Segmentation of seagrass (Enhalus acoroides) using deep learning mask R-CNN algorithm

Seagrass is an Angiosperms that live in shallow marine waters and estuaries. The method commonly used to identify seagrass is Seagrass-Watch which is done by sampling seagrass or by carrying a seagrass identification book. Technological developments in the era of the industrial revolution 4.0 made it possible to identify seagrass automatically. This research aims to apply the deep learning algorithm to detect seagrass recorded by underwater cameras. Enhalus acoroides seagrass species identification was carried out using a deep learning method with the mask region convolutional neural networks (Mask R-CNN) algorithm. The steps in the research procedure include collecting, labeling, training, testing models, and calculating the seagrass area. This study used 6000 epochs and got a measure of value generated by the model of ± 1.2. The Precision value, namely the model’s ability to correctly classify objects, reached 98.19% and the model’s ability to find all positive objects, based on system testing was able to perform recall is 95.04% and the F1 Score value of 96.58%. The results showed that the MASK R-CNN algorithm could detect and segment seagrass Enhalus acoroides.

A Practical Approach to Monitoring Marine Protected Areas: An Application to El Bajo Espíritu Santo Seamount Near La Paz, Mexico

Worldwide, marine protected areas (MPAs) are increasingly created to protect and restore selected parts of the ocean and to enhance recreation, fishing, and sustainable resources. However, this process has outpaced the development and implementation of methods for assessing and monitoring these habitats. Here, we combine data from an echosounder, a conductivity-temperature-depth probe, and underwater cameras to efficiently survey El Bajo Espíritu Santo Seamount, located in the southwest Gulf of California, Mexico. Results include a bathymetric map detailing a ridge with three peaks; oceanographic profiles showing a 35 m deep mixed layer and anoxic conditions below 200 m; mean target strength estimates for Pacific creolefish, Paranthias colonus (–34.8 dB re 1 m2, for mean total length ~33 cm), and finescale triggerfish, Balistes polylepis (–39.8 dB re 1 m2, 38 cm); baseline estimates of biomass for both species (55.7 t, 95% CI = 30.3–81.2 t and 38.9 t, 95% CI = 21.1–56.6 t, respectively) found only in the oxygenated water near the top of the seamount; and indications that these reef fishes grazed on zooplankton in the mixed layer. We conclude that acoustic-optical sampling is a practical approach for obtaining baseline information on MPAs to efficiently monitor changes resulting from natural and anthropogenic processes.

Bubble PIV technique to measure the velocity field of a free-swimming California sea lion

Fish et al. (2014) adapted laboratory PIV for safe use on larger animals. As opposed to seeding the entire flow with reflective particles and illuminating a plane of the flow with a laser, they produced a sheet of small bubbles and used sunlight for global illumination. Underwater cameras imaged the flow in a method similar to traditional PIV. This technique was used to measure the flow around a swimming dolphin and estimate the thrust produced during a tail stand maneuver (Fish et al. (2014, 2018)). In the current work, we will extend the modification of PIV of Fish et al. to measure the flow produced by a swimming sea lion also using bubbles as seeding particles and sunlight as illumination. This is the first time that the flowfield of a swimming sea lion has been directly measured. We will present an extensive extension to the image processing required to measure flow under field conditions. Finally, we will present the flow generated by propulsive strokes of an adult female (Cali) sea lion freely swimming through a pool of stationary water.

A Comparison of vision-based and CNN-based detector for fish monitoring in complex environment

Aquaculture farming can help soften the environmental impact of overfishing by fulfilling seafood demands with farmed fishes. However, to maintain big scale farms can be challenging, even with the help of underwater cameras affixed in farm cages, because there are hours’ worth of footages to sift through, which can be a laborious task if performed manually. Vision-based system therefore could be deployed to filter useful information from video footages, automatically. This work proposed to solve the above mentioned problems by deploying the; 1) Extended UTAR Aquaculture Farm Fish Monitoring System Framework (UFFMS), being the handcrafted method, and 2) Faster Region Convolutional Neural Network (Faster RCNN), being the CNN-based method, for fish detection. These two methods could extract information about fishes from video footages. Experimental results show that for well-lit footages, Faster RCNN performs better, compared to the extended-UFFMS. However, accuracy of Faster RCNN drops drastically for non-well-lit footages, at an average of 28.57%, despite still having perfect precision scores.

Autonomous Underwater Vehicles and Field of View in Underwater Operations

Submarine inspections and surveys require underwater vehicles to operate in deep waters efficiently, safely and reliably. Autonomous Underwater Vehicles employing advanced navigation and control systems present several advantages. Robust control algorithms and novel improvements in positioning and navigation are needed to optimize underwater operations. This paper proposes a new general formulation of this problem together with a basic approach for the management of deep underwater operations. This approach considers the field of view and the operational requirements as a fundamental input in the development of the trajectory in the autonomous guidance system. The constraints and involved variables are also defined, providing more accurate modelling compared with traditional formulations of the positioning system. Different case studies are presented based on commercial underwater cameras/sonars, analysing the influence of the main variables in the measurement process to obtain optimal resolution results. The application of this approach in autonomous underwater operations ensures suitable data acquisition processes according to the payload installed onboard.

Backstroke to Breaststroke Turning Performance in Age-Group Swimmers: Hydrodynamic Characteristics and Pull-Out Strategy

We compared the hydrodynamic characteristics and pull-out strategies of four backstroke-to-breaststroke turning techniques in young swimmers. Eighteen 11 and 12-year-old swimmers participated in a 4 week intervention program including 16 contextual interference sessions. The hydrodynamic variables were assessed through inverse dynamics, and the pull-out strategy kinematics were assessed with tracking markers followed by 12 land cameras and 11 underwater cameras. Swimmers randomly completed sixteen 30 m maximal backstroke-to breaststroke-open, somersault, bucket and crossover turns (four in each technique) with a 3 min rest. The data showed higher drag force, cross-sectional area and drag coefficient values for the first (compared with the second) gliding position. The crossover turn revealed the highest push-off velocity (2.17 ± 0.05 m·s−1), and the somersault turn demonstrated the lowest foot plant index (0.68 ± 0.03; 68%), which could have affected the first gliding, transition and second gliding depths (0.73 ± 0.13, 0.86 ± 0.17 and 0.76 ± 0.17 m). The data revealed the consistency of the time spent (4.86 ± 0.98 s) and breakout distance (6.04 ± 0.94 m) among the four turning techniques, and no differences were observed between them regarding time and average velocity up to 7.5 m. The hydrodynamic characteristics and pull-out strategy of the backstroke-to-breaststroke turns performed by the age group swimmers were independent of the selected technique.

Rapid coral reef assessment using 3D modelling and acoustics: acoustic indices correlate to fish abundance, diversity and environmental indicators in West Papua, Indonesia

Background Providing coral reef systems with the greatest chance of survival requires effective assessment and monitoring to guide management at a range of scales from community to government. The development of rapid monitoring approaches amenable to collection at community level, yet recognised by policymakers, remains a challenge. Technologies can increase the scope of data collection. Two promising visual and audio approaches are (i) 3D habitat models, generated through photogrammetry from video footage, providing assessment of coral cover structural metrics and (ii) audio, from which acoustic indices shown to correlate to vertebrate and invertebrate diversity, can be extracted. Methods We collected audio and video imagery using low cost underwater cameras (GoPro Hero7™) from 34 reef samples from West Papua (Indonesia). Using photogrammetry one camera was used to generate 3D models of 4 m2 reef, the other was used to estimate fish abundance and collect audio to generate acoustic indices. We investigated relationships between acoustic metrics, fish abundance/diversity/functional groups, live coral cover and reef structural metrics. Results Generalized linear modelling identified significant but weak correlations between live coral cover and structural metrics extracted from 3D models and stronger relationships between live coral and fish abundance. Acoustic indices correlated to fish abundance, species richness and reef functional metrics associated with overfishing and algal control. Acoustic Evenness (1,200–11,000 Hz) and Root Mean Square RMS (100–1,200 Hz) were the best individual predictors overall suggesting traditional bioacoustic indices, providing information on sound energy and the variability in sound levels in specific frequency bands, can contribute to reef assessment. Conclusion Acoustics and 3D modelling contribute to low-cost, rapid reef assessment tools, amenable to community-level data collection, and generate information for coral reef management. Future work should explore whether 3D models of standardised transects and acoustic indices generated from low cost underwater cameras can replicate or support ‘gold standard’ reef assessment methodologies recognised by policy makers in marine management.

Kinematic Variables of Disabled Swimmers and Their Correlation With the International Paralympic Committee Classification

This study described the kinematic variables of disabled swimmers’ performance and correlated them with their functional classification. Twenty-one impaired swimmers (S5–S10) performed 50-m maximum front-crawl swimming while being recorded by four underwater cameras. Swimming velocity, stroke rate, stroke length, intracycle velocity variation, stroke dimensions, hand velocity, and coordination index were analyzed. Kendall rank was used to correlate stroke parameters and functional classification with p < .05. Swimming velocity, stroke length, and submerged phase were positively correlated with the para swimmers functional classification (.61, .50, and .41; p < .05, respectively), while stroke rate, velocity hand for each phase, coordination index, and intracyclic velocity variation were not (τ between −.11 and .45; p > .05). Thus, some objective kinematic variables of the impaired swimmers help to support current classification. Improving hand velocity seems to be a crucial point to be improved among disabled swimmers.

Macroinvertebrata Recruitments in Artificial Reef After Two Years of Sinking on The Damas Beach, Trenggalek

The condition of coral reefs in Damas Beach is included in the bad category because there are many fragments of coral reefs (rubble) caused by fishing nets caught in the reef. Based on the problem of coral reefs contained in the waters of Damas Beach, there are innovations to sink artificial reefs. The condition of artificial reefs that have been derived in the waters of Damas Beach within 2.5 years needs to be monitored to find out the macroinvertebrates that stick to the artificial reefs. Retrieval of macroinvertebrate data using the underwater photo transect method with the help of underwater cameras to produce observational data. The results found 10 types of macroinvertebrates with a total of 9,293 individuals. The composition of attached macroinvertebrates is dominated by barnacles by 66% and the density of macroinvertebrates is dominated by barnacles by 4.50 ind / m² out of a total density of 6.88 ind / m².