Recognition of common postures in lambs (ICV-TxLamb)
- Citation Author(s):
-
Yair González Baldizón
(TecNM/IT Tuxtla Gutiérrez)
Madain Pérez-Patricio
(TecNM/IT Tuxtla Gutiérrez)
Claudia Mendoza Avendaño
(TecNM/IT Tuxtla Gutiérrez)
Andrea Magadán
(TecNM / CENIDET)
Alicia Morales-Reyes
(INAOE, Computer Science Department)
E.N. Escobar-Gómez
(TecNM/IT Tuxtla Gutiérrez)
Oscar M. Rodríguez-Elías
(TecNM/IT Hermosillo)
H.D. Vázquez-Delgado
(TecNM/IT Tuxtla Gutiérrez)
A. Fragoso-Mandujano
(TecNM/IT Tuxtla Gutiérrez)
- Submitted by:
- Yair Gonzalez Baldizon
- Last updated:
- DOI:
- 10.21227/3tyc-y227
- Data Format:
- Research Article Link:
876 views
- Categories:
- Keywords:
Abstract
Computer vision in animal monitoring has become a research application in stable or confined conditions.
Detecting animals from the top view is challenging due to barn conditions.
In this dataset called ICV-TxLamb, images are proposed for the monitoring of lamb inside a barn.
This set of data is made up of two categories, the first is lamb (classifies the only lamb), the second consists of four states of the posture of lambs, these are: eating, sleeping, lying down, and normal (standing or without activity ).
The added tags have been considered by the natural movements that the lamb makes in the observed scene.
Instructions:
In the content of the folder, there are two folders, one called "lamb" and the other "lamb Postures") previously labelled by the ICV-TX Group of TecNM / IT Tuxtla Gutiérrez, indicating the positions of the sheep such as Eating, Sleeping, Lying and Normal.
The dataset consists of images of lambs in different contexts and from a top view that contains a total of 2390 images with 6419 objects labelled in the two groups Lamb and Lamb Postures of Eating, Sleeping, Lying Down and Normal.
The labels were made using the open-source LabelImg tool which can store annotations in the machine learning annotation formats, YOLO, VOC, so both annotation formats are available. Image annotation consists of a position of the bounding box around each object of interest, the size of the bounding box in terms of width and height, and the corresponding class label (lamb, Eating, Sleeping, Lying, and Normal).
Note: The original dataset was compressed into parts of a .RAR file, but due to errors in loading large files and due to the platform requesting .zip files, it was compressed again in a ZIP folder with its corresponding name. The decompression of the two formats generates the DATASET folder. Sorry for breaking it down into so many small parts, it was just trying to upload one file.
