1 3 Evolving Systems DOI 10.1007/s12530-016-9144-x ORIGINAL PAPER Adaptive cow movement detection using evolving spiking neural network models Tao Gao 1,2 · Nikola Kasabov 2 Received: 14 October 2015 / Accepted: 25 January 2016 © Springer-Verlag Berlin Heidelberg 2016 1 Introduction At present, although the high density and centralized dairy farming system has been formed, there are still many prob- lems which influence the milk quality, such as low produc- tion efficiency, high cost and so on. The main reason is that the extensive management relies too much on human labors without enough automation process interaction. For exam- ple, dairy industry is still widely used artificial observation method for feeding, and the performance of this method is limited due to the technical quality of breeders, which not only restricts the efficiency of milk production, but also leads to more cost of resource. Because breeders can not effectively detect the physiological or psychological needs and changes of cows in the breeding process, the lag prob- lems always lead to the decreases of cow welfare and milk nutrition, and more farming resources are wasted. In growth of dairy cows, behaviors such as feeding, drinking, rumination, excretion, or inquiry throughout the process which may reflect the long-time development in different stage and different physiological status of cows. These behaviors are classified as normal behaviors and abnormal behaviors. The normal behaviors are genetic, and adapt to reflect the self instinct maintenance about self wel- fare and healthy growth performance. Abnormal behaviors are usually occurred in an adverse stimulus which shows in the unfavorable biological conditions which lead to the decline of cow welfare. Because of the schedule and energy limitation, the traditional artificial observation method is with low efficiency. Also due to the high environmen- tal demanding standards of dairy cattle breeding, breed- ers can not frequently enter into the cultivation area, some abnormal behaviors or events during the breeding pro- cess are often neglected, which may result to serious eco- nomic losses. Although some farms installed monitoring Abstract For the need of automatic and intelligent dairy cow farming, it is important to combine more informa- tion technologies with the surveillance system. Different gestures may provide different healthy status information of the cow, in order to timely detect the abnormal activ- ity and reduce the workload of breeder, a system which imitates the cognition of human brain is proposed in this paper. First, AdaBoost method is used to detect the posi- tion of cow in the surveillance video sequence, and then the 3D Evolving Spiking Neural Network Model is train- ing to recognize different gestures and classify the activity in the real-time video surveillance. Experiment shows the average accuracy of detection and classification of the pro- posed system is about 80 %, and the performance is robust in complex natural environment. The proposed method can be used as the base of whole alert system which can help breeders to discover abnormal activity and to prevent the diseases in advance, so as to improve the welfare of dairy cow and the quality of milk. Keywords Diary cow farming · Cow detection · Gesture classification · AdaBoost · NeuCube · Evolving Spiking Neural Network · Video surveillance * Tao Gao Gaotao863@163.com 1 Department of Automation, North China Electric Power University, Baoding 071003, Hebei, China 2 Knowledge Engineering and Discovery Research Institute, Auckland University of Technology, Auckland 1010, New Zealand