Developing camera-trapping protocols for wildlife monitoring in Chinese forests

AbstractCamera traps are used worldwide to monitor wildlife. Despite the increasing availability of Deep Learning (DL) models, the effective usage of this technology to support wildlife monitoring is limited. This is mainly due to the complexity of DL technology and high computing requirements. This paper presents the implementation of the light-weight and state-of-the-art YOLOv5 architecture for automated labeling of camera trap images of mammals in the Białowieża Forest (BF), Poland. The camera trapping data were organized and harmonized using TRAPPER software, an open-source application for managing large-scale wildlife monitoring projects. The proposed image recognition pipeline achieved an average accuracy of 85% F1-score in the identification of the 12 most commonly occurring medium-size and large mammal species in BF, using a limited set of training and testing data (a total of 2659 images with animals).Based on the preliminary results, we have concluded that the YOLOv5 object detection and classification model is a fine and promising DL solution after the adoption of the transfer learning technique. It can be efficiently plugged in via an API into existing web-based camera trapping data processing platforms such as e.g. TRAPPER system. Since TRAPPER is already used to manage and classify (manually) camera trapping datasets by many research groups in Europe, the implementation of AI-based automated species classification will significantly speed up the data processing workflow and thus better support data-driven wildlife monitoring and conservation. Moreover, YOLOv5 has been proven to perform well on edge devices, which may open a new chapter in animal population monitoring in real-time directly from camera trap devices.

Download Full-text

Camera-Trapping Two Felid Species: Monitoring Eurasian Lynx (Lynx lynx) and Wildcat (Felis silvestris) Populations in Mixed Temperate Forest Ecosystems

Mammal Study ◽

10.3106/ms2019-0046 ◽

2020 ◽

Vol 45 (1) ◽

pp. 41 ◽

Cited By ~ 1

Author(s):

Anıl Soyumert

Keyword(s):

Temperate Forest ◽

Forest Ecosystems ◽

Camera Trapping ◽

Lynx Lynx ◽

Felis Silvestris ◽

Eurasian Lynx ◽

Felid Species

Download Full-text

A Regional Approach to Wildlife Monitoring Related to Energy Exploration and Development in Colorado

Open-File Report ◽

10.3133/ofr20081024 ◽

2008 ◽

Author(s):

Natasha B. Kotliar ◽

Zachary H. Bowen ◽

Douglas S. Ouren ◽

Adrian H. Farmer

Keyword(s):

Wildlife Monitoring ◽

Regional Approach ◽

Exploration And Development

Download Full-text

An introduction to wildlife camera trapping monitoring from Chinese Forest Biodiversity Monitoring Network (CForBio)

Biodiversity Science ◽

10.3724/sp.j.1003.2014.14108 ◽

2014 ◽

Vol 22 (6) ◽

pp. 808

Author(s):

Xiao Zhishu

Keyword(s):

Monitoring Network ◽

Camera Trapping ◽

Forest Biodiversity ◽

Biodiversity Monitoring

Download Full-text

Coloring the Panda with an Ethnic Touch

10.1093/oso/9780199393671.003.0008 ◽

2018 ◽

Author(s):

E. Elena Songster

Keyword(s):

World Wide ◽

Scientific Research ◽

Specific Site ◽

Wildlife Monitoring ◽

International Integration ◽

Training Station ◽

And Training

Continued international integration of the post-Deng era (1990s on) transformed panda country. The specific site of the Wanglang reserve became a juncture where the local Baima villagers, international scientists, NGOs, and tourists (both foreign and domestic) competed to define the giant panda’s place in the environment and in China. Persistently pursuing its charter purposes as a scientific research base, the Wanglang reserve becomes a model and training station for wildlife monitoring and experimental conservation. One experiment, ecotourism has a dramatic impact on the area. The colorful ethnic character of the Baima people initially proved to be an asset to World Wide Fund for Nature (WWF) efforts to instigate tourism. The industry took on an identity independent of panda preservation, leading reserve staff to reemphasize Wanglang’s ties to science.

Download Full-text

Quantifying wildlife-livestock interactions and their spatio-temporal patterns: Is regular grid camera trapping a suitable approach?

Ecological Indicators ◽

10.1016/j.ecolind.2020.106565 ◽

2020 ◽

Vol 117 ◽

pp. 106565

Author(s):

Roxana Triguero-Ocaña ◽

Joaquín Vicente ◽

Pablo Palencia ◽

Eduardo Laguna ◽

Pelayo Acevedo

Keyword(s):

Temporal Patterns ◽

Camera Trapping ◽

Regular Grid ◽

Spatio Temporal

Download Full-text

Experimental Evaluation of a LoRa Wildlife Monitoring Network in a Forest Vegetation Area

Future Internet ◽

10.3390/fi13050115 ◽

2021 ◽

Vol 13 (5) ◽

pp. 115

Author(s):

Mike Oluwatayo Ojo ◽

Davide Adami ◽

Stefano Giordano

Keyword(s):

Monitoring Network ◽

Forest Vegetation ◽

Delivery Ratio ◽

Communication Link ◽

Wildlife Monitoring ◽

Design Development ◽

Sustainable Solutions ◽

Iot Applications ◽

Dense Forest ◽

Smart Agriculture

Smart agriculture and wildlife monitoring are one of the recent trends of Internet of Things (IoT) applications, which are evolving in providing sustainable solutions from producers. This article details the design, development and assessment of a wildlife monitoring application for IoT animal repelling devices that is able to cover large areas, thanks to the low power wide area networks (LPWAN), which bridge the gap between cellular technologies and short range wireless technologies. LoRa, the global de-facto LPWAN, continues to attract attention given its open specification and ready availability of off-the-shelf hardware, with claims of several kilometers of range in harsh challenging environments. At first, this article presents a survey of the LPWAN for smart agriculture applications. We proceed to evaluate the performance of LoRa transmission technology operating in the 433 MHz and 868 MHz bands, aimed at wildlife monitoring in a forest vegetation area. To characterize the communication link, we mainly use the signal-to-noise ratio (SNR), received signal strength indicator (RSSI) and packet delivery ratio (PDR). Findings from this study show that achievable performance can greatly vary between the 433 MHz and 868 MHz bands, and prompt caution is required when taking numbers at face value, as this can have implications for IoT applications. In addition, our results show that the link reaches up to 860 m in the highly dense forest vegetation environment, while in the not so dense forest vegetation environment, it reaches up to 2050 m.

Download Full-text