Animal Identification in Low Quality Camera-Trap Images Using Very Deep Convolutional Neural Networks and Confidence Thresholds

Gomez, Alexander; Díez, German; Salazar, Augusto; Díaz, Angélica M.

doi:10.1007/978-3-319-50835-1_67

Cited by 30 publications

(20 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Blanc, Lingrand, & Precioso, ; Lytle et al., ), the accuracy of these approaches was generally low (>70%). However, recent advances using new deep learning models have greatly improved model performance across a wide variety of animal taxa, from coral (Beijboom et al., ) to large mammals (Gomez, Diez, Salazar, & Diaz, ) (Table ). The majority of applications I reviewed had a particular geographic focus, for example the rodent community of the Mojave desert (Wilber et al., ).…”

Section: Identitymentioning

confidence: 99%

“…Computer vision applications, and especially deep learning approaches, require significant training and testing data. High‐quality datasets are difficult to find, and a lack of labelled data is a major obstacle in computer vision research (Belongie & Perona, ; Berg et al., ; Gomez et al., ). Packaging image datasets and making them publicly available will raise awareness of the opportunities for ecological collaboration.…”

Section: Collaboration With Computer Vision Researchersmentioning

confidence: 99%

“…The future of ecological computer vision will combine new algorithms, data and collaborations to study animals in natural environments. The rise of neural networks as the central tool in image classification (Gomez et al., ; LeCun et al., ), background subtraction (Christiansen et al., ) and image description (Mohanty, Hughes, & Salathé, ) is a key development that with bring new opportunities for ecological computer vision (Figure ). Until recently, the growth of these tools has been slowed by a lack of access to cutting edge algorithms.…”

Section: Future Growthmentioning

confidence: 99%

See 2 more Smart Citations

A computer vision for animal ecology

Weinstein

2017

Journal of Animal Ecology

338

287

View full text Add to dashboard Cite

A central goal of animal ecology is to observe species in the natural world. The cost and challenge of data collection often limit the breadth and scope of ecological study. Ecologists often use image capture to bolster data collection in time and space. However, the ability to process these images remains a bottleneck. Computer vision can greatly increase the efficiency, repeatability and accuracy of image review. Computer vision uses image features, such as colour, shape and texture to infer image content. I provide a brief primer on ecological computer vision to outline its goals, tools and applications to animal ecology. I reviewed 187 existing applications of computer vision and divided articles into ecological description, counting and identity tasks. I discuss recommendations for enhancing the collaboration between ecologists and computer scientists and highlight areas for future growth of automated image analysis.

show abstract

Section: Identitymentioning

confidence: 99%

Section: Collaboration With Computer Vision Researchersmentioning

confidence: 99%

Section: Future Growthmentioning

confidence: 99%

See 1 more Smart Citation

A computer vision for animal ecology

Weinstein

2017

Journal of Animal Ecology

338

287

View full text Add to dashboard Cite

show abstract

“…Following this work, they also utilized deep learning to improve low resolution animal species recognition by training deep CNNs on poor quality images. The data was labeled by experts into two data sets, the first classifying between birds and mammals and the second classification of different mammal species [25], [26].…”

Section: Background and Related Workmentioning

confidence: 99%

Deep Learning Object Detection Methods for Ecological Camera Trap Data

Schneider

Taylor

Kremer

2018

2018 15th Conference on Computer and Robot Vision (CRV)

135

107

View full text Add to dashboard Cite

Deep learning methods for computer vision tasks show promise for automating the data analysis of camera trap images. Ecological camera traps are a common approach for monitoring an ecosystem's animal population, as they provide continual insight into an environment without being intrusive. However, the analysis of camera trap images is expensive, labour intensive, and time consuming. Recent advances in the field of deep learning for object detection show promise towards automating the analysis of camera trap images. Here, we demonstrate their capabilities by training and comparing two deep learning object detection classifiers, Faster R-CNN and YOLO v2.0, to identify, quantify, and localize animal species within camera trap images using the Reconyx Camera Trap and the self-labeled Gold Standard Snapshot Serengeti data sets. When trained on large labeled datasets, object recognition methods have shown success. We demonstrate their use, in the context of realistically sized ecological data sets, by testing if object detection methods are applicable for ecological research scenarios when utilizing transfer learning. Faster R-CNN outperformed YOLO v2.0 with average accuracies of 93.0% and 76.7% on the two data sets, respectively. Our findings show promising steps towards the automation of the labourious task of labeling camera trap images, which can be used to improve our understanding of the population dynamics of ecosystems across the planet.

show abstract

“…The work [19] presents a study of the effectiveness of different deep learning architectures on deciding first if an image shows a bird or mammal and deciding the correct mammal set afterwards using the Snapshot Serengeti dataset [62]. Forwarding images with low confidence decisions to a human expert allows for reaching high accuracies.…”

Section: Automated Wildlife Surveillancementioning

confidence: 99%

Active and Incremental Learning with Weak Supervision

2020

View full text Add to dashboard Cite

Note: This paper is an extended version of our previous work [4], from which certain parts of sections 2 to 5 (except novel YOLO-specific methods) were taken verbatim. Section 8 contains some verbatim parts from our previous work [33].Abstract Large amounts of labeled training data are one of the main contributors to the great success that deep models have achieved in the past. Label acquisition for tasks other than benchmarks can pose a challenge due to requirements of both funding and expertise. By selecting unlabeled examples that are promising in terms of model improvement and only asking for respective labels, active learning can increase the efficiency of the labeling process in terms of time and cost.In this work, we describe combinations of an incremental learning scheme and methods of active learning. These allow for continuous exploration of newly observed unlabeled data. We describe selection criteria based on model uncertainty as well as expected model output change (E-MOC). An object detection task is evaluated in a continuous exploration context on the PASCAL VOC dataset. We also validate a weakly supervised system based on active and incremental learning in a real-world biodiversity application where images from camera traps are analyzed. Labeling only 32 images by accepting or rejecting proposals gen-

show abstract

Animal Identification in Low Quality Camera-Trap Images Using Very Deep Convolutional Neural Networks and Confidence Thresholds

Cited by 30 publications

References 10 publications

A computer vision for animal ecology

A computer vision for animal ecology

Deep Learning Object Detection Methods for Ecological Camera Trap Data

Active and Incremental Learning with Weak Supervision

Contact Info

Product

Resources

About