exploitation of river and stream landscapes such as hydropower generation. We are developing the Riverine Organism Drift Imager (RODI) to streamline how we sample the drift of aquatic organisms in rivers and streams. RODI will allow us to study the drift of small riverine organisms in unprecedented detail by recording drift within rivers using an underwater camera system. In the current prototype, all recorded images are stored and analysed later during post-processing. As a result, the storage capacity of the hardware limits operation time to 1.5 hours. We are looking for a motivated mechanical/software engineering student with an interest in applications of machine vision to extend the operating time of RODI to several days or weeks by automating the selection of images for storage. The student will perform calibration of the newest iteration of RODI and develop a software pipeline in Python to operate the system while only saving frames that contain an object of interest. Extending the deployment time of RODI will open new possibilities for scientific and practical analysis, facilitating the development of more effective river management strategies.
de Schaetzen, F., Impiö, M., Wagner, B., Nienaltowski, P., Arnold, M., Huber, M., Meyer, M., Raitoharju, J., Silva, L. G. M., & Stocker, R. (2023). The Riverine Organism Drift Imager: A new technology to study organism drift in rivers and streams. Methods in Ecology and Evolution, 14, 2341–2353. https://doi.org/10.1111/2041-210X.14130
Calibrate the novel iteration of the Riverine Organism Drift Imager (RODI) and develop a script in Python to select only images of interest for recording to extend the autonomous field deployment time.
Calibrate the novel iteration of the Riverine Organism Drift Imager (RODI) and develop a script in Python to select only images of interest for recording to extend the autonomous field deployment time.
Dr. Luiz G.M. Silva (lumartins@ethz.ch)
Dr. Frederic de Schaetzen (fdeschae@ethz.ch)
Dr. Luiz G.M. Silva (lumartins@ethz.ch) Dr. Frederic de Schaetzen (fdeschae@ethz.ch)