A computer vision system for the evaluation of the behaviour and the well-being of the dogs housed in the shelter

The project, funded by the Ministry of Health, aims to develop a computer vision system that monitors the well-being of the dogs housed in shelters with the use of 3D images acquisition sensors. The system shall analyze automatically and continuously the behavior of dogs in the short and especially long-term, detecting anomalies that may be induced by a malaise of the animals, reporting alerts to the experts and staff.

The project, funded by the Ministry of Health, aims to develop a computer vision system able to monitor the well-being of the dogs housed in shelters. It is part of a larger project called "Technological systems for the control and management of canine populations" sponsored by the “Istituto Zooprofilattico Sperimentale dell’Abruzzo e del Molise”, which aims to provide new knowledge and approaches useful for the development of integrated solutions in the biomolecular and engineering fields and for the control and management of canine populations. The aim is to develop a monitoring system able to identify the dog, recognize its body parts and analyze his behavior over the short and long period, to detect any anomalies that could be attributed to a state of malaise of the animal, providing alerts to experts and to staff taking care of the dogs. During the project we will test the use of 3D image acquisition sensors and develop ad hoc algorithms that can process the provided inputs.

Publications

1

Pistocchi, Simone; Calderara, Simone; Barnard, S.; Ferri, N.; Cucchiara, Rita "Kernelized Structural Classification for 3D Dogs Body Parts Detection" Pattern Recognition (ICPR), 2014 22nd International Conference on, Stockholm SWE, pp. 1993 -1998 , 24-28 Aug 2014, 2014 | DOI: 10.1109/ICPR.2014.348 Conference

Video Demo