Antler detection from the sky: deer sex ratio monitoring using drone‐mounted thermal infrared sensors

Ito, Yoichi; Miyazaki, Akira; Koyama, Lina; Kamada, Kisa; Nagamatsu, Dai

doi:10.1002/wlb3.01034

Cited by 6 publications

(10 citation statements)

References 29 publications

(31 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Surveys using both sensors from dawn to early morning will also be effective for research on a single species, because mammals and other objects can be distinguished using RGB images (Figure 3e). This methodology will be effective for identifying the sex of sika deer, because we can combine the merits of RGB observation, which can recognize antler shapes and body colors (Liang et al, 2020), and thermal infrared observation, which can detect animals and velvet antlers (Ito et al, 2022). For more effective species detection and identification, image analysis using machine learning also has potential (Corcoran et al, 2021; Hollings et al, 2018; Seymour et al, 2017).…”

Section: Discussionmentioning

confidence: 99%

“…Large mammal detection using thermal infrared sensors is effective (Figure 2), but the RGB sensor was virtually useless at night. Species identification using only thermal infrared images is possible between bears and ungulates if the animals are standing or moving (Figure 4) but difficult if the animals are sitting, which conceals their characteristic movements as previously shown with sika deer and wild boars (Ito et al, 2022). Identification using thermal infrared sensors alone is also not easy to distinguish between sika deer and Japanese serows, unless their shapes and movements can be observed closely.…”

Section: Identification Of Sika Deer and Japanese Serowmentioning

confidence: 94%

“…However, species identification can be done even if the animal is sitting (Figure 3e). Dawn is the best time for surveys using thermal infrared sensors, because the temperature difference between the mammals' bodies and the environment is highest, although it is possible even at night (Ito et al, 2020(Ito et al, , 2022. After sunrise, the temperature of the environment increases, which makes mammal detection more difficult.…”

Section: Identification Of Sika Deer and Japanese Serowmentioning

confidence: 99%

“…Thermal infrared sensors on UAVs are effective to detect mammals due to their homeothermy and temperature difference with their environments, such as surface grounds and waters, rocks and vegetation (Christie et al, 2016; Kays et al, 2019; Seymour et al, 2017). Species identification is also possible in cases of animals with conspicuous differences in shapes, sizes and movements based on thermal infrared images (e.g., sika deer and wild boar [ Sus scrofa ], Ito et al, 2022). However, species identification of mammals with similar sizes and shapes using only thermal infrared images is difficult due to a relatively low sensor resolution and small color difference.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Scanning sympatric sika deer and Japanese serows using drones: A methodological trial in Mt. Asama, Central Japan

Ito

Fukue²,

Minami

2023

Grassland Science

Self Cite

View full text Add to dashboard Cite

The negative impacts of sika deer (Cervus nippon) on Japanese serows (Capricornis crispus) are a subject of concern in the mountainous regions of Japan. To develop a methodology to collect information with regard to the abundance and habitat selection of sympatric large mammals using drones (unmanned aerial vehicles; UAVs) equipped with thermal infrared and red‐green‐blue (RGB) sensors, surveys were conducted at night and in the early morning in a high mountainous habitat. Despite the similar size and shape in thermal infrared images for sika deer and Japanese serows, species identification based on body color using RGB images was possible during the period from dawn to early morning. The UAV surveys also revealed the large abundance of deer (>40 deer/km2) and suggested different habitat selection between deer and serows in the alpine grassland. Our results were the first snapshots of sympathetic deer and serow distribution. UAV surveys using both thermal infrared and RGB sensors in early morning could be useful for research and monitoring of large mammals in open areas, such as grasslands, marshlands, high mountains, farmlands and drylands.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Identification Of Sika Deer and Japanese Serowmentioning

confidence: 94%

Section: Identification Of Sika Deer and Japanese Serowmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Scanning sympatric sika deer and Japanese serows using drones: A methodological trial in Mt. Asama, Central Japan

Ito

Fukue²,

Minami

2023

Grassland Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…They found that other methodological challenges included the regulations limiting UAV use and weather conditions. Another recent study also tested antler detection with drone mounted TIR cameras, successfully identifying antlered deer at night during the antler growth period (Ito et al 2022).…”

Section: Introductionmentioning

confidence: 99%

Assessment of the accuracy of counting large ungulate species (red deer Cervus elaphus) with UAV‐mounted thermal infrared cameras during night flights

Zabel

Findlay²,

White

2023

Wildlife Biology

View full text Add to dashboard Cite

Unmanned aerial vehicles (UAVs) are increasingly used in wildlife surveying, including estimation of population densities. It is essential that we evaluate and test new survey methods to guide optimal sampling strategies. This study aimed to assess the accuracy of using a UAV-mounted thermal infrared (TIR) camera to count red deer Cervus elaphus populations, and how this was influenced by flight season, height and velocity, in order to help guide future census design. We flew 57 flights across a captive population of red deer in a 13 ha deer park enclosure of semi-natural habitat, representative of the species' range in northern Germany. Flights and image assessments were performed with no prior knowledge of actual population size. Accuracy was quantified by comparing real population size (known only to deer park staff) and independently estimated population sizes from UAV TIR images. Accuracy was significantly influenced by ecological season (early and late winter, spring and early summer) and height. Across all seasons, lower flights (100 m) performed better than higher ones (120 m), with lower flights in early winter and early summer being on average accurate to within 1% of actual population counts. For the season where we had the largest range of temperatures between flights (late winter) we found that accuracy was highest when temperatures were lowest. Flights were also able to identify all five stags (defined as a male deer ≥ 2 years old) present in early summer, but not in spring. Deer appeared to avoid the landing/take-off area, but there were no noted behavioural responses to drones flying over animals when at constant height and velocity during surveys. Our results indicate that UAV-mounted TIR camera have the potential to accurately count populations of large ungulate species, but that flight season, height and potentially temperature need to be taken into account to maximise accuracy. This approach has the potential to be scaled up to more accurately estimate densities of wild populations compared to existing approaches.

show abstract

Advancing primate surveillance with image recognition techniques from unmanned aerial vehicles

He,

Zhang,

Wang

et al. 2024

American J Primatol

View full text Add to dashboard Cite

Using unmanned aerial vehicles (UAVs) for surveys on thermostatic animals has gained prominence due to their ability to provide practical and precise dynamic censuses, contributing to developing and refining conservation strategies. However, the practical application of UAVs for animal monitoring necessitates the automation of image interpretation to enhance their effectiveness. Based on our past experiences, we present the Sichuan snub‐nosed monkey (Rhinopithecus roxellana) as a case study to illustrate the effective use of thermal cameras mounted on UAVs for monitoring monkey populations in Qinling, a region characterized by magnificent biodiversity. We used the local contrast method for a small infrared target detection algorithm to collect the total population size. Through the experimental group, we determined the average optimal grayscale threshold, while the validation group confirmed that this threshold enables automatic detection and counting of target animals in similar datasets. The precision rate obtained from the experiments ranged from 85.14% to 97.60%. Our findings reveal a negative correlation between the minimum average distance between thermal spots and the count of detected individuals, indicating higher interference in images with closer thermal spots. We propose a formula for adjusting primate population estimates based on detection rates obtained from UAV surveys. Our results demonstrate the practical application of UAV‐based thermal imagery and automated detection algorithms for primate monitoring, albeit with consideration of environmental factors and the need for data preprocessing. This study contributes to advancing the application of UAV technology in wildlife monitoring, with implications for conservation management and research.

show abstract

Antler detection from the sky: deer sex ratio monitoring using drone‐mounted thermal infrared sensors

Cited by 6 publications

References 29 publications

Scanning sympatric sika deer and Japanese serows using drones: A methodological trial in Mt. Asama, Central Japan

Scanning sympatric sika deer and Japanese serows using drones: A methodological trial in Mt. Asama, Central Japan

Assessment of the accuracy of counting large ungulate species (red deer Cervus elaphus) with UAV‐mounted thermal infrared cameras during night flights

Advancing primate surveillance with image recognition techniques from unmanned aerial vehicles

Contact Info

Product

Resources

About