• Corpus ID: 6055571

Wildbook: Crowdsourcing, computer vision, and data science for conservation

@article{BergerWolf2017WildbookCC,
  title={Wildbook: Crowdsourcing, computer vision, and data science for conservation},
  author={T. Berger-Wolf and Daniel I. Rubenstein and Charles V. Stewart and Jason A. Holmberg and Jason Remington Parham and S. Menon and Jonathan P. Crall and Jon Van Oast and Emre Kıcıman and Lucas N Joppa},
  journal={ArXiv},
  year={2017},
  volume={abs/1710.08880}
}
Photographs, taken by field scientists, tourists, automated cameras, and incidental photographers, are the most abundant source of data on wildlife today. Wildbook is an autonomous computational system that starts from massive collections of images and, by detecting various species of animals and identifying individuals, combined with sophisticated data management, turns them into high resolution information database, enabling scientific inquiry, conservation, and citizen science. We have… 

Figures and Tables from this paper

Animal Wildlife Population Estimation Using Social Media Images Collections
TLDR
A new framework is presented to take into account the social media bias when using this data source to provide wildlife population size estimates and it is shown that, surprisingly, this is a learnable and potentially solvable problem.
The Role of Citizen Science and Deep Learning in Camera Trapping
TLDR
A possible pathway for isolated observations to be aggregated into a single database that respects the existing standards is suggested, incorporating a combination of the roles of experts’ and citizens’ evaluations, the way of training a neural network and adding a taxon complexity index.
Avances en foto-identificación automatizada de fauna silvestre
The wild species photo-identification is a basic resource for obtaining the necessary information for several biological research tasks. Today crowdsourcing and citizen science are beginning to play
Identifying Individual Jaguars and Ocelots via Pattern‐Recognition Software: Comparing HotSpotter and Wild‐ID
Camera‐trapping is widespread in wildlife studies, especially for species with individually unique markings to which capture–recapture analytical techniques can be applied. The large volume of data
HotSpotter: Using a computer-driven photo-id application to identify sea turtles
Mosquitoes and Public Health: Improving Data Validation of Citizen Science Contributions Using Computer Vision
TLDR
An image recognition prototype that includes image collection, training of image classifiers, specimen recognition, and expert validation and analytics is developed that can be used to assist in the mitigation of disease outbreaks.
Social media mining and photo-identification detects the shift of long-term seasonal foraging habitat for a juvenile loggerhead sea turtle
Animal imagery uploaded on social media has been identified as an important tool in wildlife research and conservation and has been used in a variety of recent studies. Here a case study is
ElephantBook: A Semi-Automated Human-in-the-Loop System for Elephant Re-Identification
TLDR
This work has built and deployed a web-based platform and database for human-in- the-loop re-identification of elephants combining manual attribute labeling and state-of-the-art computer technology.
Perspectives in machine learning for wildlife conservation
TLDR
This paper aims to demonstrate the efforts towards in-situ applicability of EMMARM, which aims to provide real-time information about the behaviour of birds to aid in the prediction of ellipsoidal behaviour.
...
1
2
3
...

References

SHOWING 1-10 OF 65 REFERENCES
Animal Population Censusing at Scale with Citizen Science and Photographic Identification
TLDR
This paper analyzes the results of two population censusing events, the Great Zebra and Giraffe Count and the Great Grevy’s Rally, where combined the authors processed over 50,000 photographs taken with more than 200 different cameras and over 300 on-the-ground volunteers.
Animal Population Estimation Using Flickr Images
While the technologies of the Information Age have produced staggering amounts of data about people, they are by and large failing the world’s wildlife. Even the simplest and most critical piece of
Snapshot Serengeti, high-frequency annotated camera trap images of 40 mammalian species in an African savanna
TLDR
This work deployed 225 camera traps across Serengeti National Park, Tanzania, to evaluate spatial and temporal inter-species dynamics and classified the images via the citizen-science website www.snapshotsereNGeti.org, yielding a final classification for each image and a measure of agreement among individual answers.
Animal Wildlife Population Estimation Using Social Media Images Collections
TLDR
A new framework is presented to take into account the social media bias when using this data source to provide wildlife population size estimates and it is shown that, surprisingly, this is a learnable and potentially solvable problem.
Unmanned Aerial Vehicles (UAVs) for Surveying Marine Fauna: A Dugong Case Study
TLDR
The results suggest that UAV systems may not be limited by sea state conditions in the same manner as sightings from manned surveys, and the overlap between images proved valuable for detecting animals that were masked by sun glitter in the corners of images, and identifying animals initially captured at awkward body angles.
A critique of density estimation from camera-trap data†
TLDR
This work critically review 47 published studies and discusses the problems associated with contemporary population estimates of elusive species from camera-trap data, and recommends greater transparency in study design and quality of the data and the recently developed spatially explicit capture–recapture models as an alternative approach.
Citizen Science as an Ecological Research Tool: Challenges and Benefits
Citizen science, the involvement of volunteers in research, has increased the scale of ecological field studies with continent-wide, centralized monitoring efforts and, more rarely, tapping of
Opportunistic citizen science data of animal species produce reliable estimates of distribution trends if analysed with occupancy models
TLDR
It is investigated whether occupancy models can correct for the observation, reporting and detection biases in opportunistic data, enabling species trends to be monitored for species groups and regions where it is not feasible to collect standardized data on a large scale.
Calibrating a camera trap–based biased mark–recapture sampling design to survey the leopard population in the N'wanetsi concession, Kruger National Park, South Africa
Estimating large carnivore abundance can be challenging. A biased leopard (Panthera pardus) population survey was conducted in the N’wanetsi concession in the Kruger National Park (KNP), South
Statistical solutions for error and bias in global citizen science datasets
...
1
2
3
4
5
...