Description of movement sensor dataset for dog behavior classification

• DOI: 10.1016/j.dib.2022.107822
• Publication Date: 2022-01-11T02:15:57Z
• Authors: Antti Vehkaoja; Sanni Somppi; Heini Törnqvist; Anna Valldeoriola...; Pekka Kumpulainen; Heli Väätäjä; Päivi Majaranta; Veikko Surakka; Miiamaaria V. Kujala; Outi Vainio
• Keywords: Sitting; Post hoc; Sniffing

Movement sensor data from seven static and dynamic dog behaviors (sitting, standing, lying down, trotting, walking, playing, (treat) searching i.e. sniffing) was collected 45 middle to large sized dogs with six degree-of-freedom movement sensors attached the collar harness. With 17 collection procedure repeated. The duration of each approximately three minutes. order tasks varied between two repetitions (for dogs). were annotated post-hoc based on video recordings made camcorders during tests one second resolution. annotations accurately synchronized raw data. originally used for training behavior classification machine learning algorithms classifying behaviors. developed signal processing are provided together measurement reference annotations. description results original investigation that dataset relates found in: P. Kumpulainen, A. Valldeoriola Cardó, S. Somppi, H. Törnqvist, Väätäjä, Majaranta, Y. Gizatdinova, C. Hoog Antink, V. Surakka, M. Kujala, O. Vainio, Vehkaoja, Dog placed harness collar, Applied Animal Science, 241 (2021), 105,393.