Using automated video analysis to study fish escapement through escape panels in active fishing gears: Application to the effect of net colour

Abstract Quantification of the escapement rate of unwanted fish through a selective device is usually based on catch comparison. This study proposes a new, time efficient method to automatically compare two selective devices by automated counting of fish escapements through each selective device based on video sequences. First, sea trials were conducted to record video sequences of fish escaping a white and black square mesh panel. Then, all of the underwater sequences were automatically analysed by a computer vision software for automated object detection and tracking. Finally, the algorithm was assessed using 150 min of video sequences analysed by humans. We observed that the variability in escapements rate between all the observers on reference video sequences could reach 5%. As the difference in escapements rate between the algorithm and the observers was lower than the variability between observers, the automated approach was validated. The software detected a significant difference in fish escapement rate according to the net colour in the camera field of view: 60% of all fish escaped through the white panel. Our results suggest that net colour influences the escape rates of fish. The colour of the selective device should therefore be investigated further with the aim of increasing their efficiency. Further development of the software could be done to identify species and size of the fish and assess the effectiveness of a selective device by species and size.

[1]  J. Robertson,et al.  The effect of twine thickness on cod-end selectivity of trawls for haddock in the North Sea , 1996 .

[2]  Jenq-Neng Hwang,et al.  Tracking Live Fish From Low-Contrast and Low-Frame-Rate Stereo Videos , 2015, IEEE Transactions on Circuits and Systems for Video Technology.

[3]  Pascal Poncin,et al.  Video multitracking of fish behaviour: a synthesis and future perspectives , 2013 .

[4]  R. Fryer,et al.  Square mesh panels in demersal trawls: further data relating haddock and whiting selectivity to panel position , 2003 .

[5]  Ludvig Ahm Krag,et al.  Improving escape panel selectivity in Nephrops-directed fisheries by actively stimulating fish behavior , 2017 .

[6]  Alessandro Lucchetti,et al.  The influence of twine thickness on the size selectivity of polyamide codends in a Mediterranean bottom trawl , 2007 .

[7]  C. S. Wardle,et al.  Comparison of the reactions of fish to a trawl gear, at high and low light intensities , 1989 .

[8]  Dominique Pelletier,et al.  Underwater video techniques for observing coastal marine biodiversity: A review of sixty years of publications (1952–2012) , 2014 .

[9]  C. S. Wardle,et al.  Studies on the use of visual stimuli to control fish escape from codends. II. The effect of a black tunnel on the reaction behaviour of fish in otter trawl codends , 1995 .

[10]  Norval J. C. Strachan,et al.  Automated measurement of species and length of fish by computer vision , 2006 .

[11]  Ling Shao,et al.  Recent advances and trends in visual tracking: A review , 2011, Neurocomputing.

[12]  N. Rawlinson,et al.  An assessment of Bycatch Reduction Devices in a tropical Australian prawn trawl fishery , 1998 .

[13]  R. J. Fryer,et al.  Selectivity of a 120 mm diamond cod-end and the effect of inserting a rigid grid or a square mesh panel , 2004 .

[14]  Shale Rosen,et al.  DeepVision in-trawl imaging: Sampling the water column in four dimensions , 2013 .

[15]  T. Thünken,et al.  Visual prey detection by near-infrared cues in a fish , 2012, Naturwissenschaften.

[16]  Xi En Cheng,et al.  Automatically Detect and Track Multiple Fish Swimming in Shallow Water with Frequent Occlusion , 2014, PloS one.

[17]  Pietro Perona,et al.  Automated image-based tracking and its application in ecology. , 2014, Trends in ecology & evolution.

[18]  Nils Olav Handegard,et al.  Fish species identification using a convolutional neural network trained on synthetic data , 2018, ICES Journal of Marine Science.

[19]  K. Stokesbury,et al.  Development of a Video Trawl Survey System for New England Groundfish , 2017 .

[20]  RosenShale,et al.  DeepVision: a stereo camera system provides highly accurate counts and lengths of fish passing inside a trawl , 2013 .

[21]  Dorothée Kopp,et al.  Using underwater video to assess megabenthic community vulnerability to trawling in the Grande Vasière (Bay of Biscay) , 2017, Environmental Conservation.

[22]  Camille Vogel,et al.  From discard ban to exemption: How can gear technology help reduce catches of undersized Nephrops and hake in the Bay of Biscay trawling fleet? , 2017, Journal of environmental management.

[23]  R. Kynoch,et al.  Square mesh panels in demersal trawls: some data on haddock selectivity in relation to mesh size and position , 2001 .

[24]  Junita Diana Karlsen,et al.  Understanding the release efficiency of Atlantic cod (Gadus morhua) from trawls with a square mesh panel: effects of panel area, panel position, and stimulation of escape response , 2015 .

[25]  C. S. Wardle,et al.  Behavioural studies of the principles underlying mesh penetration by fish , 1993 .

[26]  N. Madsen,et al.  A study of fish behaviour in the extension of a demersal trawl using a multi-compartment separator frame and SIT camera system , 2009 .

[27]  Ajmal Mian,et al.  A review of techniques for the identification and measurement of fish in underwater stereo-video image sequences , 2013, Optical Metrology.

[28]  Steven H. D. Haddock,et al.  Using red light for in situ observations of deep-sea fishes , 2005 .

[29]  R. Fryer,et al.  The effects of mesh size, cod-end extension length and cod-end diameter on the selectivity of Scottish trawls and seines , 1992 .