Deep learning for environmental conservation

The last decade has transformed the field of artificial intelligence, with deep learning at the forefront of this development. With its ability to 'self-learn' discriminative patterns directly from data, deep learning is a promising computational approach for automating the classification of visual, spatial and acoustic information in the context of environmental conservation. Here, we first highlight the current and future applications of supervised deep learning in environmental conservation. Next, we describe a number of technical and implementation-related challenges that can potentially impede the real-world adoption of this technology in conservation programmes. Lastly, to mitigate these pitfalls, we discuss priorities for guiding future research and hope that these recommendations will help make this technology more accessible to environmental scientists and conservation practitioners.

[1]  Z. Buřivalová,et al.  The sound of a tropical forest , 2019, Science.

[2]  Kate E. Jones,et al.  CityNet—Deep learning tools for urban ecoacoustic assessment , 2018, Methods in Ecology and Evolution.

[3]  Gary Marcus,et al.  Deep Learning: A Critical Appraisal , 2018, ArXiv.

[4]  Francisco Herrera,et al.  Deep-Learning Convolutional Neural Networks for scattered shrub detection with Google Earth Imagery , 2017, ArXiv.

[5]  Oliver R. Wearn,et al.  Responsible AI for conservation , 2019, Nature Machine Intelligence.

[6]  C. Jayawardena,et al.  TreeSpirit: Illegal logging detection and alerting system using audio identification over an IoT network , 2017, 2017 11th International Conference on Software, Knowledge, Information Management and Applications (SKIMA).

[7]  Ella Browning,et al.  Predicting animal behaviour using deep learning: GPS data alone accurately predict diving in seabirds , 2017 .

[8]  Michael A. Tabak,et al.  Machine learning to classify animal species in camera trap images: applications in ecology , 2018, bioRxiv.

[9]  Alex Rogers,et al.  AudioMoth: Evaluation of a smart open acoustic device for monitoring biodiversity and the environment , 2018 .

[10]  Anurag Agarwal,et al.  The Internet of Things—A survey of topics and trends , 2014, Information Systems Frontiers.

[11]  K. S. Willis,et al.  Remote sensing change detection for ecological monitoring in United States protected areas , 2015 .

[12]  Yuan Yu,et al.  TensorFlow: A system for large-scale machine learning , 2016, OSDI.

[13]  Hoo-Chang Hoo-Chang Shin Shin,et al.  Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning , 2016, Ieee Transactions on Medical Imaging.

[14]  R. Kays,et al.  Emerging Technologies to Conserve Biodiversity. , 2015, Trends in ecology & evolution.

[15]  S. Wich,et al.  Dawn of Drone Ecology: Low-Cost Autonomous Aerial Vehicles for Conservation , 2012 .

[16]  Peng Hao,et al.  Transfer learning using computational intelligence: A survey , 2015, Knowl. Based Syst..

[17]  Siu-Ming Yiu,et al.  Multi-key privacy-preserving deep learning in cloud computing , 2017, Future Gener. Comput. Syst..

[18]  Christoph Fink,et al.  Machine learning for tracking illegal wildlife trade on social media , 2018, Nature Ecology & Evolution.

[19]  Ramakant Nevatia,et al.  SPOT Poachers in Action: Augmenting Conservation Drones With Automatic Detection in Near Real Time , 2018, AAAI.

[20]  Ronald M. Summers,et al.  Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning , 2016, IEEE Transactions on Medical Imaging.

[21]  Bradley J. Erickson,et al.  Toolkits and Libraries for Deep Learning , 2017, Journal of Digital Imaging.

[22]  William J. Dally,et al.  The GPU Computing Era , 2010, IEEE Micro.

[23]  Pieter Abbeel,et al.  Image Object Label 3 D CAD Model Candidate Grasps Google Object Recognition Engine Google Cloud Storage Select Feasible Grasp with Highest Success Probability Pose EstimationCamera Robots Cloud 3 D Sensor , 2014 .

[24]  Ben. G. Weinstein.  A computer vision for animal ecology. , 2018, The Journal of animal ecology.

[25]  Margaret Kosmala,et al.  Automatically identifying, counting, and describing wild animals in camera-trap images with deep learning , 2017, Proceedings of the National Academy of Sciences.

[26]  Christoph Fink,et al.  A framework for investigating illegal wildlife trade on social media with machine learning , 2018, Conservation biology : the journal of the Society for Conservation Biology.

[27]  Mark A. Girolami,et al.  Bat detective—Deep learning tools for bat acoustic signal detection , 2017, bioRxiv.

[28]  Amit Agarwal,et al.  CNTK: Microsoft's Open-Source Deep-Learning Toolkit , 2016, KDD.

[29]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[30]  John Joseph Valletta,et al.  Applications of machine learning in animal behaviour studies , 2017, Animal Behaviour.

[31]  Chandra Krintz,et al.  Where's the Bear? - Automating Wildlife Image Processing Using IoT and Edge Cloud Systems , 2017, 2017 IEEE/ACM Second International Conference on Internet-of-Things Design and Implementation (IoTDI).

[32]  Bram van Ginneken,et al.  A survey on deep learning in medical image analysis , 2017, Medical Image Anal..

[33]  Lian Pin Koh,et al.  Futurecasting ecological research: The rise of technoecology , 2018 .

[34]  Peter Corcoran,et al.  Deep Learning for Consumer Devices and Services: Pushing the limits for machine learning, artificial intelligence, and computer vision. , 2017, IEEE Consumer Electronics Magazine.

[35]  Amanda J. C. Sharkey,et al.  Combining Artificial Neural Nets: Ensemble and Modular Multi-Net Systems , 1999 .

[36]  Julio Hernandez-Castro,et al.  Assessing the extent and nature of wildlife trade on the dark web , 2016, Conservation biology : the journal of the Society for Conservation Biology.

[37]  L. Joppa.  The case for technology investments in the environment. , 2017 .