Guzhva, Oleksiy
- Department of Biosystems and Technology, Swedish University of Agricultural Sciences
Conference paper2015Peer reviewed
Guzhva, O.; Ardö, H.; Herlin, A.; Nilsson, M.; Aström, K.; Bergsten, C.
A well-planned waiting area is crucial for automatic milking systems. In an enclosed waiting area, cows of different rank compete to enter the milking station and are exposed to a variety of social interactions. Such interactions could increase standing time and delay milking, which may result in stress, lameness, impaired welfare and reduced performance. The aim was to monitor the waiting area in a free stall dairy by means of three video cameras in order to detect the occurrence of social interactions using improved image segmentation and tracking methods. The surveillance system observed 250 cows with free access to any of four milking stations during 24 hours over a period of two weeks. A two-step pattern recognition approach was used. In the first step geometric features (distances) were extracted from every pair of cows in every frame. These features form the input to the second step in which a cow behaviour classifier was developed. A support vector machine was used to create this classifier. The social interactions were identified based on collision of geometrical shapes segmented from the image and positively identified as cows by experienced observers. The results showed that the proposed system was capable of reasonably accurate detection of social interactions.
Automatic milking; Computer vision; Lameness; Object recognition; Social interactions
Title: Precision Livestock Farming '15 : papers presented at the 7th conference on precision livestock farming
ISBN: 9788890975325Publisher: European Conference on Precision Livestock Farming
Precision Livestock Farming '15 : 7th conference on precision livestock farming, 2015-09-15--2015-09-18, Milan, Italy
Animal and Dairy Science
Computer Vision and Robotics (Autonomous Systems)
https://res.slu.se/id/publ/132007