Computer vision in aquaculture: a case study of juvenile fish counting.

Autor:	Babu KM; The New Zealand Institute for Plant and Food Research Limited, Lincoln, New Zealand., Bentall D; The New Zealand Institute for Plant and Food Research Limited, Lincoln, New Zealand., Ashton DT; The New Zealand Institute for Plant and Food Research Limited, Nelson, New Zealand., Puklowski M; The New Zealand Institute for Plant and Food Research Limited, Nelson, New Zealand., Fantham W; The New Zealand Institute for Plant and Food Research Limited, Nelson, New Zealand., Lin HT; The New Zealand Institute for Plant and Food Research Limited, Lincoln, New Zealand., Tuckey NPL; The New Zealand Institute for Plant and Food Research Limited, Nelson, New Zealand., Wellenreuther M; The New Zealand Institute for Plant and Food Research Limited, Nelson, New Zealand.; The University of Auckland, Auckland, New Zealand., Jesson LK; The New Zealand Institute for Plant and Food Research Limited, Lincoln, New Zealand.
Jazyk:	angličtina
Zdroj:	Journal of the Royal Society of New Zealand [J R Soc N Z] 2022 Aug 03; Vol. 53 (1), pp. 52-68. Date of Electronic Publication: 2022 Aug 03 (Print Publication: 2023).
DOI:	10.1080/03036758.2022.2101484
Abstrakt:	In aquaculture breeding or production programmes, counting juvenile fish represents a considerable cost in terms of the human hours needed. In this study, we explored the use of two state-of-the-art machine learning architectures (Single Shot Detection, hereafter SSD and Faster Regions with convolutional neural networks, hereafter Faster R-CNN) to augment a manual image-based juvenile fish counting method for the Australasian snapper ( Chrysophrys auratus ) bred at The New Zealand Institute for Plant and Food Research Limited. We tested model accuracy after tuning for confidence thresholds and non-maximal suppression overlap parameters, and implementing a bias correction using a Poisson regression model. Validation of image data showed that after tuning, bias-corrected SSD and Faster R-CNN models had mean absolute percent errors (MAPE) of less than 10%, with SSD having MAPE of less than 5%. Comparison of the results with those from manual counts showed that, while manual counts are slightly more accurate (MAPE = 1.56), the machine learning methods allow for more rapid assessment of counts and thus facilitating a higher throughput. This work represents a first step for deploying machine learning applications to an existing real-life aquaculture scenario and provides a useful starting point for further developments, such as real-time counting of fish or collecting additional phenotypic data from the source images. Competing Interests: No potential conflict of interest was reported by the author(s). (© 2022 The New Zealand Institute for Plant and Food Research Limited.)
Databáze:	MEDLINE
Externí odkaz:	Zobrazit plný text záznamu