Automated detection of wildlife using drones: Synthesis, opportunities and constraints

Methods in Ecology and Evolution, Journal Year: 2021, Volume and Issue: 12(6), P. 1103 - 1114

Published: Feb. 22, 2021

Abstract Accurate detection of individual animals is integral to the management vulnerable wildlife species, but often difficult and costly achieve for species that occur over wide or inaccessible areas engage in cryptic behaviours. There a growing acceptance use drones (also known as unmanned aerial vehicles, UAVs remotely piloted aircraft systems, RPAS) detect wildlife, largely because capacity rapidly cover large compared ground survey methods. While can aid capture amounts imagery, requires either manual evaluation imagery automated using machine learning algorithms. drone‐acquired possible sometimes necessary, powerful combination with this much faster and, some cases, more accurate than human observers. Despite great potential emerging approach, most attention date has been paid development algorithms, little about constraints around successful (P. W. J. Baxter, G. Hamilton, 2018, Ecosphere , 9 e02194). We reviewed studies were conducted last 5 years which detected automatically understand how technological constraints, environmental conditions ecological traits target impact From review, we found could be achieved wider range under greater variety reported previous reviews imagery. A high probability efficiently fixed‐wing platforms RGB sensors occurred open homogeneous environments vegetation variation topography while infrared multirotor necessary successfully small, elusive complex habitats. The insight gained review allow conservation managers algorithms accurately conduct abundance data on populations critical their conservation.

Language: Английский

---

The Drone Revolution of Shark Science: A Review

Drones, Journal Year: 2021, Volume and Issue: 5(1), P. 8 - 8

Published: Jan. 21, 2021

Over the past decade, drones have become a popular tool for wildlife management and research. Drones shown significant value animals that were often difficult or dangerous to study using traditional survey methods. In five years drone technology has commonplace shark research with their use above, more recently, below water helping minimise knowledge gaps about these cryptic species. enhanced our understanding of behaviour are critically important tools, not only due importance conservation in ecosystem, but also help encounters humans. To provide some guidance future relation sharks, this review provides an overview how currently used critical context monitoring. We show been fill around fundamental behaviours movements, social interactions, predation across multiple species scenarios. further detail advancement sensors, automation, artificial intelligence improving abilities data collection analysis opening opportunities shark-related beach safety. An investigation shark-based potential underwater (ROV/AUV) is provided. Finally, baseline observations pioneered recommendations might be enhance future.

Language: Английский

---

Opportunities and risks in the use of drones for studying animal behaviour

Methods in Ecology and Evolution, Journal Year: 2022, Volume and Issue: 14(8), P. 1864 - 1872

Published: June 18, 2022

Abstract In the last decade, drones have become an affordable technology offering highly mobile aerial platforms that can carry a range of sensory equipment into hitherto uncharted areas. Drones thus widely applicable tool for surveying animal populations and habitats to assist conservation efforts or study behavioural ecology species by monitoring individual group behaviour. Here, we review current applications drone surveys potential recently developed computer algorithms automatic detection tracking in footage. We further which factors are reportedly associated with disturbance during presentations how may be used anti‐predator Drone their environments allow scientists create digital terrain models habitats, estimate abundance, monitor behaviour composition, spatial organization movement groups. As influence many bird mammal directly, they also provide experimental responses novel situations, including itself. conclude combined use automated software population estimates opens new possibilities collective With regard drone‐related as predator models, recommend interpret results against background population‐specific predation pressure sources anthropogenic disturbance.

Language: Английский

---

A Survey on Applications of Unmanned Aerial Vehicles Using Machine Learning

IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 117582 - 117621

Published: Jan. 1, 2023

Unmanned Aerial Vehicles (UAVs) play an important role in many applications, including health, transport, telecommunications and safe rescue operations. Their adoption can improve the speed precision of applications when compared to traditional solutions based on handwork. The use UAVs brings scientific technological challenges. In this context, Machine Learning (ML) techniques provide several problems concerning civil military applications. An increasing number papers ML context have been published academic journals. work, we present a literature review UAVs, outlining most recurrent areas commonly used UAV results reveal that environment, communication security are among main research topics.

Language: Английский

---

Counting animals in aerial images with a density map estimation model

Ecology and Evolution, Journal Year: 2023, Volume and Issue: 13(4)

Published: April 1, 2023

Animal abundance estimation is increasingly based on drone or aerial survey photography. Manual postprocessing has been used extensively; however, volumes of such data are increasing, necessitating some level automation, either for complete counting, as a labour-saving tool. Any automated processing can be challenging when using tools species that nest in close formation Pygoscelis penguins. We present here customized CNN-based density map method counting penguins from low-resolution Our model, an indirect regression algorithm, performed significantly better terms accuracy than standard detection algorithm (Faster-RCNN) small objects images and gave error rate only 0.8 percent. Density methods demonstrated vastly improve our ability to count animals tight aggregations demonstrably monitoring efforts imagery.

Language: Английский

---

Risks of Drone Use in Light of Literature Studies

Sensors, Journal Year: 2024, Volume and Issue: 24(4), P. 1205 - 1205

Published: Feb. 13, 2024

This article aims to present the results of a bibliometric analysis relevant literature and discuss main research streams related topic risks in drone applications. The methodology conducted consisted five procedural steps, including planning research, conducting systematic review literature, proposing classification framework corresponding contemporary trends risk applications, compiling characteristics publications assigned each highlighted thematic groups. used PRISMA method. A total 257 documents comprising articles conference proceedings were analysed. On this basis, eight categories use drones associated with their operation distinguished. Due high content within two these categories, further division into subcategories was proposed illustrate topics better. investigation made it possible identify current pointed out existing gaps, both area assessment its application areas. obtained from can provide interesting material for industry academia.

Language: Английский

---

Going Batty: The Challenges and Opportunities of Using Drones to Monitor the Behaviour and Habitat Use of Rays

Drones, Journal Year: 2021, Volume and Issue: 5(1), P. 12 - 12

Published: Feb. 2, 2021

The way an animal behaves in its habitat provides insight into ecological role. As such, collecting robust, accurate datasets a time-efficient manner is ever-present pressure for the field of behavioural ecology. Faced with shortcomings and physical limitations traditional ground-based data collection techniques, particularly marine studies, drones offer low-cost efficient approach range coastal environments. Despite being widely used to monitor animals, they currently remain underutilised ray research. innovative application environmental studies has presented novel opportunities observation assessment, although this emerging faces substantial challenges. we consider possibility rays using drones, face challenges related local aviation regulations, weather environment, as well sensor platform limitations. Promising solutions continue be developed, however, growing potential drone-based monitoring behaviour use rays. While barriers enter may appear daunting researchers little experience technology becoming increasingly accessible, helping obtain wide highly useful data.

Language: Английский

---

An automated work-flow for pinniped surveys: A new tool for monitoring population dynamics

Frontiers in Ecology and Evolution, Journal Year: 2022, Volume and Issue: 10

Published: Aug. 11, 2022

Detecting changes in population trends depends on the accuracy of estimated mean growth rates and thus quality input data. However, monitoring wildlife populations poses economic logistic challenges especially complex remote habitats. Declines can remain undetected for years unless effective techniques are developed, guiding appropriate management actions. We developed an automated survey workflow using unmanned aerial vehicles (drones) to quantify number size individual animals, well-studied Scandinavian harbour seal ( Phoca vitulina ) as a model species. compared ground-based counts telescopes with manual flights, zoom photo/video, pre-programmed flights producing orthomosaic photo maps. used machine learning identify count both pups older seals we present new method measuring body automatically. evaluate population’s reproductive success drone data, historical predictions from Leslie matrix model. The most accurate time-efficient results were achieved by performing where identified their sizes measured detector was 95–97% classification error 4.6 ± 2.9 3.1 2.1 during good light conditions. There clear distinction between breeding time. 320 season 2021 drone, which is well beyond expected number, based data pup production. high confirms earlier indications deteriorating rate this important colony. show that drones powerful tools inaccessible areas be assess annual recruitment seasonal variations condition.

Language: Английский

---

Collectively advancing deep learning for animal detection in drone imagery: Successes, challenges, and research gaps

Ecological Informatics, Journal Year: 2024, Volume and Issue: unknown, P. 102842 - 102842

Published: Oct. 1, 2024

Language: Английский

---

Coexisting with sharks: a novel, socially acceptable and non-lethal shark mitigation approach

Scientific Reports, Journal Year: 2020, Volume and Issue: 10(1)

Published: Oct. 15, 2020

Abstract Conflict between humans and large predators is a longstanding challenge that can present negative consequences for wildlife. Sharks have global distribution are considered to pose potential threat humans; concurrently many shark species themselves threatened. Developing strategies coexistence this keystone group imperative. We assess blimp surveillance as technique simply effectively reduce encounters at ocean beaches determine the social acceptance of compared an established mitigation strategy—shark meshing. demonstrate suitability blimps risk mitigation, with detection probabilities analogues by professional lifeguards 0.93 in ideal swimming conditions. Social surveys indicate strong preference non-lethal mitigation. show continuous aerial provide measurable reduction from sharks, improving beach safety facilitating people

Language: Английский

---