Filters








4,274 Hits in 8.4 sec

A deep active learning system for species identification and counting in camera trap images [article]

Mohammad Sadegh Norouzzadeh, Dan Morris, Sara Beery, Neel Joshi, Nebojsa Jojic, Jeff Clune
2019 arXiv   pre-print
We combine the power of machine intelligence and human intelligence to build a scalable, fast, and accurate active learning system to minimize the manual work required to identify and count animals in  ...  Motion-activated cameras, also known as camera traps, are a critical tool for population surveys, as they are cheap and non-intrusive.  ...  For example, species identification in camera trap images is an image classification problem in which the input is the camera trap image and the output is the probability of the presence of each species  ... 
arXiv:1910.09716v1 fatcat:xqrxsaf5obbqxpksictfildjwa

Automatically identifying, counting, and describing wild animals in camera-trap images with deep learning

Mohammad Sadegh Norouzzadeh, Anh Nguyen, Margaret Kosmala, Alexandra Swanson, Meredith S. Palmer, Craig Packer, Jeff Clune
2018 Proceedings of the National Academy of Sciences of the United States of America  
We train deep convolutional neural networks to identify, count, and describe the behaviors of 48 species in the 3.2 millionimage Snapshot Serengeti dataset.  ...  Those efficiency gains highlight the importance of using deep neural networks to automate data extraction from camera-trap images, reducing a roadblock for this widely used technology.  ...  We thank Sarah Benson-Amram, the SS volunteers, and the members of the Evolving AI Laboratory at the University of Wyoming for valuable feedback, especially Joost Huizinga, Tyler Jaszkowiak, Roby Velez  ... 
doi:10.1073/pnas.1719367115 pmid:29871948 pmcid:PMC6016780 fatcat:uaflbasnznbzfescvn2vw4bnl4

Automatically identifying, counting, and describing wild animals in camera-trap images with deep learning [article]

Mohammed Sadegh Norouzzadeh, Anh Nguyen, Margaret Kosmala, Ali Swanson, Meredith Palmer, Craig Packer, Jeff Clune
2017 arXiv   pre-print
We train deep convolutional neural networks to identify, count, and describe the behaviors of 48 species in the 3.2-million-image Snapshot Serengeti dataset.  ...  More importantly, if our system classifies only images it is confident about, our system can automate animal identification for 99.3% of the data while still performing at the same 96.6% accuracy as that  ...  Deep neural networks can successfully identify, count, and describe animals in camera-trap images.  ... 
arXiv:1703.05830v5 fatcat:5ip5ine4czhshieovcue2ellli

Automatic Detection and Monitoring of Insect Pests—A Review

Matheus Cardim Ferreira Lima, Maria Elisa Damascena de Almeida Leandro, Constantino Valero, Luis Carlos Pereira Coronel, Clara Oliva Gonçalves Bazzo
2020 Agriculture  
The paper focuses on the methods for identification of pests based in infrared sensors, audio sensors and image-based classification, presenting the different systems available, examples of applications  ...  Future trends of automatic traps and decision support systems are also discussed.  ...  [36] developed a model for the counting and identification of aphids based on machine learning and an adapted smartphone.  ... 
doi:10.3390/agriculture10050161 fatcat:jilapwwc5ncexnoj3u3nblcd3q

Deep Learning Object Detection Methods for Ecological Camera Trap Data [article]

Stefan Schneider, Graham W. Taylor, Stefan C. Kremer
2018 arXiv   pre-print
Recent advances in the field of deep learning for object detection show promise towards automating the analysis of camera trap images.  ...  Deep learning methods for computer vision tasks show promise for automating the data analysis of camera trap images.  ...  ACKNOWLEDGMENTS The authors would like to thank all the Snapshot Serengeti volunteers and the camera trap community at large for uploading their data for public access.  ... 
arXiv:1803.10842v1 fatcat:5ptcazjymjgfdbc3qairowsyjy

Deep learning and computer vision will transform entomology

Toke T. Høye, Johanna Ärje, Kim Bjerge, Oskar L. P. Hansen, Alexandros Iosifidis, Florian Leese, Hjalte M. R. Mann, Kristian Meissner, Claus Melvad, Jenni Raitoharju
2021 Proceedings of the National Academy of Sciences of the United States of America  
Here, we connect recent developments in deep learning and computer vision to the urgent demand for more cost-efficient monitoring of insects and other invertebrates.  ...  Advances in computer vision and deep learning provide potential new solutions to this global challenge.  ...  Other solutions have embedded a digital camera and a microprocessor that can count trapped individuals in real time using object detection based on a deep learning model (37) .  ... 
doi:10.1073/pnas.2002545117 pmid:33431561 fatcat:3m4xtz5365awlpv6nlmunkuwl4

Three critical factors affecting automated image species recognition performance for camera traps

Stefan Schneider, Saul Greenberg, Graham W. Taylor, Stefan C. Kremer
2020 Ecology and Evolution  
The success of deep learning systems using camera trap images has been previously explored in preliminary stages. These studies, however, are lacking in their practicality.  ...  Our goal was to test the capabilities of deep learning systems trained on camera trap images using modestly sized training data, compare performance when considering unseen background locations, and quantify  ...  their own-deep learning systems for camera trap images (Schneider github animal classification tool)  ... 
doi:10.1002/ece3.6147 pmid:32274005 pmcid:PMC7141055 fatcat:m3mabgop2zf2dp7dsogprm3rui

Image-Based Animal Detection and Breed Identification Using Neural Networks

2020 Journal of science and technolgy  
to the world of ecology and trap camera images analysis .  ...  of filtering animal images and identifying species automatically and counting the number of species.  ...  We also thank all friends for being a constant source of our support.  ... 
doi:10.46243/jst.2020.v5.i5.pp130-134 fatcat:rf2tlgi6y5edfbqrwq6q4heo34

Deep learning and computer vision will transform entomology [article]

Toke Thomas Hoye, Johanna Arje, Kim Bjerge, Oskar LP Hansen, Alexandros Iosifidis, Florian Leese, Hjalte Mann, Kristian Meissner, Claus Melvad, Jenni Raitoharju
2020 bioRxiv   pre-print
Here, we connect recent developments in deep learning and computer vision to the urgent demand for more cost-efficient monitoring of insects and other invertebrates.  ...  Advances in computer vision and deep learning provide potential new solutions to this global challenge.  ...  and identification of species recorded with cameras in the field pose a 332 critical challenge for implementing deep learning tools in entomology.  ... 
doi:10.1101/2020.07.03.187252 fatcat:wv3sn4jet5dr3llion5ua34ssu

A light trap and computer vision system to detect and classify live moths (Lepidoptera) using tracking and deep learning [article]

Kim Bjerge, Martin Videbaek Sepstrup, Jakob Bonde Nielsen, Flemming Helsing, Toke Thomas Hoye
2020 bioRxiv   pre-print
A computer vision algorithm referred to as Moth Classification and Counting, based on deep learning analysis of the captured images then tracked and counted the number of insects and identified moth species  ...  A light trap with multiple illuminations and a camera was designed to attract and monitor live insects during twilight and night hours.  ...  Developing a deep learning model for classifying species was an iterative process with an alternation between selecting images and training CNN models.  ... 
doi:10.1101/2020.03.18.996447 fatcat:lsxwsgqk6rh55jlft4kvu2obna

An Automated Light Trap to Monitor Moths (Lepidoptera) Using Computer Vision-Based Tracking and Deep Learning

Kim Bjerge, Jakob Bonde Nielsen, Martin Videbæk Sepstrup, Flemming Helsing-Nielsen, Toke Thomas Høye
2021 Sensors  
A computer vision algorithm referred to as Moth Classification and Counting (MCC), based on deep learning analysis of the captured images, tracked and counted the number of insects and identified moth  ...  Insect monitoring methods are typically very time-consuming and involve substantial investment in species identification following manual trapping in the field.  ...  Acknowledgments: The authors like to acknowledge Freia Martensen for language and proof reading the article and Mads Dyrmann for inputs and review.  ... 
doi:10.3390/s21020343 pmid:33419136 fatcat:g4nxswllofc6vhpmd4l2qrd5hy

Perspectives on individual animal identification from biology and computer vision [article]

Maxime Vidal and Nathan Wolf and Beth Rosenberg and Bradley P. Harris and Alexander Mathis
2021 arXiv   pre-print
We conclude by offering recommendations for starting an animal identification project, illustrate current limitations and propose how they might be addressed in the future.  ...  In response to some of the limitations of current identification methods, new automated computer vision approaches have emerged with strong performance.  ...  Support for MV, BR, NW, and BPH was provided by Alaska Education Tax  ... 
arXiv:2103.00560v1 fatcat:6xdsiojn7vamxonhwmklse3tja

Perspectives on individual animal identification from biology and computer vision

Maxime Vidal, Nathan Wolf, Beth Rosenberg, Bradley P Harris, Alexander Mathis
2021 Integrative and Comparative Biology  
We conclude by offering recommendations for starting an animal identification project, illustrate current limitations and propose how they might be addressed in the future.  ...  In response to some of the limitations of current identification methods, new automated computer vision approaches have emerged with strong performance.  ...  Images from camera traps can be used both for model training and monitored for inference.  ... 
doi:10.1093/icb/icab107 pmid:34050741 pmcid:PMC8490693 fatcat:zp32mrr56fcvda4ubc3r2xotja

Seeing biodiversity: perspectives in machine learning for wildlife conservation [article]

Devis Tuia, Benjamin Kellenberger, Sara Beery, Blair R. Costelloe, Silvia Zuffi, Benjamin Risse, Alexander Mathis, Mackenzie W. Mathis, Frank van Langevelde, Tilo Burghardt, Roland Kays, Holger Klinck (+6 others)
2021 arXiv   pre-print
We argue that machine learning, and especially deep learning approaches, can meet this analytic challenge to enhance our understanding, monitoring capacity, and conservation of wildlife species.  ...  approaches and train a new generation of data scientists in ecology and conservation.  ...  SB would like to thank the Microsoft AI for Earth initiative, the Idaho Department of Fish and Game, and Wildlife Protection Solutions for insightful discussions and providing data for figures.  ... 
arXiv:2110.12951v1 fatcat:hzuz5aeja5dfxpte6xprg3yjti

Iterative Human and Automated Identification of Wildlife Images [article]

Zhongqi Miao, Ziwei Liu, Kaitlyn M. Gaynor, Meredith S. Palmer, Stella X. Yu, Wayne M. Getz
2021 arXiv   pre-print
These two drawbacks can be overcome through a hybrid combination of machine learning and humans in the loop.  ...  Our proposed iterative human and automated identification approach is capable of learning from wildlife imagery data with a long-tailed distribution.  ...  Automatic image recognition systems The use of deep learning (a subset of AI technology) to automatically identify animals in camera trap images has recently drawn considerable attention from the ecological  ... 
arXiv:2105.02320v2 fatcat:ja3wbtfkpbedrinyfxlgnsmuui
« Previous Showing results 1 — 15 out of 4,274 results