Evaluating the use of onboard cameras in the Shark Gillnet Fishery in South Australia



Download 4.23 Mb.
Page5/13
Date28.03.2018
Size4.23 Mb.
#43396
1   2   3   4   5   6   7   8   9   ...   13

2.1.1.3 Footage analyses

Quality


Footage quality was analysed for the 640 net hauls recorded during this trial (corresponding to 274 fishing days). Records were made of any issues that affected an analyst’s ability to review the video footage, including visibility level, presence of camera obstruction, recording failure during a fishing event and lighting problems. The frequency of issues experienced by an analyst was later used to assess how well electronic monitoring systems recorded information on fishing activities.

Marine mammals


A total of 640 individual net hauls were recorded and analysed to obtain the number of direct interactions between marine mammals and gillnets. This included instances where animals fell out or were otherwise removed from the net between the water surface and the boat. Electronic monitoring system footage was analysed by a number of analysts that included an independent data analyst (D&S Data Fix), and an experienced AFMA observer. The independent data analyst with low-level species identification skills took screen shots and captured short clips of relevant video when interactions were detected. These short clips were subsequently reviewed by the experienced AFMA observer to identify individuals to the finest taxonomical level possible. As analysts gain further experience and expertise in species identification it is possible that identification could be done during the analysis phase.

Analysis of electronic monitoring footage was undertaken using VLC Media Player and electronic monitoring Interpret Pro™, which is a dedicated video analysis software package produced by AMR. Both pieces of software allowed electronic monitoring system footage to be reviewed at different speeds. Due to the large size of marine mammals, which allows easy detection in the video footage, speeds up to 6 times that of real-time were used for video analyses to minimise costs. Speed was slowed down when interactions were detected and analysed on a frame-by-frame basis to identify individuals to the finest taxonomical level possible. Time and position information was recorded for each interaction as well as the nature of the interactions.





Download 4.23 Mb.

Share with your friends:
1   2   3   4   5   6   7   8   9   ...   13




The database is protected by copyright ©ininet.org 2024
send message

    Main page