Automatic lameness detection based on consecutive 3D-video recordings

The objective of this study was to evaluate if a multi-sensor system (milk, activity, body posture) was a better classifier for lameness than the single-sensor-based detection models. Between September 2013 and August 2014, 3629 cow observations were collected on a commercial dairy farm in Belgium. Human locomotion scoring was used as reference for the model development and evaluation. Cow behaviour and performance was measured with existing sensors that were already present at the farm. A prototype of three-dimensional-based video recording system was used to quantify automatically the back posture of a cow. For the single predictor comparisons, a receiver operating characteristics curve was made. For the multivariate detection models, logistic regression and generalized linear mixed models (GLMM) were developed. The best lameness classification model was obtained by the multi-sensor analysis (area under the receiver operating characteristics curve (AUC) = 0.757 ± 0.029), containing a combination of milk and milking variables, activity and gait and posture variables from videos. Second, the multivariate videobased system (AUC = 0.732 ± 0.011) performed better than the multivariate milk sensors (AUC = 0.604 ± 0.026) and the multivariate behaviour sensors (AUC = 0.633 ± 0.018). The video-based system performed better than the combined behaviour and performance-based detection model (AUC = 0.669 ± 0.028), indicating that it is worthwhile to consider a video-based lameness detection system, regardless the presence of other existing sensors in the farm. The results suggest that Θ 2 , the feature variable for the back curvature around the hip joints, with an AUC of 0.719 is the best single predictor variable for lameness detection based on locomotion scoring. In general, this study showed that the video-based back posture monitoring system is outperforming the behaviour and performance sensing techniques for locomotion scoring-based lameness detection. A GLMM with seven specific variables (walking speed, back posture measurement, daytime activity, milk yield, lactation stage, milk peak flow rate and milk peak conductivity) is the best combination of variables for lameness classification. The accuracy on four-level lameness classification was 60.3%. The accuracy improved to 79.8% for binary lameness classification. The binary GLMM obtained a sensitivity of 68.5% and a specificity of 87.6%, which both exceed the sensitivity (52.1% ± 4.7%) and specificity (83.2% ± 2.3%) of the multi-sensor logistic regression model. This shows that the repeated measures analysis in the GLMM, taking into account the individual history of the animal, outperforms the classification when thresholds based on herd level (a statistical population) are used.

Section: Discussionsupporting

confidence: 92%

Section: Discussionsupporting

confidence: 85%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Lameness detection in dairy cattle: single predictor v. multivariate analysis of image-based posture processing and behaviour and performance sensing

Hertem

Bahr

Tello

et al. 2016

“…As lameness is one of the most costly health problems in dairy cows also technology to detect lame cows is being investigated. Several sensors have been tested for their ability to register cow locomotion variables that are related to lameness; for example, weight distribution (Pastell and Kujala, 2007), gait pattern (Maertens et al, 2011) or posture pattern like arching of the back (Van Hertem et al, 2014) (reviewed by Van Nuffel et al, 2015). Such lameness detection systems are based on the assumption that the lameness-relevant-variables change when a cow develops lameness, for example, shorter step length or more arching of the back.…”

Section: Introductionmentioning

confidence: 99%

Environmental and cow-related factors affect cow locomotion and can cause misclassification in lameness detection systems

Nuffel¹,

Gucht²,

Saeys

et al. 2016

To tackle the high prevalence of lameness, techniques to monitor cow locomotion are being developed in order to detect changes in cows' locomotion due to lameness. Obviously, in such lameness detection systems, alerts should only respond to locomotion changes that are related to lameness. However, other environmental or cow factors can contribute to locomotion changes not related to lameness and hence, might cause false alerts. In this study the effects of wet surfaces, dark environment, age, production level, lactation and gestation stage on cow locomotion were investigated. Data was collected at Institute for Agricultural and Fisheries Research research farm (Melle, Belgium) during a 5-month period. The gait variables of 30 non-lame and healthy Holstein cows were automatically measured every day. In dark environments and on wet walking surfaces cows took shorter, more asymmetrical strides with less step overlap. In general, older cows had a more asymmetrical gait and they walked slower with more abduction. Lactation stage or gestation stage also showed significant association with asymmetrical and shorter gait and less step overlap probably due to the heavy calf in the uterus. Next, two lameness detection algorithms were developed to investigate the added value of environmental and cow data into detection models. One algorithm solely used locomotion variables and a second algorithm used the same locomotion variables and additional environmental and cow data. In the latter algorithm only age and lactation stage together with the locomotion variables were withheld during model building. When comparing the sensitivity for the detection of non-lame cows, sensitivity increased by 10% when the cow data was added in the algorithm (sensitivity was 70% and 80% for the first and second algorithm, respectively). Hence, the number of false alerts for lame cows that were actually non-lame, decreased. This pilot study shows that using knowledge on influencing factors on cow locomotion will help in reducing the number of false alerts for lameness detection systems under development. However, further research is necessary in order to better understand these and many other possible influencing factors (e.g. trimming, conformation) of non-lame and hence 'normal' locomotion in cows.

“…Viazzi et al (2014) compared the Microsoft Kinect 1 3D camera mounted in top view position to a 2D camera in side view position with regard to their applicability in lameness detection. Afterwards, the algorithms were improved and the classification results were optimized in Van Hertem et al (2014). The Kinect uses the measurement principle 'Structured Light' (Andersen et al, 2012).…”

mentioning

confidence: 99%

Developing a multi-Kinect-system for monitoring in dairy cows: object recognition and surface analysis using wavelets

Salau

Haas

Thaller

et al. 2016

Camera-based systems in dairy cattle were intensively studied over the last years. Different from this study, single camera systems with a limited range of applications were presented, mostly using 2D cameras. This study presents current steps in the development of a camera system comprising multiple 3D cameras (six Microsoft Kinect cameras) for monitoring purposes in dairy cows. An early prototype was constructed, and alpha versions of software for recording, synchronizing, sorting and segmenting images and transforming the 3D data in a joint coordinate system have already been implemented. This study introduced the application of two-dimensional wavelet transforms as method for object recognition and surface analyses. The method was explained in detail, and four differently shaped wavelets were tested with respect to their reconstruction error concerning Kinect recorded depth maps from different camera positions. The images' high frequency parts reconstructed from wavelet decompositions using the haar and the biorthogonal 1.5 wavelet were statistically analyzed with regard to the effects of image fore-or background and of cows' or persons' surface. Furthermore, binary classifiers based on the local high frequencies have been implemented to decide whether a pixel belongs to the image foreground and if it was located on a cow or a person. Classifiers distinguishing between image regions showed high (⩾0.8) values of Area Under reciever operation characteristic Curve (AUC). The classifications due to species showed maximal AUC values of 0.69.Keywords: dairy cattle, monitoring system, 3D camera, object recognition, wavelet transform ImplicationThe prototype of a multi-Kinect system is presented that captures 3D data of dairy cows in motion for health monitoring. When fully developed, it will provide an inexpensive means of automated early lameness detection and livestock monitoring, thereby helping to increase animal welfare. The large amount of recorded data will provide a solid basis for scientific analyses and can help to deepen the understanding of connections between traits and genetics. Object recognition and methods to analyze the animals' surfaces from the 3D data are important developmental steps that are approached using the wavelet transform. IntroductionSeveral camera-based studies were conducted over the last years yielding high rates of correct classification in lameness detection in dairy cows. Song et al. (2008), Pluk et al. (2012) and Viazzi et al. (2013) used digital 2D cameras to record walking cows from the side at distances of 3 to 6 m and analyzed step overlap, fetlock joints' angles or back posture. There were also several camera-based studies dealing with body condition score (BCS) determination. Azzaro et al. (2011) used linear and polynomial kernel principal analysis to reconstruct cows' shapes. Bercovich et al. (2012) presented BCS-prediction models that were based on five anatomical points. In both studies the animals were observed from a top view position using 2D cameras. The definitio...