Location: Cool and Cold Water Aquaculture ResearchTitle: MortCam: An Artificial Intelligence-aided fish mortality detection and alert system for recirculating aquaculture
|RANJAN, RAKESH - Freshwater Institute
|SHARRER, KATA - Freshwater Institute
|TSUKUDA, SCOTT - Freshwater Institute
|GOOD, CHRISTOPHER - Freshwater Institute
Submitted to: Aquacultural Engineering
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/9/2023
Publication Date: 5/15/2023
Citation: Ranjan, R., Sharrer, K., Tsukuda, S., Good, C. 2023. MortCam: An Artificial Intelligence-aided fish mortality detection and alert system for recirculating aquaculture. Aquacultural Engineering. 102:102341. https://doi.org/10.1016/j.aquaeng.2023.102341.
Interpretive Summary: Unusual mortality patterns may be an indicator of abiotic or biotic stresses on fish. Automated real or near real-time mortality tracking in recirculating aquaculture system (RAS) could provide valuable inputs to farm managers to make informed management decisions to address the root cause of the issue, and thereby minimize mortality and prevent mass mortality events, but this technology is currently not available for producers. This study was conducted to develop an artificial intelligence-aided mortality alert system (‘MortCam’) to enable real-time mortality tracking in commercial RAS tanks. After optimization, the system performed well in terms of mortality count accuracy and was robust to changes in imaging conditions. This system can help in making informed decisions on mortality management and provide valuable inputs to farm managers by automatically logging the instantaneous, daily, and cumulative mortality data.
Technical Abstract: Mortality is an important production and fish welfare indicator in aquaculture. Unusual mortality patterns can be associated with abiotic or/and biotic stresses on fish in recirculating aquaculture systems (RAS). Real or near real-time mortality tracking can provide valuable inputs to farm managers, to inform RAS management decisions and address root causes in an effort to prevent mass mortality events. While traditional systems use infrequent human operator observation and tracking - often in conjunction with an underwater camera - the proposed tool (i.e., ‘MortCam’) augments this approach with Artificial Intelligence (AI) and Internet of Things (IoT) deployed at the Edge to provide round-the-clock mortality monitoring to trigger alerts when mortality thresholds are exceeded. MortCam consists of an imaging sensor integrated with an edge computing device, customized for underwater applications. MortCam was deployed in a 150 m3 circular dual-drain RAS tank at 0.6 m above the bottom drain plate to acquire the imagery data in both ambient and supplemental light conditions. The images were collected every fifteen minutes for 90 days. Acquired images were annotated as ‘alive’ and ‘dead’ fish and split into training (70%), validation (20%), and test (10%) datasets to train a custom Yolo V7 mortality detection model. The optimized mixed model achieved a mean average precision (mAP) and F1 score of 93.4% and 0.89, respectively. Additionally, the model performed well in terms of mortality count and was found robust toward the change in the imaging conditions. The model was deployed on the MortCam to achieve round-the-clock autonomous mortality monitoring. The system reliably generated email and text alerts to notify fish production staff of unusual mortality events.