Automated identification of animal species in camera trap images

By Xiaoyuan Yu, Jiangping Wang, R. Kays, Patrick A. Jansen, Tianjiang Wang, Thomas S. Huang. 2013

243 citations recorded by the source corpus.

Image sensors are increasingly being used in biodiversity monitoring, with each study generating many thousands or millions of pictures. Efficiently identifying the species captured by each image is a critical challenge for the advancement of this field. Here, we present an automated species identification method for wildlife pictures captured by remote camera traps. Our process starts with images that are cropped out of the background. We then use improved sparse coding spatial pyramid matching (ScSPM), which extracts dense SIFT descriptor and cell-structured LBP (cLBP) as the local features, that generates global feature via weighted sparse coding and max pooling using multi-scale pyramid kernel, and classifies the images by a linear support vector machine algorithm. Weighted sparse coding is used to enforce both sparsity and locality of encoding in feature space. We tested the method on a dataset with over 7,000 camera trap images of 18 species from two different field cites, and achieved an average classification accuracy of 82%. Our analysis demonstrates that the combination of SIFT and cLBP can serve as a useful technique for animal species recognition in real, complex scenarios.

Subfields: Computer Vision and Pattern Recognition, Media Technology, Molecular Biology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Media Technology

In Molecular Biology

Citations in the corpus

Recognition in Terra Incognita (2018, background): “[17] showed classification results on data that provides image sequences of 1̃0 frames.”
Towards automatic wild animal monitoring: Identification of animal species in camera-trap images using very deep convolutional neural networks (2016, methodology): “An important difference from [4] is that they use an automatic segmentation method (Ensemble Video Object Cut) for cropping the animals from the images and use this crops to train and test their system.”
Use of object detection in camera trap image identification: assessing a method to rapidly and accurately classify human and animal detections for research and application in recreation ecology (2022, background): “Researchers have proposed the use of machine learning to identify species in camera trap images (Tabak et al., 2019; Willi et al., 2019; Yu et al., 2013), but tests thus far show performance of these models is generally unsuitable for robust use in real world ecological research (Schneider et al.,…”
Background Subtraction in Real Applications: Challenges, Current Models and Future Directions (2019, methodology): “In these application, the detection done by background subtraction is followed by a tracking phase or a recognition phase [32][398][385].”
Efficient Pipeline for Camera Trap Image Review (2019, methodology): “The computer vision community has been making strides towards automating the species classification challenge in camera traps [1, 2, 4–16], but it has proven difficult to to apply models trained in one region to images collected in different geographic areas.”
The iWildCam 2019 Challenge Dataset (2019, background): “Automating camera trap labeling is not a new challenge for the computer vision community [1, 3–10, 12–17].”
Machine learning to classify animal species in camera trap images: applications in ecology (2018, background): “Recently, machine learning has emerged as a potential solution for automatically classifying images from camera traps (Chen, Han, He, Kays, & Forrester, 2014; Gomez Villa, Salazar, & Vargas, 2017; Norouzzadeh et al., 2018; Swinnen, Reijniers, Breno, & Leirs, 2014; Yu et al., 2013).”
Fast human-animal detection from highly cluttered camera-trap images using joint background modeling and deep learning classification (2017, methodology): “Dense SIFT descriptors and cell-structured local binary patterns as local features were extracted from improved sparse coding spatial pyramid matching (ScSPM) and used to recognize 18 species of animals on a dataset with over 7,000 camera trap manually cropped images [7].”
Identifying animal species in camera trap images using deep learning and citizen science (2018, result): “Even so, the accuracies that were achieved, for example, 82% by Yu et al. (2013), were not comparable to human accuracies of 96.6% (Swanson et al., 2016).”
camtrapR: an R package for efficient camera trap data management (2016, background): “…as automatic identiﬁcation tools are currently still too unreliable and need reference data for all species potentially present in the study area (Yu et al. 2013; but see McShea et al. 2016). camtrapR supports two diﬀerent ways of identifying species: (i) by assigning species tags to images in…”
Automatically identifying wild animals in camera trap images with deep learning (2017, background): “(26) extracted the features with sparse coding spatial pyramid matching (28) and harnessed a linear support vector machine to classify the images.”