Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 Mar;102(3):102412.
doi: 10.1016/j.psj.2022.102412. Epub 2022 Dec 9.

Passive radio frequency identification and video tracking for the determination of location and movement of broilers

J E Doornweerd  1 , G Kootstra  2 , R F Veerkamp  3 , B de Klerk  4 , I Fodor  3 , M van der Sluis  3 , A C Bouwman  3 , E D Ellen  3
Affiliations

Passive radio frequency identification and video tracking for the determination of location and movement of broilers

J E Doornweerd et al. Poult Sci. 2023 Mar.

Abstract

Phenotypes on individual animals are required for breeding programs to be able to select for traits. However, phenotyping individual animals can be difficult and time-consuming, especially for traits related to health, welfare, and performance. Individual broiler behavior could serve as a proxy for these traits when recorded automatically and reliably on many animals. Sensors could record individual broiler behavior, yet different sensors can differ in their assessment. In this study a comparison was made between a passive radio frequency identification (RFID) system (grid of antennas underneath the pen) and video tracking for the determination of location and movement of 3 color-marked broilers at d 18. Furthermore, a systems comparison of derived behavioral metrics such as space usage, locomotion activity and apparent feeding and drinking behavior was made. Color-marked broilers simplified the computer vision task for YOLOv5 to detect, track, and identify the animals. Animal locations derived from the RFID-system and based on video were largely in agreement. Most location differences (77.5%) were within the mean radius of the antennas' enclosing circle (≤128 px, 28.15 cm), and 95.3% of the differences were within a one antenna difference (≤256 px, 56.30 cm). Animal movement was not always registered by the RFID-system whereas video was sensitive to detection noise and the animal's behavior (e.g., pecking). The method used to determine location and the systems' sensitivities to movement led to differences in behavioral metrics. Behavioral metrics derived from video are likely more accurate than RFID-system derived behavioral metrics. However, at present, only the RFID-system can provide individual identification for non-color marked broilers. A combination of verifiable and detailed video with the unique identification of RFID could make it possible to identify, describe, and quantify a wide range of individual broiler behaviors.

Keywords: activity; broiler; deep learning; sensors; tracking.

PubMed Disclaimer

Figures

Figure 1
Figure 1
Schematic top view of the radio frequency identification (RFID)-grid. One of the fifteen PVC panels is outlined. Distances between antenna center points are shown. The defective antenna (A2) is crossed out. The feeders were positioned on antennas E1, F1, E5 and F5. The drinker was positioned on antenna D1. An antenna's enclosing circle is illustrated at antenna D3. Circular sectors are indicated for the feeders. Reproduced and adapted with permission from van der Sluis et al. (2020).
Figure 2
Figure 2
Input and output YOLOv5s. YOLOv5s returns bounding boxes and classifications for the objects of interests, in this case three color-marked broilers. See YOLOv5s repository for detailed architecture (Jocher et al., 2021).
Figure 3
Figure 3
Location of the black color-marked broiler over time as derived from video (A) and radio frequency identification (RFID; B). The green square represents the track's start, and the red square represents the track's end. The color transition depicts the location's progression from dark blue to dark red over time. The white dots represent the antenna corners. The plot is projected over a darkened video freeze frame.
Figure 4
Figure 4
Differences in location expressed as Euclidean distance between locations as derived from radio frequency identification (RFID) and video. Differences in location are shown for A) the black color-marked broiler, B) the light blue color-marked broiler, and C) the pink color-marked broiler. Locations on video within antenna A2 were excluded. The vertical dashed lines indicate the mean radius of the antennas’ enclosing circle (128 px) and a multiple (256 px). Values to the right of the first vertical dashed line signify a difference that crosses antennas, that is, a broiler is assumed to be on another antenna than its current position on video. The black circle in the boxplot shows the mean difference.
Figure 5
Figure 5
Heatmaps of the black color-marked broiler as derived from video (A) and radio frequency identification (RFID; B). The numbers represent the total amount of time spent on each antenna in seconds. RFID antenna A2 was defective. The animal's bounding box center point locations are included in the heatmap derived from video locations (A) as black dots. The plot is projected over a darkened video freeze frame.
Figure A-1
Figure A-1
Antenna location of the black color-marked broiler over time as derived from radio frequency identification (RFID; solid line) and video-as-RFID (dashed line). Highlighted with red boxes are 3 events; 1) nonregistered small movement, 2) antenna switches caused by small movements of bounding box's center point, 3) movement to nonadjacent antenna. Highlighted by horizontal dashed lines are the defective antenna (A2; gray dashed line) and drinker antenna (D1; blue dashed line). Nonadjacent antenna location differences between RFID and video shown in bottom plot.
Figure A-2
Figure A-2
Difference between radio frequency identification (RFID) and video-as-RFID in total time spent per antenna per color-marked broiler on a symmetric log-scale.
See this image and copyright information in PMC

References

    1. Caplen G., Hothersall B., Murrell J.C., Nicol C.J., Waterman-Pearson A.E., Weeks C.A., Colborne G.R. Kinematic analysis quantifies gait abnormalities associated with lameness in broiler chickens and identifies evolutionary gait differences. PLoS One. 2012;7:1–11. - PMC - PubMed
    1. Derakhshani S.M., Overduin M., van Niekerk T.G.C.M., Groot Koerkamp P.W.G. Implementation of inertia sensor and machine learning technologies for analyzing the behavior of individual laying hens. Animals. 2022;12:536. - PMC - PubMed
    1. Doornweerd J.E., Fodor I., Kootstra G., Veerkamp R.F., De Klerk B., Van der Sluis M., Bouwman A.C., Ellen E.D. Proc. 12th WCGALP. 2022. Automated individual walking distance of group-housed broilers; a comparison between ground-truth, RFID, and video. - PubMed
    1. Ellen E.D., Van Der Sluis M., Siegford J., Guzhva O., Toscano M.J., Bennewitz J., Van Der Zande L.E., Van Der Eijk J.A.J., De Haas E.N., Norton T., Piette D., Tetens J., de Klerk B., Visser B., Bas Rodenburg T. Review of sensor technologies in animal breeding: phenotyping behaviors of laying hens to select against feather pecking. Animals. 2019;9:108. - PMC - PubMed
    1. Gebhardt-Henrich S.G., Fröhlich E.K.F., Burose F., Fleurent J., Gantner M., Zähner M. Individual tracking of laying hens with an RFID-System. Landtechnik. 2014;69:301–307.