Real-time Assessment of Swine Thermal Comfort by Computer Vision
Interactive assessment and control of swine thermal comfort by real-time computer image analysis of pig resting patterns is investigated. The new approach overcomes pitfalls inherent of conventional method in that it uses the animals themselves as an integral biosensor versus the partial air temperature as used in conventional method. The behavior-based system consists of a CCD camera; an image grabber board inside a PC; and a visual C++ program that performs acquisition, processing and classification of animal behavioral images. Critical and challenging to robust operation of this real-time system is proper image feature selections that have little dependence on growth and thus changing body size of the pigs. Image processing algorithms of the system were developed with paper-cut pigs in laboratory settings and briefly validated with live pigs in production settings. Further refinement of the prototype is needed before it can be used in commercial production settings.
This proceeding is from Pp. 362-369 in Proceedings of the World Congress of Computers in Agriculture and Natural Resources (13-15, March 2002, Iguacu Falls, Brazil) 701P0301.