Automatic detection, tracking and counting of birds in marine video content

T'Jampens, Roeland; Hernández, F.; Vandecasteele, Florian; Verstockt, Steven

doi:10.1109/ipta.2016.7821031

Cited by 17 publications

(8 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another improvement over the rangefinder is a fixed field of view that does away with human intervention to point at a flying animal of interest. Thus, one may use a programmable camera for motion detection to automatically trigger the other cameras without being present (T'Jampens et al ., 2016). Rangefinders are difficult to use for small, fast‐moving birds, and likely create inherent biases for close, consistently moving, and larger birds (Desholm et al ., 2006; Harwood et al ., 2018).…”

Section: Discussionmentioning

confidence: 99%

Estimating bird flight height using 3‐D photogrammetry

et al. 2021

View full text Add to dashboard Cite

Harnessing wind or solar power have become popular “green” options for energy production. However, colliding with wind turbine blades or being burned by concentrated solar flux around power towers can present a substantial threat to birds. Assessing the severity of this risk to different bird species requires accurate estimates of their flight height. We developed a three‐dimensional (3‐D) stereophotogrammetric approach to determine bird flight heights. The accuracy of four varying stereophotogrammetric camera layouts was compared between each other and against laser‐based rangefinder measurements of static structures. Bird flight heights were measured and compared between species, and repetitive photographic captures over short time periods were tested for autocorrelation. Three out of four camera layouts performed equally well when measuring static structures at distances of up to 100 m (0.0 ± 0.3%; or 0.00 ± 0.03 m error), better than laser‐based rangefinders (0.3 ± 4.8%; or 0.12 ± 0.51 m error) on a small target. Photogrammetrically measured flight heights were precise to 0.07 ± 0.05 m up to ~275 m away and to within 1 m at 400 m, and measurable up to ~535 m away. Using this tested approach, repetitive, sequential flight heights of moving birds were significantly autocorrelated compared to random flight heights (P = 0.001). Species‐specific flight heights were distinct, practically demonstrating the approach’s potential application, however, scarcity of flight height data prompts further application of the approach to record distributions of flight height. This stereophotogrammetric method was accurate, cost‐effective, objective, and relatively simple to apply. It could measure flight heights, and potentially micro‐avoidance behaviour in 3‐D flight patterns, to ultimately identify species that are at potential risk of collision or burning with wind turbines and solar towers.

show abstract

Section: Discussionmentioning

confidence: 99%

Estimating bird flight height using 3‐D photogrammetry

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Studies [ 28 , 29 ] discuss the behavior of birds in the sky and on water, in addition to evaluating different types of classifiers.…”

Section: Previous Researchmentioning

confidence: 99%

A System Using Artificial Intelligence to Detect and Scare Bird Flocks in the Protection of Ripening Fruit

Marcoň

Janoušek

Pokorny

et al. 2021

Sensors

View full text Add to dashboard Cite

Flocks of birds may cause major damage to fruit crops in the ripening phase. This problem is addressed by various methods for bird scaring; in many cases, however, the birds become accustomed to the distraction, and the applied scaring procedure loses its purpose. To help eliminate the difficulty, we present a system to detect flocks and to trigger an actuator that will scare the objects only when a flock passes through the monitored space. The actual detection is performed with artificial intelligence utilizing a convolutional neural network. Before teaching the network, we employed videocameras and a differential algorithm to detect all items moving in the vineyard. Such objects revealed in the images were labeled and then used in training, testing, and validating the network. The assessment of the detection algorithm required evaluating the parameters precision, recall, and F1 score. In terms of function, the algorithm is implemented in a module consisting of a microcomputer and a connected videocamera. When a flock is detected, the microcontroller will generate a signal to be wirelessly transmitted to the module, whose task is to trigger the scaring actuator.

show abstract

“…Here, we provide an overview of how deep neural networks can be used to detect every bird inside an image. Examples of similar techniques have been for detection of marine birds from videos [51], tracking of Serengeti wild animals in camera-trap images [52], detection of wild birds using UAV imagery [53], detection of Interior Least Tern in uncontrolled outdoor videos [54]. A deep neural network represents a relatively complex hypothesis function that maps features (e.g., image pixels) to the desired output (e.g., pixel class labels for classification, bounding boxes for detection) by computations in multiple layers.…”

Section: Automated Bird Detectionmentioning

confidence: 99%

Automated Bird Counting with Deep Learning for Regional Bird Distribution Mapping

Akçay

Kabasakal

Aksu

et al. 2020

Animals

View full text Add to dashboard Cite

A challenging problem in the field of avian ecology is deriving information on bird population movement trends. This necessitates the regular counting of birds which is usually not an easily-achievable task. A promising attempt towards solving the bird counting problem in a more consistent and fast way is to predict the number of birds in different regions from their photos. For this purpose, we exploit the ability of computers to learn from past data through deep learning which has been a leading sub-field of AI for image understanding. Our data source is a collection of on-ground photos taken during our long run of birding activity. We employ several state-of-the-art generic object-detection algorithms to learn to detect birds, each being a member of one of the 38 identified species, in natural scenes. The experiments revealed that computer-aided counting outperformed the manual counting with respect to both accuracy and time. As a real-world application of image-based bird counting, we prepared the spatial bird order distribution and species diversity maps of Turkey by utilizing the geographic information system (GIS) technology. Our results suggested that deep learning can assist humans in bird monitoring activities and increase citizen scientists’ participation in large-scale bird surveys.

show abstract

Automatic detection, tracking and counting of birds in marine video content

Cited by 17 publications

References 11 publications

Estimating bird flight height using 3‐D photogrammetry

Estimating bird flight height using 3‐D photogrammetry

A System Using Artificial Intelligence to Detect and Scare Bird Flocks in the Protection of Ripening Fruit

Automated Bird Counting with Deep Learning for Regional Bird Distribution Mapping

Contact Info

Product

Resources

About