Application of a computer vision technique to animal-borne video data: extraction of head movement to understand sea turtles’ visual assessment of surroundings

Okuyama, Junichi; Nakajima, Kana; Matsui, Kenta; Nakamura, Yūichi; Kondo, Kazuaki; Koizumi, Toru; Arai, Nobuaki

doi:10.1186/s40317-015-0079-y

Cited by 10 publications

(3 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Video recording observations combined with collection of GPS data improves the collection of precise data, both for identifying the species involved and for determining location and circumstances of the collisions (train speed, bird behavior, etc.). Techniques like video recording from fixed points, vehicles, or animal-borne cameras have been applied to study animal behavior and monitor wildlife (Whorff and Griffing, 1992;Thompson et al, 1999;Okuyama et al, 2015), but rarely to assess mortality risk (Desholm et al, 2006;Cryan et al, 2014;Furness, 2014;Doppler et al, 2015).…”

Section: Discussionmentioning

confidence: 99%

On-Board Video Recording Unravels Bird Behavior and Mortality Produced by High-Speed Trains

et al. 2017

View full text Add to dashboard Cite

Large high-speed railway (HSR) networks are planned for the near future to accomplish increased transport demand with low energy consumption. However, high-speed trains produce unknown avian mortality due to birds using the railway and being unable to avoid approaching trains. Safety and logistic difficulties have precluded until now mortality estimation in railways through carcass removal, but information technologies can overcome such problems. We present the results obtained with an experimental on-board system to record bird-train collisions composed by a frontal recording camera, a GPS navigation system and a data storage unit. An observer standing in the cabin behind the driver controlled the system and filled out a form with data of collisions and bird observations in front of the train. Photographs of the train front taken before and after each journey were used to improve the record of killed birds. Trains running the 321.7 km line between Madrid and Albacete (Spain) at speeds up to 250-300 km/h were equipped with the system during 66 journeys along a year, totaling approximately 14,700 km of effective recording. The review of videos produced 1,090 bird observations, 29.4% of them corresponding to birds crossing the infrastructure under the catenary and thus facing collision risk. Recordings also showed that 37.7% bird crossings were of animals resting on some element of the infrastructure moments before the train arrival, and that the flight initiation distance of birds (mean ± SD) was between 60 ± 33 m (passerines) and 136 ± 49 m (raptors). Mortality in the railway was estimated to be 60.5 birds/km year on a line section with 53 runs per day and 26.1 birds/km year in a section with 25 runs per day. Our results are the first published estimation of bird mortality in a HSR and show the potential of information technologies to yield useful data for monitoring the impact of trains on birds via on-board recording systems. Moreover, recordings point to the use of the infrastructure by birds as a key issue leading to bird train-kill.

show abstract

Section: Discussionmentioning

confidence: 99%

On-Board Video Recording Unravels Bird Behavior and Mortality Produced by High-Speed Trains

et al. 2017

View full text Add to dashboard Cite

show abstract

“…Therefore, in the present preliminary study, we utilized a behavioral observation method by attaching a wearable camera to calves. Several studies have reported their application in goats [ 29 ] and sea turtles [ 30 ], but there is no study on the daily behavioral time budgets. The camera was chosen based on its small size and its ability to be worn on the animal’s body.…”

Section: Introductionmentioning

confidence: 99%

Evaluation of Wearable Cameras for Monitoring and Analyzing Calf Behavior: A Preliminary Study

Saitoh

Kato

2021

Animals

View full text Add to dashboard Cite

Understanding cattle behavior is important for discerning their health and management status. However, manual observations of cattle are time-consuming and labor-intensive. Moreover, during manual observations, the presence or position of a human observer may alter the normal behavior of the cattle. Wearable cameras are small and lightweight; therefore, they do not disturb cattle behavior when attached to their bodies. Thus, this study aimed to evaluate the suitability of wearable cameras for monitoring and analyzing cattle behavior. From December 18 to 27, 2017, this study used four 2-month-old, group-housed Holstein calves at the Field Science Center of the Obihiro University of Agriculture and Veterinary Medicine, Japan. Calf behavior was recorded every 30 s using a wearable camera (HX-A1H, Panasonic, Japan) from 10:00 to 15:30 and observed directly from 11:00 to 12:00 and 14:00 to 15:00. In addition, the same observer viewed the camera recordings corresponding to the direct observation periods, and the results were compared. The correlation coefficients of all behavioral data from direct and wearable camera video observations were significant (p < 0.01). We conclude that wearable cameras are suitable for observing calf behavior, particularly their posture (standing or lying), as well as their ruminating and feeding behaviors.

show abstract

“…Two notable examples include the use of object tracking on the prey of falcons and template matching to determine the head position of sea turtles, though in both cases using 2D images [13,14].…”

Section: Introductionmentioning

confidence: 99%

Joining the dots: reconstructing 3D environments and movement paths using animal-borne devices

McClune

2018

Anim Biotelemetry

View full text Add to dashboard Cite

Background: Animal-attached sensors are increasingly used to provide insights on behaviour and physiology. However, such tags usually lack information on the structure of the surrounding environment from the perspective of a study animal and thus may be unable to identify potentially important drivers of behaviour. Recent advances in robotics and computer vision have led to the availability of integrated depth-sensing and motion-tracking mobile devices. These enable the construction of detailed 3D models of an environment within which motion can be tracked without reliance on GPS. The potential of such techniques has yet to be explored in the field of animal biotelemetry. This report trials an animal-attached structured light depth-sensing and visual-inertial odometry motion-tracking device in an outdoor environment (coniferous forest) using the domestic dog (Canis familiaris) as a compliant test species.Results: A 3D model of the forest environment surrounding the subject animal was successfully constructed using point clouds. The forest floor was labelled using a progressive morphological filter. Trees trunks were modelled as cylinders and identified by random sample consensus. The predicted and actual presence of trees matched closely, with an object-level accuracy of 93.3%. Individual points were labelled as belonging to tree trunks with a precision, recall, and F β score of 1.00, 0.88, and 0.93, respectively. In addition, ground-truth tree trunk radius measurements were not significantly different from random sample consensus model coefficient-derived values. A first-person view of the 3D model was created, illustrating the coupling of both animal movement and environment reconstruction. Conclusions:Using data collected from an animal-borne device, the present study demonstrates how terrain and objects (in this case, tree trunks) surrounding a subject can be identified by model segmentation. The device pose (position and orientation) also enabled recreation of the animal's movement path within the 3D model. Although some challenges such as device form factor, validation in a wider range of environments, and direct sunlight interference remain before routine field deployment can take place, animal-borne depth sensing and visual-inertial odometry have great potential as visual biologging techniques to provide new insights on how terrestrial animals interact with their environments.

show abstract

Application of a computer vision technique to animal-borne video data: extraction of head movement to understand sea turtles’ visual assessment of surroundings

Cited by 10 publications

References 28 publications

On-Board Video Recording Unravels Bird Behavior and Mortality Produced by High-Speed Trains

On-Board Video Recording Unravels Bird Behavior and Mortality Produced by High-Speed Trains

Evaluation of Wearable Cameras for Monitoring and Analyzing Calf Behavior: A Preliminary Study

Joining the dots: reconstructing 3D environments and movement paths using animal-borne devices

Contact Info

Product

Resources

About