Computer Vision for Detection of Body Posture and Behavior of Red Foxes

Abstract

The behavior of animals is related to their health and welfare status. The latter plays a particular role in animal experiments, where continuous monitoring is essential for animal welfare. In this study, we focus on red foxes in an experimental setting and study their behavior. Although animal behavior is a complex concept, it can be described as a combination of body posture and activity. To measure body posture and activity, video monitoring can be used as a non-invasive and cost-efficient tool. While it is possible to analyze the video data resulting from the experiment manually, this method is time consuming and costly. We therefore use computer vision to detect and track the animals over several days. The detector is based on a neural network architecture. It is trained to detect red foxes and their body postures, i.e., 'lying', 'sitting', and 'standing'. The trained algorithm has a mean average precision of 99.91%. The combination of activity and posture results in nearly continuous monitoring of animal behavior. Furthermore, the detector is suitable for real-time evaluation. In conclusion, evaluating the behavior of foxes in an experimental setting using computer vision is a powerful tool for cost-efficient real-time monitoring.

Keywords: YOLOv4; animal activity; animal behavior; animal monitoring; animal welfare; body posture; computer vision.

Figure 1

Workflow: video data evaluation. White: raw video material. Gray: detection and analysis. Yellow: results.

Figure 2

Detection of red foxes in single frames. (a–d) detection examples of sitting foxes; (e–h), standing foxes; and (i–l) lying foxes. Two day scene images (left) and night scene images (right), respectively.

Figure 3

Activity overview of six days from a time period of 11 days.

Figure 3

Activity overview of six days from a time period of 11 days.

Figure 4

Posture overview of 6 days from a period of 11 days.

Figure 4

Posture overview of 6 days from a period of 11 days.

Figure 5

Number of changes on day 1 in half hour steps.

Figure 6

Postures of the red fox over 30 min during two different periods. Each vertical line represents a posture change. (a) Day 1—07:30 to 08:00; (b) Day 1—11:00 to 11:30.

Figure 7

Decision tree: Determination of behavior (yellow) through a combined analysis of body posture (blue) and activity level (gray). For example, if a fox is classified as ‘active’ and ‘standing’, it shows the behavior ‘active standing’. This refers to a standing fox with an activity level of ‘active’, and this could be scratching, eating, etc.

Figure 8

Overview of the exhibited behavior of the red fox in timelines of 30 min for two different periods. (a) Day 1—09:00 to 09:30; (b) Day 1—11:00 to 11:30. Note: The fox does not show two behaviors at the same time at any point, although it may appear like that in the figure due to the high density of data points.