1
|
Taghavi M, Russello H, Ouweltjes W, Kamphuis C, Adriaens I. Cow key point detection in indoor housing conditions with a deep learning model. J Dairy Sci 2024; 107:2374-2389. [PMID: 37863288 DOI: 10.3168/jds.2023-23680] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/01/2023] [Accepted: 10/02/2023] [Indexed: 10/22/2023]
Abstract
Lameness in dairy cattle is a costly and highly prevalent problem that affects all aspects of sustainable dairy production, including animal welfare. Automation of gait assessment would allow monitoring of locomotion in which the cows' walking patterns can be evaluated frequently and with limited labor. With the right interpretation algorithms, this could result in more timely detection of locomotion problems. This in turn would facilitate timely intervention and early treatment, which is crucial to reduce the effect of abnormal behavior and pain on animal welfare. Gait features of dairy cows can potentially be derived from key points that locate crucial anatomical points on a cow's body. The aim of this study is 2-fold: (1) to demonstrate automation of the detection of dairy cows' key points in a practical indoor setting with natural occlusions from gates and races, and (2) to propose the necessary steps to postprocess these key points to make them suitable for subsequent gait feature calculations. Both the automated detection of key points as well as the postprocessing of them are crucial prerequisites for camera-based automated locomotion monitoring in a real farm environment. Side-view video footage of 34 Holstein-Friesian dairy cows, captured when exiting the milking parlor, were used for model development. From these videos, 758 samples of 2 successive frames were extracted. A previously developed deep learning model called T-LEAP was trained to detect 17 key points on cows in our indoor farm environment with natural occlusions. To this end, the dataset of 758 samples was randomly split into a train (n = 22 cows; no. of samples = 388), validation (n = 7 cows; no. of samples = 108), and test dataset (n = 15 cows; no. of samples = 262). The performance of T-LEAP to automatically assign key points in our indoor situation was assessed using the average percentage of correctly detected key points using a threshold of 0.2 of the head length (PCKh0.2). The model's performance on the test set achieved a good result with PCKh0.2: 89% on all 17 key points together. Detecting key points on the back (n = 3 key points) of the cow had the poorest performance PCKh0.2: 59%. In addition to the indoor performance of the model, a more detailed study of the detection performance was conducted to formulate postprocessing steps necessary to use these key points for gait feature calculations and subsequent automated locomotion monitoring. This detailed study included the evaluation of the detection performance in multiple directions. This study revealed that the performance of the key points on a cows' back were the poorest in the horizontal direction. Based on this more in-depth study, we recommend the implementation of the outlined postprocessing techniques to address the following issues: (1) correcting camera distortion, (2) rectifying erroneous key point detection, and (3) establishing the necessary procedures for translating hoof key points into gait features.
Collapse
Affiliation(s)
- M Taghavi
- Wageningen Livestock Research, Wageningen University and Research, 6708 WD Wageningen, the Netherlands.
| | - H Russello
- Agricultural Biosystems Engineering, Wageningen University and Research, 6700 AA Wageningen, the Netherlands
| | - W Ouweltjes
- Wageningen Livestock Research, Wageningen University and Research, 6708 WD Wageningen, the Netherlands
| | - C Kamphuis
- Wageningen Livestock Research, Wageningen University and Research, 6708 WD Wageningen, the Netherlands
| | - I Adriaens
- Wageningen Livestock Research, Wageningen University and Research, 6708 WD Wageningen, the Netherlands; Department of Biosystems Engineering, Livestock Technology, KU Leuven, 3001 Leuven, Belgium; Department of Mathematical Modelling and Data Analysis, BioVisM, Ghent University, B-9000 Ghent, Belgium
| |
Collapse
|
2
|
Gait Analysis in Walking and Trotting Dairy Cows on Different Flooring Types with Novel Mobile Pressure Sensors and Inertial Sensors. Animals (Basel) 2022; 12:ani12182457. [PMID: 36139317 PMCID: PMC9495103 DOI: 10.3390/ani12182457] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/11/2022] [Revised: 09/13/2022] [Accepted: 09/14/2022] [Indexed: 11/30/2022] Open
Abstract
Mechanical overburdening is a major risk factor that provokes non-infectious claw diseases. Moreover, lameness-causing lesions often remain undetected and untreated. Therefore, prevention of claw tissue overburdening is of interest, especially by analyzing harmful effects within dairy cows’ housing environment. However, objective “on-cow” methods for bovine gait analysis are underdeveloped. The purpose of the study was to apply an innovative mobile pressure sensor system attached at the claws to perform pedobarometric gait analysis. A further goal was the supplementation with accelerative data, generated simultaneously by use of two inertial measurement units (IMUs), attached at metatarsal level. IMU data were analyzed with an automatic step detection algorithm. Gait analysis was performed in ten dairy cows, walking and trotting on concrete flooring and rubber mats. In addition to the basic applicability of the sensor systems and with the aid of the automatic step detection algorithm for gait analysis in cows, we were able to determine the impact of the gait and flooring type on kinematic and kinetic parameters. For pressure sensor output, concrete was associated with significantly (p < 0.001) higher maximum and average pressure values and a significantly smaller contact area, compared to rubber mats. In contrast to walking, trotting led to a significantly higher force, especially under the medial claw. Further, IMU-derived parameters were significantly influenced by the gait. The described sensor systems are useful tools for detailed gait analysis in dairy cows. They allow the investigation of factors which may affect claw health negatively.
Collapse
|
3
|
Silva SR, Araujo JP, Guedes C, Silva F, Almeida M, Cerqueira JL. Precision Technologies to Address Dairy Cattle Welfare: Focus on Lameness, Mastitis and Body Condition. Animals (Basel) 2021; 11:2253. [PMID: 34438712 PMCID: PMC8388461 DOI: 10.3390/ani11082253] [Citation(s) in RCA: 7] [Impact Index Per Article: 2.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/29/2021] [Revised: 07/28/2021] [Accepted: 07/28/2021] [Indexed: 01/28/2023] Open
Abstract
Specific animal-based indicators that can be used to predict animal welfare have been the core of protocols for assessing the welfare of farm animals, such as those produced by the Welfare Quality project. At the same time, the contribution of technological tools for the accurate and real-time assessment of farm animal welfare is also evident. The solutions based on technological tools fit into the precision livestock farming (PLF) concept, which has improved productivity, economic sustainability, and animal welfare in dairy farms. PLF has been adopted recently; nevertheless, the need for technological support on farms is getting more and more attention and has translated into significant scientific contributions in various fields of the dairy industry, but with an emphasis on the health and welfare of the cows. This review aims to present the recent advances of PLF in dairy cow welfare, particularly in the assessment of lameness, mastitis, and body condition, which are among the most relevant animal-based indications for the welfare of cows. Finally, a discussion is presented on the possibility of integrating the information obtained by PLF into a welfare assessment framework.
Collapse
Affiliation(s)
- Severiano R. Silva
- Veterinary and Animal Research Centre (CECAV), Associate Laboratory of Animal and Veterinary Sciences (AL4AnimalS), University of Trás-os-Montes e Alto Douro, Quinta de Prados, 5000-801 Vila Real, Portugal; (S.R.S.); (C.G.); (F.S.); (M.A.)
| | - José P. Araujo
- Escola Superior Agrária do Instituto Politécnico de Viana do Castelo, Rua D. Mendo Afonso, 147, Refóios do Lima, 4990-706 Ponte de Lima, Portugal;
- Mountain Research Centre (CIMO), Instituto Politécnico de Viana do Castelo, Rua D. Mendo Afonso, 147, Refóios do Lima, 4990-706 Ponte de Lima, Portugal
| | - Cristina Guedes
- Veterinary and Animal Research Centre (CECAV), Associate Laboratory of Animal and Veterinary Sciences (AL4AnimalS), University of Trás-os-Montes e Alto Douro, Quinta de Prados, 5000-801 Vila Real, Portugal; (S.R.S.); (C.G.); (F.S.); (M.A.)
| | - Flávio Silva
- Veterinary and Animal Research Centre (CECAV), Associate Laboratory of Animal and Veterinary Sciences (AL4AnimalS), University of Trás-os-Montes e Alto Douro, Quinta de Prados, 5000-801 Vila Real, Portugal; (S.R.S.); (C.G.); (F.S.); (M.A.)
| | - Mariana Almeida
- Veterinary and Animal Research Centre (CECAV), Associate Laboratory of Animal and Veterinary Sciences (AL4AnimalS), University of Trás-os-Montes e Alto Douro, Quinta de Prados, 5000-801 Vila Real, Portugal; (S.R.S.); (C.G.); (F.S.); (M.A.)
| | - Joaquim L. Cerqueira
- Veterinary and Animal Research Centre (CECAV), Associate Laboratory of Animal and Veterinary Sciences (AL4AnimalS), University of Trás-os-Montes e Alto Douro, Quinta de Prados, 5000-801 Vila Real, Portugal; (S.R.S.); (C.G.); (F.S.); (M.A.)
- Escola Superior Agrária do Instituto Politécnico de Viana do Castelo, Rua D. Mendo Afonso, 147, Refóios do Lima, 4990-706 Ponte de Lima, Portugal;
| |
Collapse
|
4
|
Kang X, Zhang XD, Liu G. A Review: Development of Computer Vision-Based Lameness Detection for Dairy Cows and Discussion of the Practical Applications. SENSORS 2021; 21:s21030753. [PMID: 33499381 PMCID: PMC7866151 DOI: 10.3390/s21030753] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.7] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 11/16/2020] [Revised: 01/18/2021] [Accepted: 01/20/2021] [Indexed: 01/29/2023]
Abstract
The computer vision technique has been rapidly adopted in cow lameness detection research due to its noncontact characteristic and moderate price. This paper attempted to summarize the research progress of computer vision in the detection of lameness. Computer vision lameness detection systems are not popular on farms, and the accuracy and applicability still need to be improved. This paper discusses the problems and development prospects of this technique from three aspects: detection methods, verification methods and application implementation. The paper aims to provide the reader with a summary of the literature and the latest advances in the field of computer vision detection of lameness in dairy cows.
Collapse
Affiliation(s)
- Xi Kang
- Key Lab of Modern Precision Agriculture System Integration Research, Ministry of Education of China, China Agricultural University, Beijing 100083, China; (X.K.); (X.D.Z.)
- Key Lab of Agricultural Information Acquisition Technology, Ministry of Agricultural of China, China Agricultural University, Beijing 100083, China
| | - Xu Dong Zhang
- Key Lab of Modern Precision Agriculture System Integration Research, Ministry of Education of China, China Agricultural University, Beijing 100083, China; (X.K.); (X.D.Z.)
- Key Lab of Agricultural Information Acquisition Technology, Ministry of Agricultural of China, China Agricultural University, Beijing 100083, China
| | - Gang Liu
- Key Lab of Modern Precision Agriculture System Integration Research, Ministry of Education of China, China Agricultural University, Beijing 100083, China; (X.K.); (X.D.Z.)
- Key Lab of Agricultural Information Acquisition Technology, Ministry of Agricultural of China, China Agricultural University, Beijing 100083, China
- Correspondence: ; Tel.: +86-010-62736741
| |
Collapse
|
5
|
Mokhtarnazif S, Smid AMC, Weary DM, Mohamadnia A, von Keyserlingk MAG. Short communication: Motivation to walk affects gait attributes. J Dairy Sci 2020; 103:9481-9487. [PMID: 32747100 DOI: 10.3168/jds.2019-18060] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/14/2019] [Accepted: 05/12/2020] [Indexed: 11/19/2022]
Abstract
Lameness is a major welfare problem in the dairy industry. Environmental factors, such as flooring surface, as well as cow-level factors, such as udder fill, can influence gait. The aim of the current study was to test whether motivation to walk affects gait attributes and whether this effect differs between lame and sound cows. We trained cows to walk down an alley for a food reward and assessed walking speed, stride length, head bob, and back arch of cows previously identified as either lame (n = 7) or sound (n = 10). Cows were assessed when they walked toward a food reward and toward no reward. Cows walked faster and had longer stride length and less variation in head bob when approaching the reward; these effects were similar in both sound and lame cows. We concluded that motivation to walk affects several gait attributes of dairy cows.
Collapse
Affiliation(s)
- Shabnaz Mokhtarnazif
- Animal Welfare Program, University of British Columbia, 2357 Mall, Vancouver, BC, Canada, V6T 1Z4; Department of Clinical Science, Faculty of Veterinary Medicine, Ferdowsi University of Mashhad, Azadi square, Mashhad, Iran. 9177948974
| | - Anne-Marieke C Smid
- Animal Welfare Program, University of British Columbia, 2357 Mall, Vancouver, BC, Canada, V6T 1Z4
| | - Daniel M Weary
- Animal Welfare Program, University of British Columbia, 2357 Mall, Vancouver, BC, Canada, V6T 1Z4
| | - Ahmadreza Mohamadnia
- Department of Clinical Science, Faculty of Veterinary Medicine, Ferdowsi University of Mashhad, Azadi square, Mashhad, Iran. 9177948974
| | | |
Collapse
|
6
|
O'Leary N, Byrne D, O'Connor A, Shalloo L. Invited review: Cattle lameness detection with accelerometers. J Dairy Sci 2020; 103:3895-3911. [DOI: 10.3168/jds.2019-17123] [Citation(s) in RCA: 31] [Impact Index Per Article: 7.8] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/18/2019] [Accepted: 12/30/2019] [Indexed: 01/08/2023]
|
7
|
Zillner JC, Tücking N, Plattes S, Heggemann T, Büscher W. Using walking speed for lameness detection in lactating dairy cows. Livest Sci 2018. [DOI: 10.1016/j.livsci.2018.10.005] [Citation(s) in RCA: 5] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/28/2022]
|
8
|
Van De Gucht T, Saeys W, Van Meensel J, Van Nuffel A, Vangeyte J, Lauwers L. Farm-specific economic value of automatic lameness detection systems in dairy cattle: From concepts to operational simulations. J Dairy Sci 2017; 101:637-648. [PMID: 29102143 DOI: 10.3168/jds.2017-12867] [Citation(s) in RCA: 12] [Impact Index Per Article: 1.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/13/2017] [Accepted: 09/01/2017] [Indexed: 11/19/2022]
Abstract
Although prototypes of automatic lameness detection systems for dairy cattle exist, information about their economic value is lacking. In this paper, a conceptual and operational framework for simulating the farm-specific economic value of automatic lameness detection systems was developed and tested on 4 system types: walkover pressure plates, walkover pressure mats, camera systems, and accelerometers. The conceptual framework maps essential factors that determine economic value (e.g., lameness prevalence, incidence and duration, lameness costs, detection performance, and their relationships). The operational simulation model links treatment costs and avoided losses with detection results and farm-specific information, such as herd size and lameness status. Results show that detection performance, herd size, discount rate, and system lifespan have a large influence on economic value. In addition, lameness prevalence influences the economic value, stressing the importance of an adequate prior estimation of the on-farm prevalence. The simulations provide first estimates for the upper limits for purchase prices of automatic detection systems. The framework allowed for identification of knowledge gaps obstructing more accurate economic value estimation. These include insights in cost reductions due to early detection and treatment, and links between specific lameness causes and their related losses. Because this model provides insight in the trade-offs between automatic detection systems' performance and investment price, it is a valuable tool to guide future research and developments.
Collapse
Affiliation(s)
- Tim Van De Gucht
- Technology and Food Sciences Unit, Institute for Agricultural and Fisheries Research, Burg. van Gansberghelaan 115, 9820 Merelbeke, Belgium; Department of Biosystems, Katholieke Universiteit Leuven, Kasteelpark Arenberg 30, 3001 Leuven, Belgium
| | - Wouter Saeys
- Department of Biosystems, Katholieke Universiteit Leuven, Kasteelpark Arenberg 30, 3001 Leuven, Belgium
| | - Jef Van Meensel
- Social Sciences Unit, Institute for Agricultural and Fisheries Research, Burg. van Gansberghelaan 115, 9820 Merelbeke, Belgium
| | - Annelies Van Nuffel
- Technology and Food Sciences Unit, Institute for Agricultural and Fisheries Research, Burg. van Gansberghelaan 115, 9820 Merelbeke, Belgium.
| | - Jurgen Vangeyte
- Technology and Food Sciences Unit, Institute for Agricultural and Fisheries Research, Burg. van Gansberghelaan 115, 9820 Merelbeke, Belgium
| | - Ludwig Lauwers
- Social Sciences Unit, Institute for Agricultural and Fisheries Research, Burg. van Gansberghelaan 115, 9820 Merelbeke, Belgium; Ghent University, Department of Agricultural Economics, Coupure Links 653, 9000 Ghent, Belgium
| |
Collapse
|
9
|
Van De Gucht T, Van Weyenberg S, Van Nuffel A, Lauwers L, Vangeyte J, Saeys W. Supporting the Development and Adoption of Automatic Lameness Detection Systems in Dairy Cattle: Effect of System Cost and Performance on Potential Market Shares. Animals (Basel) 2017; 7:ani7100077. [PMID: 28991188 PMCID: PMC5664036 DOI: 10.3390/ani7100077] [Citation(s) in RCA: 7] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/05/2017] [Revised: 09/19/2017] [Accepted: 09/28/2017] [Indexed: 11/23/2022] Open
Abstract
Simple Summary Most prototypes of systems to automatically detect lameness in dairy cattle are still not available on the market. Estimating their potential adoption rate could support developers in defining development goals towards commercially viable and well-adopted systems. We simulated the potential market shares of such prototypes to assess the effect of altering the system cost and detection performance on the potential adoption rate. We found that system cost and lameness detection performance indeed substantially influence the potential adoption rate. In order for farmers to prefer automatic detection over current visual detection, the usefulness that farmers attach to a system with specific characteristics should be higher than that of visual detection. As such, we concluded that low system costs and high detection performances are required before automatic lameness detection systems become applicable in practice. Abstract Most automatic lameness detection system prototypes have not yet been commercialized, and are hence not yet adopted in practice. Therefore, the objective of this study was to simulate the effect of detection performance (percentage missed lame cows and percentage false alarms) and system cost on the potential market share of three automatic lameness detection systems relative to visual detection: a system attached to the cow, a walkover system, and a camera system. Simulations were done using a utility model derived from survey responses obtained from dairy farmers in Flanders, Belgium. Overall, systems attached to the cow had the largest market potential, but were still not competitive with visual detection. Increasing the detection performance or lowering the system cost led to higher market shares for automatic systems at the expense of visual detection. The willingness to pay for extra performance was €2.57 per % less missed lame cows, €1.65 per % less false alerts, and €12.7 for lame leg indication, respectively. The presented results could be exploited by system designers to determine the effect of adjustments to the technology on a system’s potential adoption rate.
Collapse
Affiliation(s)
- Tim Van De Gucht
- Institute for Agricultural and Fisheries Research-ILVO, Technology and Food Sciences Unit, Burg, van Gansberghelaan 115, 9820 Merelbeke, Belgium.
- KU Leuven Department of Biosystems, MeBioS, Kasteelpark Arenberg 30 Box 2456, 3001 Leuven, Belgium.
| | - Stephanie Van Weyenberg
- Institute for Agricultural and Fisheries Research-ILVO, Technology and Food Sciences Unit, Burg, van Gansberghelaan 115, 9820 Merelbeke, Belgium.
| | - Annelies Van Nuffel
- Institute for Agricultural and Fisheries Research-ILVO, Technology and Food Sciences Unit, Burg, van Gansberghelaan 115, 9820 Merelbeke, Belgium.
| | - Ludwig Lauwers
- Institute for Agricultural and Fisheries Research-ILVO, Social Sciences Unit, Burg, van Gansberghelaan 115, 9820 Merelbeke, Belgium.
- Department of Agricultural Economics, Faculty of Bio-Engineering, Ghent University, Coupure Links 653, 9000 Gent, Belgium.
| | - Jürgen Vangeyte
- Institute for Agricultural and Fisheries Research-ILVO, Technology and Food Sciences Unit, Burg, van Gansberghelaan 115, 9820 Merelbeke, Belgium.
| | - Wouter Saeys
- KU Leuven Department of Biosystems, MeBioS, Kasteelpark Arenberg 30 Box 2456, 3001 Leuven, Belgium.
| |
Collapse
|