Automatic whale counting in satellite images with deep learning

Guirado, Emilio; Tabik, Siham; Ml, Rivas; Alcaraz‐Segura, Domingo; Herrera, Francisco

doi:10.1101/443671

Cited by 18 publications

(16 citation statements)

References 56 publications

(30 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As deep learning is used to detect, identify and classify individuals in automatic monitoring data, such tools can be scaled up to help monitor populations. For instance population size can be estimated by counting individuals (Guirado, Tabik, Rivas, Alcaraz‐Segura, & Herrera, ; Norouzzadeh et al, ). By extension, information such as population distribution or density can also be calculated from this data as it has already been done with traditional methods (Rovero et al, ).…”

Section: Overview Of Applications In Ecologymentioning

confidence: 99%

“…Therefore, training datasets often contain thousands to millions of examples, depending on the task, the number of items to detect and the desired performance (Marcus, ). Good results have been obtained with smaller training datasets with only several hundred examples per class (Abrams et al, ; Fairbrass et al, ; Guirado et al, ) opening the approach for most fields of ecology. Yet, overall, the bigger the training dataset, the better the classification accuracy will be (Marcus, ).…”

Section: Implementing Deep Learning: Challenges and Guidelinesmentioning

confidence: 99%

See 1 more Smart Citation

Applications for deep learning in ecology

2019

View full text Add to dashboard Cite

A lot of hype has recently been generated around deep learning, a novel group of artificial intelligence approaches able to break accuracy records in pattern recognition. Over the course of just a few years, deep learning has revolutionized several research fields such as bioinformatics and medicine with its flexibility and ability to process large and complex datasets. As ecological datasets are becoming larger and more complex, we believe these methods can be useful to ecologists as well. In this paper, we review existing implementations and show that deep learning has been used successfully to identify species, classify animal behaviour and estimate biodiversity in large datasets like camera‐trap images, audio recordings and videos. We demonstrate that deep learning can be beneficial to most ecological disciplines, including applied contexts, such as management and conservation. We also identify common questions about how and when to use deep learning, such as what are the steps required to create a deep learning network, which tools are available to help, and what are the requirements in terms of data and computer power. We provide guidelines, recommendations and useful resources, including a reference flowchart to help ecologists get started with deep learning. We argue that at a time when automatic monitoring of populations and ecosystems generates a vast amount of data that cannot be effectively processed by humans anymore, deep learning could become a powerful reference tool for ecologists.

show abstract

Section: Overview Of Applications In Ecologymentioning

confidence: 99%

Section: Implementing Deep Learning: Challenges and Guidelinesmentioning

confidence: 99%

Applications for deep learning in ecology

2019

View full text Add to dashboard Cite

show abstract

“…In addition, several studies are being conducted to investigate the use of deep-learning applications for wildlife monitoring with the aid of aerial photographs. Maire et al [51] showed the feasibility of using the simple linear iterative clustering (SLIC) and CNN methods for the detection of wild marine mammals, and Guirado et al [52] detected whales using satellite images in combination with a CNN-based method to detect the presence of whales and for whale counting.…”

Section: Introductionmentioning

confidence: 99%

Application of Deep-Learning Methods to Bird Detection Using Unmanned Aerial Vehicle Imagery

Hong

Han

Kim

et al. 2019

Sensors

155

114

View full text Add to dashboard Cite

Wild birds are monitored with the important objectives of identifying their habitats and estimating the size of their populations. Especially in the case of migratory bird, they are significantly recorded during specific periods of time to forecast any possible spread of animal disease such as avian influenza. This study led to the construction of deep-learning-based object-detection models with the aid of aerial photographs collected by an unmanned aerial vehicle (UAV). The dataset containing the aerial photographs includes diverse images of birds in various bird habitats and in the vicinity of lakes and on farmland. In addition, aerial images of bird decoys are captured to achieve various bird patterns and more accurate bird information. Bird detection models such as Faster Region-based Convolutional Neural Network (R-CNN), Region-based Fully Convolutional Network (R-FCN), Single Shot MultiBox Detector (SSD), Retinanet, and You Only Look Once (YOLO) were created and the performance of all models was estimated by comparing their computing speed and average precision. The test results show Faster R-CNN to be the most accurate and YOLO to be the fastest among the models. The combined results demonstrate that the use of deep-learning-based detection methods in combination with UAV aerial imagery is fairly suitable for bird detection in various environments.

show abstract

“…However, one of the most timeconsuming aspects of biological observation has been identifying species, historically requiring taxonomic experts. Deep learning techniques enable automated classification of species from a variety of platforms, including: opportunistic citizen science visual observations (e.g., redmap.org.au; iNaturalist.org; Pimm et al, 2015); benthic photo quadrats (BisQue; Rahimi et al, 2014, Fedorov et al, 2017; cabled video observatories (Marini et al, 2018); unmanned underwater vehicles (Qin et al, 2015;Sung et al, 2017); acoustic-sensing hydrophones (Dugan et al, 2015;McQuay et al, 2017); plankton-sensing flow cytometers (Göröcs et al, 2018); and satellite imagery (Guirado et al, 2018). Taxonomic experts are still very much needed for developing datasets as inputs to this modeling approach.…”

Section: Artificial Intelligencementioning

confidence: 99%

Global Observational Needs and Resources for Marine Biodiversity

et al. 2019

View full text Add to dashboard Cite

The diversity of life in the sea is critical to the health of ocean ecosystems that support living resources and therefore essential to the economic, nutritional, recreational, and health needs of billions of people. Yet there is evidence that the biodiversity of many marine habitats is being altered in response to a changing climate and human activity. Understanding this change, and forecasting where changes are likely to occur, requires monitoring of organism diversity, distribution, abundance, and health. It requires a minimum of measurements including productivity and ecosystem function, species composition, allelic diversity, and genetic expression. These observations need to be complemented with metrics of environmental change and socioeconomic drivers. However, existing global ocean observing infrastructure and programs often do not explicitly consider observations of marine biodiversity and associated processes. Much effort has focused on physical, chemical and some biogeochemical measurements.

show abstract

Automatic whale counting in satellite images with deep learning

Cited by 18 publications

References 56 publications

Applications for deep learning in ecology

Applications for deep learning in ecology

Application of Deep-Learning Methods to Bird Detection Using Unmanned Aerial Vehicle Imagery

Global Observational Needs and Resources for Marine Biodiversity

Contact Info

Product

Resources

About