Work package 1

The use of animal based warning signals on farm to predict outbreaks of injurious behaviour

Task 1.1: (Leader: WLR, partners: FiBL, INRA, FBN, DMRI) Eyes of the stockman

The stockman plays a crucial role in the early identification of behavioural problems. Farmer Focus Groups will be set up in the participating countries of the project. They will discuss and improve existing protocols for visual observations of injurious behaviours (FareWellDock and EUWelNet outcomes). Two to three meetings will be used for this. The protocols will then be tested on 2 farms of focus group members in each participating country. Daily observations according to the protocol will be carried out by the stockmen on the participating farms, and the results discussed in the focus groups.

Task 1.2 (Leader: INRA, partners: CEA, FBN) Automated assessment of injurious behaviours on farm

The main objective of this task is to develop automated techniques based on the use of sensors to detect patterns of activity that could indicate or predict injurious behaviours. Ten animals will be fitted with ear tags containing a wireless accelerometer and will be observed by video cameras. They will be kept under conventional circumstances, and subjected to a mixing period to provoke aggression. Behaviour records obtained through video will be compared to the sensor data, allowing algorithms to link sensor data and video observations to be developed. Predictive behaviours highlighted by farmers in T1.1 will be used. Finally, the method will be validated on groups of finishing pigs in an experimental setting, so the algorithms can be adjusted and finalised.

Task 1.3 (Leader: INRA, partner: FiBL) Automated assessment of lesion on farm

An automated technique based on digital pictures of pigs using a multi-spectrum camera will be developed. This camera will take simultaneously six pictures allowing to detect the presence of haemoglobin and hence the existence and location of fresh lesions. It will ignore other discolorations e.g. through dirtiness. A prototype of the multi-spectrum camera will be built, so it can be used in farms on live animals. We will validate that the images allow a good detection of the lesions. Secondly, we will optimize the basic settings: distance between the camera and the animal, position of the animal. Thirdly, we will develop and validate algorithms for automatic assessment of the lesions, the output being the presence/absence of lesions, their location, and the % of the body surface presenting lesions as revealed by the presence of haemoglobin.

Task 1.4 (Leader: FiBL, partners: INRA, WLR, DMRI, FBN) On-farm application across participating countries

The Farmer Focus Groups will develop uniform dissemination materials with clear and practical instructions and pictures on how to recognize early signs preceding tail biting and aggression. These instructions will primarily be based on the work in T1.1, but also include descriptions of the technologies developed in T1.2 and T1.3. They will be written in English, and the focus group languages (FR, DE, DK, NL). All participating farmers will then be trained in the use of the ‘Eyes of the farmer’ protocol (T1.1). In addition, in Switzerland, two farmers will receive instructions on the lesion scoring technology (T1.3), and in Germany two farmers will be instructed on the application of the behaviour scoring system (T1.2). All protocols will be applied over a period of one year, and the results discussed within the respective focus groups. In addition to the on-farm assessments of WP1, the abattoir results from T2.2 and T2.3 will be included in farmer focus groups of CH and NL+DK respectively. These results will also be shared across the PigWatch focus groups.