Decentralized Machine Learning Intelligence > Topics > Intelligent Fisheries Management

Intelligent Fisheries Management

Fisheries management aims to balance the sustainability of fisheries stocks and the impacts of fishing on the environment with the economic opportunities they offer. Annual estimates of removals from fish stocks are a fundamental requirement for any fisheries management regime. The collection of quantitative information from the recreational fishing sector is especially problematic, as these fisheries are poorly defined, widespread, and are not subject to any form of reporting regime.

New Zealand’s National Institute of Water and Atmospheric Research (NIWA) has established a network of web cameras overlooking key boat ramps throughout New Zealand, on behalf of the Ministry for Primary Industries (MPI), to monitor trends in recreational fishing efforts over time. In this monitoring system, one image is captured per minute for each web camera, providing 1440 images of a monitored ramp each day. These images are viewed in series by a technician who manually interprets the images and records a count of returning boats for that day. This process is very onerous and time-consuming; moreover, manual counting is subjective, so the counting results are unstable in some complicated cases.

The objective of this project is to automate the process of counting boats returning to boat ramps and interpret web camera images using computer vision based 24-hour boat-flow analysis. This research is funded by MPI in collaboration with NIWA.