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

Ecologists often study wildlife populations by deploying camera traps. Large datasets are generated using this approach which can be difficult for research teams to manually evaluate. Researchers increasingly enlist volunteers from the general public as citizen scientists to help classify images. The growing number of camera trap studies, however, makes it ever more challenging to find enough volunteers to process all projects in a timely manner. Advances in machine learning, especially deep learning, allow for accurate automatic image classification. By training models using existing datasets of images classified by citizen scientists and subsequent application of such models on new studies, human effort may be reduced substantially. The goals of this study were to (a) assess the accuracy of deep learning in classifying camera trap data, (b) investigate how to process datasets with only a few classified images that are generally difficult to model, and (c) apply a trained model on a live online citizen science project. Convolutional neural networks (CNNs) were used to differentiate among images of different animal species, images of humans or vehicles, and empty images (no animals, vehicles, or humans). We used four different camera trap datasets featuring a wide variety of species, different habitats, and a varying number of images. All datasets were labelled by citizen scientists on Zooniverse. Accuracies for identifying empty images across projects ranged between 91.2% and 98.0%, whereas accuracies for identifying specific species were between 88.7% and 92.7%. Transferring information from CNNs trained on large datasets (“transfer‐learning”) was increasingly beneficial as the size of the training dataset decreased and raised accuracy by up to 10.3%. Removing low‐confidence predictions increased model accuracies to the level of citizen scientists. By combining a trained model with classifications from citizen scientists, human effort was reduced by 43% while maintaining overall accuracy for a live experiment running on Zooniverse. Ecology researchers can significantly reduce image classification time and manual effort by combining citizen scientists and CNNs, enabling faster processing of data from large camera trap studies.

show abstract

“…The most recent advances by Norouzzadeh et al. () have reported accuracies of 93.8% and have matched human accuracy on over 99% of all images.…”

Section: Introductionmentioning

confidence: 88%

Identifying animal species in camera trap images using deep learning and citizen science

et al. 2018

Self Cite

View full text Add to dashboard Cite

show abstract

“…Image classifications used for species identification have dramatically increased in accuracy, performance, and in the number of taxa analyzed (Marques et al, 2018;Martineau et al, 2017;Norouzzadeh et al, 2018;Schneider, Taylor, & Kremer, 2018;Van Horn et al, 2017).…”

Section: Introductionmentioning

confidence: 99%

Species‐level image classification with convolutional neural network enables insect identification from habitus images

Hansen

Svenning

Olsen

et al. 2019

Ecology and Evolution

View full text Add to dashboard Cite

1. Changes in insect biomass, abundance, and diversity are challenging to track at sufficient spatial, temporal, and taxonomic resolution. Camera traps can capture habitus images of ground-dwelling insects. However, currently sampling involves manually detecting and identifying specimens. Here, we test whether a convolutional neural network (CNN) can classify habitus images of ground beetles to species level, and estimate how correct classification relates to body size, number of species inside genera, and species identity.2. We created an image database of 65,841 museum specimens comprising 361 carabid beetle species from the British Isles and fine-tuned the parameters of a pretrained CNN from a training dataset. By summing up class confidence values within genus, tribe, and subfamily and setting a confidence threshold, we trade-off between classification accuracy, precision, and recall and taxonomic resolution.3. The CNN classified 51.9% of 19,164 test images correctly to species level and 74.9% to genus level. Average classification recall on species level was 50.7%.Applying a threshold of 0.5 increased the average classification recall to 74.6% at the expense of taxonomic resolution. Higher top value from the output layer and larger sized species were more often classified correctly, as were images of species in genera with few species. 4. Fine-tuning enabled us to classify images with a high mean recall for the whole test dataset to species or higher taxonomic levels, however, with high variability.This indicates that some species are more difficult to identify because of properties such as their body size or the number of related species.5. Together, species-level image classification of arthropods from museum collections and ecological monitoring can substantially increase the amount of occurrence data that can feasibly be collected. These tools thus provide new opportunities in understanding and predicting ecological responses to environmental change. Jens-Christian Svenninghttps://orcid.

show abstract

“…Moreover, as insect infestation may not be localized on a building basis but be widespread, by centralizing the server infrastructure and providing geolocational information for each smart trap, an interactive infestation map may be created, adding the system to the smart city concept. Future work relates detection of urban insects and deep learning classification algorithms that have become a standard in visual-based applications of artificial intelligence as in the paradigm of Zhong [29], where the device automatically counts the insects and reports their identity and as in Sadegh et al [39] where the identity of wild animals is automatically inferred based on their picture.…”

Section: Resultsmentioning

confidence: 99%

A “Smart” Trap Device for Detection of Crawling Insects and Other Arthropods in Urban Environments

et al. 2018

View full text Add to dashboard Cite

Abstract:We introduce a device for the automatic detecting and reporting of crawling insects in urban environments. It is a monitoring device for urban pests that complies with the context of smart homes and smart cities, and is compatible with the emerging discipline of the Internet of Things (IoT). We believe it can find its place in every room of a hotel, hospital, military camp, and residence. This box-shaped device attracts targeted insect pests, senses the entering insect, and takes automatically a picture of the internal space of the box. The e-trap includes strong attractants (pheromone and/or food) to increase capture efficiency and traps the insect on its sticky floor. The device carries the necessary optoelectronic sensors to monitor all entrances of the trap. As the insect enters it interrupts the infrared light source. This triggers a detection event; a picture is taken, and a time-stamp is set before delivering the picture through the Wi-Fi to an authorized person/stakeholder. The device can be integrated seamlessly in urban environments and operates unobtrusively to human activities. We report results on various insect pests and depending on the insect species, can reach a detection accuracy ranging from 96 to 99%.

show abstract

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

Cited by 722 publications

References 59 publications

Identifying animal species in camera trap images using deep learning and citizen science

Identifying animal species in camera trap images using deep learning and citizen science

Species‐level image classification with convolutional neural network enables insect identification from habitus images

A “Smart” Trap Device for Detection of Crawling Insects and Other Arthropods in Urban Environments

Contact Info

Product

Resources

About