EventFinder: a program for screening remotely captured images

Janzen, Michael; Ritter, Ashley Z.; Walker, Philip D.; Visscher, Darcy R.

doi:10.1007/s10661-019-7518-9

Cited by 6 publications

(3 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Increasingly, artificial intelligence (AI) solutions are becoming available for identifying species in imagery and filtering out empty images (Beery et al 2019, Falzon et al 2019, Janzen et al 2019). Theoretically, in order for an AI system to identify animals to species, it requires the AI to be trained for each site, requiring a foundational set of imagery already identified to species.…”

Section: Discussionmentioning

confidence: 99%

Estimating species richness with camera traps: modeling the effects of delay period, deployment length, number of sites, and interference imagery

Mashintonio

Harris

Stewart

et al. 2022

Wildlife Society Bulletin

View full text Add to dashboard Cite

Biologists commonly use camera traps for estimating species richness to inform conservation actions, steer land protection, and reveal effects of climate change. Long-term studies using short delay periods (≤1 min) and numerous cameras produce voluminous amounts of redundant imagery. Thus, cameratrapping procedures maximizing richness estimates while minimizing data collection need development. We used imagery of mammals spanning 4 deserts in the United States to model the effects of delay, deployment length (i.e., study duration), number of sampling sites, and interference events on the proportion of known species richness detected (R p ). We also determined the proportion of subsamples containing each species (SR) under different sampling conditions to inform subsequent occupancy estimation. We generated contour plots describing the optimal configuration of sites and deployment length that minimized the image acquisition required to estimate R p = 0.9. The optimal configuration was independent of delay (requiring ~50 sites and 13 months). The shortest delay (10 sec) generated ~8 times more images than the longest (3600 sec) without substantially improving R p and rare species detection. The shortest duration to acquire R p = 0.9 was 10 months but required ~70 sites. The fewest

show abstract

Section: Discussionmentioning

confidence: 99%

Estimating species richness with camera traps: modeling the effects of delay period, deployment length, number of sites, and interference imagery

Mashintonio

Harris

Stewart

et al. 2022

Wildlife Society Bulletin

View full text Add to dashboard Cite

show abstract

“…With the rapid development of machine vision in various fields [5][6][7], it has also been used in wildlife diversity monitoring to provide important data support for the research, conservation, and management of wildlife [8][9][10][11]. The images captured in various wildlife reserves are accumulating in millions of units, and these images contain a large amount of key information about wild species and populations [12], individual activities [13], illness and injury [14], etc. Currently, these data mainly rely on artificial visual screening, which is far behind the speed of image accumulation and seriously restricts the effective application of data in research, protection, and management work.…”

Section: Related Workmentioning

confidence: 99%

MammalClub: An Annotated Wild Mammal Dataset for Species Recognition, Individual Identification, and Behavior Recognition

Lu,

Zhao,

Wang

et al. 2023

Electronics

View full text Add to dashboard Cite

Mammals play an important role in conserving species diversity and maintaining ecological balance, so research on mammal species composition, individual identification, and behavioral analysis is of great significance for optimizing the ecological environment. Due to their great capabilities for feature extraction, deep learning networks have gradually been applied to wildlife monitoring. However, training a network requires a large number of animal image samples. Although a few wildlife datasets contain many mammals, most mammal images in these datasets are not annotated. In particular, selecting mammalian images from vast and comprehensive datasets is still a time-consuming task. Therefore, there is currently a lack of specialized datasets of images of wild mammals. To address these limitations, this article created a mammal image dataset (named MammalClub), which contains three sub-datasets (i.e., a species recognition sub-dataset, an individual identification sub-dataset, and a behavior recognition sub-dataset). This study labeled the bounding boxes of the images used for species recognition and the coordinates of the mammals’ skeletal joints for behavior recognition. This study also captured images of each individual from different points of view for individual mammal identification. This study explored novel intelligent animal recognition models and compared and analyzed them with the mainstream models in order to test the dataset.

show abstract

“…The EventFinder suite was used to facilitate the removal of non-target (i.e. vegetation and empty frames) images and then used to collapse individual images into independent events for classification (for full details see Janzen et al, 2019). Photo metadata including camera name, location, date, and temperature were recorded and the events were tagged with species name, age class, sex, and number of individuals.…”

Section: Camera Trappingmentioning

confidence: 99%

Host spatiotemporal overlap in a park with high endemicity of Echinococcus multilocularis

Visscher¹,

Toews²,

Pattison³

et al. 2023

Front. Parasitol.

Self Cite

View full text Add to dashboard Cite

BackgroundThere has been a spate of recent cases of human alveolar echinococcosis (AE) in Alberta, Canada. Alveolar echinococcosis is caused by Echinococcus multilocularis, which is prevalent among coyote populations and present in domestic dogs in Alberta.Methods and resultsUsing qPCR, we estimated the seasonal fecal prevalence of E. multilocularis in coyotes and dogs in a multiuse recreation area close to Edmonton, Alberta, where we also setup remote cameras to model seasonal changes in the overlap in temporal activity and the spatial intensity of use among coyotes, humans, and dogs, as a proxy of potential transmission. We detected E. multilocularis in 18 of 137 wild canid feces and none in 44 dog feces. After correcting for the qPCR test’s sensitivity and specificity, we estimated at 15.7% (9.7-22.7%, 95% CrI) the true fecal prevalence for coyotes. Temporal overlap between coyotes and both humans and dogs increased in the fall and winter relative to the spring and summer. Coyote intensity of use showed seasonal variations and was higher on maintained trails and locations closer to visitor parking and at sites with high intensity of dog use.ConclusionsOur results reinforce the need of an integrated approach, typical of both One-Health and Eco-Health, to park management for minimizing the likelihood of transmission where human and dog activity results in significant overlap with the one of the natural definitive hosts of zoonotic parasites.

show abstract

EventFinder: a program for screening remotely captured images

Cited by 6 publications

References 17 publications

Estimating species richness with camera traps: modeling the effects of delay period, deployment length, number of sites, and interference imagery

Estimating species richness with camera traps: modeling the effects of delay period, deployment length, number of sites, and interference imagery

MammalClub: An Annotated Wild Mammal Dataset for Species Recognition, Individual Identification, and Behavior Recognition

Host spatiotemporal overlap in a park with high endemicity of Echinococcus multilocularis

Contact Info

Product

Resources

About