A Caltech Library Service

A deep active learning system for species identification and counting in camera trap images

Norouzzadeh, Mohammad Sadegh and Morris, Dan and Beery, Sara and Joshi, Neel and Jojic, Nebojsa and Clune, Jeff (2021) A deep active learning system for species identification and counting in camera trap images. Methods in Ecology and Evolution, 12 (1). pp. 150-161. ISSN 2041-210X.

[img] PDF - Supplemental Material
See Usage Policy.


Use this Persistent URL to link to this item:


1. A typical camera trap survey may produce millions of images that require slow, expensive manual review. Consequently, critical conservation questions may be answered too slowly to support decision‐making. Recent studies demonstrated the potential for computer vision to dramatically increase efficiency in image‐based biodiversity surveys; however, the literature has focused on projects with a large set of labeled training images, and hence many projects with a smaller set of labeled images cannot benefit from existing machine learning techniques. Furthermore, even sizable projects have struggled to adopt computer vision methods because classification models overfit to specific image backgrounds (i.e., camera locations). 2. In this paper, we combine the power of machine intelligence and human intelligence via a novel active learning system to minimize the manual work required to train a computer vision model. Furthermore, we utilize object detection models and transfer learning to prevent overfitting to camera locations. To our knowledge, this is the first work to apply an active learning approach to camera trap images. 3. Our proposed scheme can match state‐of‐the‐art accuracy on a 3.2 million image dataset with as few as 14,100 manual labels, which means decreasing manual labeling effort by over 99.5%. Our trained models are also less dependent on background pixels, since they operate only on cropped regions around animals. 4. The proposed active deep learning scheme can significantly reduce the manual labor required to extract information from camera trap images. Automation of information extraction will not only benefit existing camera trap projects, but can also catalyze the deployment of larger camera trap arrays.

Item Type:Article
Related URLs:
URLURL TypeDescription ItemPeer review history ItemTrained models and the live version of the source code ItemDatasets
Norouzzadeh, Mohammad Sadegh0000-0002-2983-9374
Beery, Sara0000-0002-2544-1844
Additional Information:© 2020 British Ecological Society. Issue Online: 10 January 2021; Version of Record online: 20 November 2020; Accepted manuscript online: 14 October 2020; Manuscript accepted: 04 September 2020; Manuscript received: 02 February 2020. Peer Review: The peer review history for this article is available at‐210X.13504. Data Availability Statement: The released version of the source code is available at (Norouzzadeh, 2020). Trained models and the live version of the source code are available at (‐et‐al‐2020/research/active_learning). The North American Camera Trap Images (NACTI), Snapshot Serengeti (SS) and Caltech Camera Traps (CCT) datasets are accessible in the Labeled Information Library of Alexandria: Biology & Conservation (LILA BC) digital repository (
Subject Keywords:active learning; camera trap images; computer vision; deep learning; deep neural networks
Issue or Number:1
Record Number:CaltechAUTHORS:20201015-152731988
Persistent URL:
Official Citation:Norouzzadeh, MS, Morris, D, Beery, S, Joshi, N, Jojic, N, Clune, J. A deep active learning system for species identification and counting in camera trap images. Methods Ecol Evol. 2021; 12: 150–161.
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:106086
Deposited By: George Porter
Deposited On:16 Oct 2020 19:55
Last Modified:15 Jan 2021 21:09

Repository Staff Only: item control page