University of Twente Student Theses

Login

Generating IMU Data for Horse Activity Recognition from Horse Videos

Wan, Xin (2021) Generating IMU Data for Horse Activity Recognition from Horse Videos.

[img] PDF
428kB
Abstract:Inertial Measurement Units (IMU) which including accelerometers, gyroscopes and sometimes contains magnetometers, these data are widely used to recognize animal activities in the farm, when it comes to the wildlife, it is costly and dangerous to install the IMU sensors on the wild animals, Moreover, it is also a challenge to monitor and track these wild animals, therefore this makes it difficult to obtain a larger number of labelled IMU data sets to train a desirable machine learning model for Animal activity recognition (AAR). This paper will conduct research on generating IMU data from horse videos of which contains 3 activities of horses (standing, walking and trotting) and then evaluate whether these generated IMU data can be used to classify these three horse activities with a classification model that trained with the real IMU data for horse activity classification. Since accelerometers are used more often than gyroscopes and magnetometers for Animal activity recognition (AAR), therefore, in this research, I only consider generating the accelerations from horse videos. In the future, it will be improved for generating a greater variety of synthesis IMU data (also including the Gyroscope data) from videos of more animal species and more animal activities.
Item Type:Essay (Bachelor)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:54 computer science
Programme:Computer Science BSc (56964)
Link to this item:https://purl.utwente.nl/essays/87374
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page