Detection of Algae in Water Bodies Using Computer Vision and Deep Learning

Arabinda Samantaray, Purdue University


The large-scale proliferation of algae, in water bodies all across the globe, is a rising cause for concern amongst researchers, administrators and even common civilians. Harmful Algal Blooms (HABs) have been responsible for the release of hazardous tox- ins in water bodies, contaminating drinking water, killing shellfish, marine mammals and also destroying the aesthetic appeal of many tourist hot-spots. This has resulted in the loss of millions of dollars to the tourism and shing industry and even the deaths of humans. Due to the large scale negative impacts of HABs, it is important to constantly monitor water bodies and identify any algae build-up, so that prompt action against its accumulation can be taken and the harmful consequences can be avoided. While, considerable work has been done in the eld of algae monitoring, current methodologies suffer from various limitations such as requiring signicant manpower, making them economically unfeasible (e.g., it was projected that a budget of $3.5 million was required to monitor the 100 largest lakes in Oklahoma once a month ). Alternatively, methods may depend upon the hardware available, which can be in the form of drones, robotic fish, underwater cameras or satellites, which limits their applicability in varying environmental, topological and socio-economic conditions. To counter the shortcomings of the aforementioned methods, in this paper we propose a computer vision system based on deep learning for algae monitoring. The proposed system is fast, accurate and cheap, and it can be installed on any robotic platforms such as USVs and UAVs for autonomous algae monitoring. The experimental results demonstrate that the proposed system can detect algae in distinct environments regardless of the underlying hardware with high accuracy and in real time.




Min, Purdue University.

Subject Area

Computer science|Artificial intelligence|Robotics

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server