A computer vision system for the evaluation of swine welfare based on analysis of lesions on carcases

Human-caused bruises, tail lesions caused by tail biting, ear lesions, among other things, in slaughter pigs are detrimental to animal welfare, a violation of the Animal Welfare Act, and a source of public concern. Animal-based indicators of welfare are quantifiable characteristics on the animal that are reflective of what the animal has experienced. The evaluation of animal welfare indicators on swine carcasses could offer a cost-effective method to welfare analysis and quality assurance. In this project, we propose a system which will perform welfare assessing images from carcasses, obtained by camera installed in a slaughter. The purpose of this project is to develop a computer vision system that uses deep learning and indicators of welfare on swine carcasses to accurately inform about the welfare of swine on-farm, and the relationship between welfare indicators and total or partial condemnations.

Faculty Supervisor:

Yolande Seddon;Seok-bum Ko

Student:

Partner:

Maple Leaf Foods (Mississauga, ON);Olymel

Discipline:

Engineering

Sector:

Agriculture; Wholesale trade

University:

University of Saskatchewan

Program: