Abstract
Monitoring and understanding fish behavior is crucial for achieving precision in everyday husbandry practices (i.e. for optimizing farm performance), and for improving fish welfare in aquaculture. Various intelligent monitoring and control methods, using mathematical models, acoustic methods and computer vision, have been recently developed for this reason. Here, a tracking algorithm based on computer vision that extracts short trajectories of individual European seabass in both recirculating aquaculture systems and sea cages was developed using videos from network cameras. Using this methodology, parameters such as instantaneous normalized speed, travel direction and preference for the tank surface by European seabass could be quantified. When testing the sensitivity of this algorithm for detecting fish swimming variations under different husbandry scenarios, we found that the algorithm could detect variations in all of the abovementioned parameters and could potentially be a useful tool for monitoring the behavioral state of European seabass.
Highlights
Fish can display a wide spectrum of behavioral patterns that emerge from complex interactions with their conspecifics and their environment (Brown, 2015; Macaulay et al, 2021a)
The aims of the current study were to: (a) develop automated routines that can track European seabass (Dicentrarchus labrax) with individual-level approaches in recirculating aquaculture systems (RAS) and sea cages using single cameras; (b) extract feature parameters that could be used for the detection of variations in swimming behavior; and (c) provide application examples of the developed methodology using different husbandry scenarios
Pre-processing The image frame is converted to grayscale, and contrast limited adaptive histogram equalization (CLAHE) is applied to smooth the effect of lighting on the image and achieve local contrast enhancement
Summary
Fish can display a wide spectrum of behavioral patterns that emerge from complex interactions with their conspecifics and their environment (Brown, 2015; Macaulay et al, 2021a). Changes in behavior can result from different environmental or physiological conditions and these could be used as an indicator of fish welfare. The method detects as foreground any significant background motion caused by sudden light variation or the irrelevant motion of air bubbles or organic particles. To filter out this falsely detected foreground and keep only the true fish objects, morphological operations, and contour analysis (shape and size filtering) are applied (see contour analysis section). At each iteration, each pixel’s i value (I(i)) in the original image is replaced according to Equation 1 and a binary foreground image is extracted
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
