Abstract:Habitat loss is a serious issue threatening biodiversity across the planet, including coastal habitats that support important fish populations. Many coastal areas have been extensively modified by the construction of infrastructure such as ports, seawalls, docks, and armored shorelines. In addition, habitat restoration and enhancement projects often include constructed breakwaters or reefs. Such infrastructure may have incidental or intended habitat values for fish, yet their physical complexity makes quantita… Show more
“…The metrics of MaxN T and species richness had an even lower ratio of 0.5:1, as the entire video can be watched at 2× speed for RUV data in these low abundance habitats. A recent study on RUVs suggested using Frequency of Occurrence, a presence/absence metric derived from species richness, in situations where a quick and robust assessment of fish assemblage composition is required ( Baker et al, 2022 ). Researchers could consider this method over subsampling when fast data processing is required.…”
Section: Discussionmentioning
confidence: 99%
“…Other field and simulation studies have shown that it can be less precise than MaxN, and potentially over-inflate zero counts ( Stobart et al, 2015 ; Campbell et al, 2015 ). Their relative value can change based on useage, as for RUVs in complex habitats there is high correlation between the two for structure-oriented species, but less so for mobile species ( Baker et al, 2022 ).…”
Background
Assessing fish assemblages in subtidal and intertidal habitats is challenging due to the structural complexity of many of these systems. Trapping and collecting are regarded as optimal ways to sample these assemblages, but this method is costly and destructive, so researchers also use video techniques. Underwater visual census and baited remote underwater video stations are commonly used to characterise fish communities in these systems. More passive techniques such as remote underwater video (RUV) may be more appropriate for behavioural studies, or for comparing proximal habitats where the broad attraction caused by bait plumes could be an issue. However, data processing for RUVs can be time consuming and create processing bottlenecks.
Methods
Here, we identified the optimal subsampling method to assess fish assemblages on intertidal oyster reefs using RUV footage and bootstrapping techniques. We quantified how video subsampling effort and method (systematic vs random) affect the accuracy and precision of three different fish assemblage metrics; species richness and two proxies for the total abundance of fish, MaxNT and MeanCountT, which have not been evaluated previously for complex intertidal habitats.
Results
Results suggest that MaxNT and species richness should be recorded in real time, whereas optimal sampling for MeanCountT is every 60 s. Systematic sampling proved to be more accurate and precise than random sampling. This study provides valuable methodology recommendations which are relevant for the use of RUV to assess fish assemblages in a variety of shallow intertidal habitats.
“…The metrics of MaxN T and species richness had an even lower ratio of 0.5:1, as the entire video can be watched at 2× speed for RUV data in these low abundance habitats. A recent study on RUVs suggested using Frequency of Occurrence, a presence/absence metric derived from species richness, in situations where a quick and robust assessment of fish assemblage composition is required ( Baker et al, 2022 ). Researchers could consider this method over subsampling when fast data processing is required.…”
Section: Discussionmentioning
confidence: 99%
“…Other field and simulation studies have shown that it can be less precise than MaxN, and potentially over-inflate zero counts ( Stobart et al, 2015 ; Campbell et al, 2015 ). Their relative value can change based on useage, as for RUVs in complex habitats there is high correlation between the two for structure-oriented species, but less so for mobile species ( Baker et al, 2022 ).…”
Background
Assessing fish assemblages in subtidal and intertidal habitats is challenging due to the structural complexity of many of these systems. Trapping and collecting are regarded as optimal ways to sample these assemblages, but this method is costly and destructive, so researchers also use video techniques. Underwater visual census and baited remote underwater video stations are commonly used to characterise fish communities in these systems. More passive techniques such as remote underwater video (RUV) may be more appropriate for behavioural studies, or for comparing proximal habitats where the broad attraction caused by bait plumes could be an issue. However, data processing for RUVs can be time consuming and create processing bottlenecks.
Methods
Here, we identified the optimal subsampling method to assess fish assemblages on intertidal oyster reefs using RUV footage and bootstrapping techniques. We quantified how video subsampling effort and method (systematic vs random) affect the accuracy and precision of three different fish assemblage metrics; species richness and two proxies for the total abundance of fish, MaxNT and MeanCountT, which have not been evaluated previously for complex intertidal habitats.
Results
Results suggest that MaxNT and species richness should be recorded in real time, whereas optimal sampling for MeanCountT is every 60 s. Systematic sampling proved to be more accurate and precise than random sampling. This study provides valuable methodology recommendations which are relevant for the use of RUV to assess fish assemblages in a variety of shallow intertidal habitats.
“…The experiment presented by (28) cannot be used to detect the exact location and species of fish that the proposed model has performed. The model built by (10) could not detect the species, has reduced performance at a lower resolution, and cannot pinpoint the exact location of the fish. The presented model has addressed all of these parameters.…”
Section: Comparison With the State-of-the-artmentioning
An intelligent detection and recognition model for the fish species from camera footage is urgently required
as fishery contributes to a large portion of the world economy, and these kinds of advanced models can
aid fishermen on a large scale. Such models incorporating a pick-and-place machine can be beneficial to
sorting different fish species in bulk without human intervention, significantly reducing costs for large-scale
fishing industries. Existing methods for detecting and recognizing fish species have many limitations,
such as limited scalability, detection accuracy, failure to detect multiple species, degraded performance at
a lower resolution, or pinpointing the exact location of the fish. Modifying the head of a compelling deep
learning model, namely VGG-16, with pre-trained weights, can be used to detect both the species of the
fish and find the exact location of the fish in an image by implementing a modified YOLO to incorporate
the bounding box regression head. We have proposed using the ESRGAN algorithm and the proposed
neural network to amplify the image resolution by a factor of 4. With this method, an overall detection
accuracy of 96.5% has been obtained. The experiment has been conducted based on a total of 9460
images spread across 9 species. After further improving the model, a pick-and-place machine could be
integrated to quickly sort the fish according to their species in different large-scale fish industries.
“…The experiment presented by [73] cannot be used to detect the exact location and species of fish that the presented model has performed. The model built by [74] could not detect the species, has reduced performance at a lower resolution, and cannot pinpoint the exact location of the fish and all of these parameters have been addressed by the presented model. [75,81] had performance degradation at lower resolution images where the method proposed in the experiment had obtained higher accuracy, which is 96.5% for lower resolution as well.…”
Section: Comparison With the State-of-the-artmentioning
A smart detection and recognition model for the species of a fish from camera footage is of urgent requirement as fishery contributes to a large portion of the world economy and these kinds of advanced models can aid fishermen on a large scale. Such models incorporating a pick-and-place machine can be very useful to sort different fish species in a bulk without human intervention and this can greatly reduce costs for large scale fishing industries. Existing methods for detecting and recognizing fish species have many limitations such as limited scalability, detection accuracy, failing to detect multiple species, degraded performance at a lower resolution, or pinpointing the exact location of the fish. By modifying the head of a very powerful deep learning model namely VGG-16 with pre-trained weights can be used to detect both the species of the fish and find the exact location of the fish in an image by implementing a modified YOLO to incorporate the bounding box regression head. We have proposed the usage of the ESRGAN algorithm along with the proposed neural network to amplify the image resolution by a factor of 4. With this method, the overall detection accuracy of 96.5% has been obtained. The experiment has been conducted based on a total of 9460 images spread across 9 species. After further improving the model, a pick-and-place machine could be integrated for very fast sorting of the fish according to their species in different large-scale fish industries.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.