An Adaptive Automatic Approach to Filtering Empty Images from Camera Traps Using a Deep Learning Model. (14th May 2021)

• Record Type: Journal Article
• Authors: Yang, Deng‐Qi; Ren, Guo‐Peng; Tan, Kun; Huang, Zhi‐Pang; Li, De‐Pin; Li, Xiao‐Wei; Wang, Jian‐Ming; Chen, Ben‐Hui; Xiao, Wen
• Is Part Of: Wildlife Society bulletin. Volume 45:Number 2(2021)
• Page Start: 230
• Page End: 236
• Publication Date: 2021-05-14
• Subjects: Artificial intelligence -- camera traps -- deep learning -- empty images -- image recognition -- wildlife monitoring; Wildlife management -- Periodicals; Wildlife conservation -- Periodicals; 333.9540973
• DOI: 10.1002/wsb.1176
• Languages: English
• ISSNs: 0091-7648

Abstract

ABSTRACT: Camera traps are widely used in wildlife surveys because they are non‐invasive, low‐cost, and highly efficient. Camera traps deployed in the wild often produce large datasets, making it increasingly difficult to manually classify images. Deep learning is a machine learning method that provides a tool to automatically identify images, but it requires labeled training samples and high‐performance servers with multiple Graphics Processing Units (GPUs). However, manually preparing large‐scale training images for training deep learning models is labor intensive, and the high‐performance servers with multiple GPUs are often not available for wildlife management agencies and field researchers. Our study explores an adaptive deep learning method to use small‐scale training sets and a commonly‐available, desktop personal computer (PC) to achieve automatic filtering of empty camera images. Our results showed that by using 29, 192 training samples, the overall error, commission error, and omission error of the proposed method on a PC were 2.69%, 6.82%, and 6.45%, respectively. Moreover, the accuracy of our method can be adaptively improved on PCs in actual ecological monitoring projects, which would benefit researchers in field settings when only a PC is available. © 2021 The Wildlife Society. : An adaptive automatic method of empty camera trap images was proposed, which used a small‐scale training set to complete the training of a deep learning model on a common PC. The method will benefit researchers in the field setting when only one PC is available. It can also greatly reduce the workload of manually labeling training samples.