Our cookies

We use cookies, which are small text files, to improve your experience on our website.
You can allow or reject non essential cookies or manage them individually.

Reject allAllow all

More options  •  Cookie policy

Our cookies

Allow all

We use cookies, which are small text files, to improve your experience on our website. You can allow all or manage them individually.

You can find out more on our cookie page at any time.

EssentialThese cookies are needed for essential functions such as logging in and making payments. Standard cookies can’t be switched off and they don’t store any of your information.
AnalyticsThese cookies help us collect information such as how many people are using our site or which pages are popular to help us improve customer experience. Switching off these cookies will reduce our ability to gather information to improve the experience.
FunctionalThese cookies are related to features that make your experience better. They enable basic functions such as social media sharing. Switching off these cookies will mean that areas of our website can’t work properly.

Save preferences

pigs faceEvaluation of the impact of age and experience on chick's emotional response using computer vision algorithms

 

Year: 2023

Sarah Paola Juárez Jerez
University of Veterinary Medicine Vienna, Austria

Supervisor(s): Dr Sara Hintze, University of Natural Resources and Life Sciences Vienna (BOKU) & Dr Maciej Oczak, University of Veterinary Medicine Vienna, Austria

 


 

Behavioural indicators of emotion are extremely useful for animals’ emotions’ assessment as animals under minimal or no stress should be able to display their natural behaviour patterns. Considering the challenges of behavioural observation, this study aimed to implement a skeleton-based Action Recognition model to detect and classify five behaviours in a group of 24 broiler chicks: Sitting Alert, Standing Alert, Preening, Exploration Pecks, and Feeding Pecks, which added up to 2687 total instances labelled across 144 videos.

A three-stage approach was used for action detection and classification, where object detection (1) and pose estimation (2) models were trained to identify individual chicks and their key body points and subsequently detect and classify the chick’s behaviours (3). A significant aspect of the study was evaluating how various approaches to data pre-processing affected the model's performance.