Deep Census: AUV-Based Scallop Population Monitoring

Christopher Rasmussen, Jiayi Zhao, Danielle Ferraro, Arthur Trembanis; Proceedings of the IEEE International Conference on Computer Vision (ICCV), 2017, pp. 2865-2873


We describe an integrated system for vision-based counting of wild scallops in order to measure population health, particularly pre- and post-dredging in fisheries areas. Sequential images collected by an autonomous underwater vehicle (AUV) are independently analyzed by a convolutional neural network based on the YOLOv2 architecture, which offers state-of-the-art object detection accuracy at real-time speeds. To augment the training dataset, a denoising auto-encoder network is used to automatically upgrade manually-annotated approximate object positions to full bounding boxes, increasing the detection network's performance. The system can act as a tool to improve or even replace an existing offline manual annotation workflow, and is fast enough to function "in the loop" for AUV control.

Related Material

author = {Rasmussen, Christopher and Zhao, Jiayi and Ferraro, Danielle and Trembanis, Arthur},
title = {Deep Census: AUV-Based Scallop Population Monitoring},
booktitle = {Proceedings of the IEEE International Conference on Computer Vision (ICCV) Workshops},
month = {Oct},
year = {2017}