- Animal Breeding & Genomics (4)
- Animal Breeding and Genetics (4)
- Animal Breeding and Genomics (4)
- LR - Animal Breeding & Genomics (4)
- WIAS (4)
- PE&RC (3)
- VLAG (3)
- Farm Technology (2)
- Farm Technology Group (2)
- GTB Teelt & Gewasfysiologie A (2)
- Physical Chemistry and Colloid Science (2)
- Physical Chemistry and Soft Matter (2)
- GTB Tuinbouw Technologie (1)
- Resource Ecology (1)
- Vakgroep Sociologie (1)
- WPR GTB AgroFoodRobotics (1)
- WUR GTB Tuinbouw Technologie (1)
- Robert B. Fisher (1)
- Peter Biber (1)
- Michael Blaich (1)
- J. Brouwer (1)
- Thomas Brox (1)
- Krassimir D. Danov (1)
- Simeon D. Stoyanov (2)
- K.D. Danov (1)
- Bart Ducro (2)
- B. Ducro (1)
- B.J. Ducro (1)
- Roel F. Veerkamp (2)
- Krastanka G. Marinova (1)
- Eddie G. Pelan (1)
- Theo Gevers (1)
- Egiel Hanenberg (2)
- Jochen Hemming (2)
- Eldert Henten Van (1)
- Eldert Henten van (1)
- Joris IJsselmuiden (1)
- Dejan Kaljaca (1)
- P.A. Kralchevsky (1)
- Lydia M. Dimitrova (2)
- G.H. Mehl (1)
- Angelo Mencarelli (1)
- A.L. Mok (1)
- Mariana P. Boneva (1)
- V.N. Paunov (1)
- Nicolai Petkov (4)
- N. Petkov (1)
- J. Petkov (1)
- J.T. Petkov (1)
- Marc Pollefeys (1)
- Q. Qin (1)
- G.M. Radulova (1)
- P.A. Rubini (1)
- M. Rutkevicius (1)
- Elka S. Basheva (1)
- Torsten Sattler (1)
- S.D. Stoyanov (2)
- Nicola Strisciuglio (1)
- Jordan T. Petkov (1)
- Toon Tielen (1)
- Bart Tuijl Van (1)
- Radim Tylecek (1)
- Plamen V. Petkov (1)
- R.F. Veerkamp (2)
- Pieter Visser De (1)
Estimation of Muscle Scores of Live Pigs Using a Kinect Camera
Alsahaf, Ahmad ; Azzopardi, George ; Ducro, Bart ; Hanenberg, Egiel ; Veerkamp, Roel F. ; Petkov, Nicolai - \ 2019
IEEE Access 7 (2019). - ISSN 2169-3536 - p. 52238 - 52245.
Computer vision - machine learning - precision farming - RGB-D imaging
The muscle grading of livestock is a primary component of valuation in the meat industry. In pigs, the muscularity of a live animal is traditionally estimated by visual and tactile inspection from an experienced assessor. In addition to being a time-consuming process, scoring of this kind suffers from inconsistencies inherent to the subjectivity of human assessment. On the other hand, accurate, computer-driven methods for carcass composition estimation, such as magnetic resonance imaging (MRI) and computed tomography scans (CT-scans), are expensive and cumbersome to both the animals and their handlers. In this paper, we propose a method that is fast, inexpensive, and non-invasive for estimating the muscularity of live pigs, using RGB-D computer vision and machine learning. We used morphological features extracted from the depth images of pigs to train a classifier that estimates the muscle scores that are likely to be given by a human assessor. The depth images were obtained from a Kinect v1 camera which was placed over an aisle through which the pigs passed freely. The data came from 3246 pigs, each having 20 depth images, and a muscle score from 1 to 7 (reduced later to 5 scores) assigned by an experienced assessor. The classification based on morphological features of the pig's body shape-using a gradient boosted classifier-resulted in a mean absolute error of 0.65 in tenfold cross-validation. Notably, the majority of the errors corresponded to pigs being classified as having muscle scores adjacent to the groundtruth labels given by the assessor. According to the end users of this application, the proposed approach could be used to replace expert assessors at the farm.
Predicting slaughter weight in pigs with regression tree ensembles
Alsahaf, A. ; Azzopardi, G. ; Ducro, B. ; Veerkamp, R.F. ; Petkov, N. - \ 2018
In: Applications of Intelligent Systems - Proceedings of the 1st International APPIS Conference 2018, APPIS 2018. - IOS Press (Frontiers in Artificial Intelligence and Applications ) - ISBN 9781614999287 - p. 1 - 9.
Animal production - Ensemble learning - Gradient boosting - Pigs - Random forest - XGBoost
Domestic pigs vary in the age at which they reach slaughter weight even under the controlled conditions of modern pig farming. Early and accurate estimates of when a pig will reach slaughter weight can lead to logistic efficiency in farms. In this study, we compare four methods in predicting the age at which a pig reaches slaughter weight (120 kg). Namely, we compare the following regression tree-based ensemble methods: random forest (RF), extremely randomized trees (ET), gradient boosted machines (GBM), and XGBoost. Data from 32979 pigs is used, comprising a combination of phenotypic features and estimated breeding values (EBV). We found that the boosting ensemble methods, GBM and XGBoost, achieve lower prediction errors than the parallel ensembles methods, RF and ET. On the other hand, RF and ET have fewer parameters to tune, and perform adequately well with default parameter settings.
Trimbot cutting tools and manipulator
Hemming, Jochen ; Tuijl, Bart Van; Tielen, Toon ; Kaljaca, Dejan ; IJsselmuiden, Joris ; Henten, Eldert Van; Mencarelli, Angelo ; Visser, Pieter De - \ 2018
In: Applications of Intelligent Systems - Proceedings of the 1st International APPIS Conference 2018, APPIS 2018. - IOS Press (Frontiers in Artificial Intelligence and Applications ) - ISBN 9781614999287 - p. 89 - 93.
Agriculture - End-effector - Path planning - Pruning - Robot - Trimming
This article describes the tasks and first results of the work package "Manipulator and Control" of the EU project Trimbot2020. This project develops a mobile robot for outdoor hedge, rose and bush trimming. The Kinova Jaco 2 robotic arm was selected as manipulator. Two different types of robotic end-effectors have been developed. The tool for trimming topiaries uses two custom designed circular contra-rotating blades. The tool for single stem cutting is based on a commercial electrical pruner. The arm and the tools can all be controlled by using the Robot Operating System (ROS). The motion planning algorithm of the arm for the bush trimming action is divided into the planning setup module, the coverage planning module and the trajectory planning module. The path planning is modelled as a traveling salesman problem. In the first phase of the project the trimming control is performed open loop. A positioning genetic algorithm was developed that minimizes the needed number of vehicle poses for one target object. In the next phase of the project a vision feedback mechanism will be implemented.
Assigning pigs to uniform target weight groups using machine learning
Alsahaf, Ahmad ; Azzopardi, George ; Ducro, B.J. ; Veerkamp, R.F. ; Petkov, Nicolai - \ 2018
In: World Congress on Genetics Applied to Livestock Production. - - p. 112 - 112.
|Trimbot2020 : An outdoor robot for automatic gardening
Strisciuglio, Nicola ; Tylecek, Radim ; Blaich, Michael ; Petkov, Nicolai ; Biber, Peter ; Hemming, Jochen ; Henten, Eldert van; Sattler, Torsten ; Pollefeys, Marc ; Gevers, Theo ; Brox, Thomas ; Fisher, Robert B. - \ 2018
In: 50th International Symposium on Robotics, ISR 2018. - VDE Verlag GmbH - ISBN 9781510870314 - 1 p.
Robots are increasingly present in modern industry and also in everyday life. Their applications range from health-related situations, for assistance to elderly people or in surgical operations, to automatic and driver-less vehicles (on wheels or flying) or for driving assistance. Recently, an interest towards robotics applied in agriculture and gardening has arisen, with applications to automatic seeding and cropping or to plant disease control, etc. Autonomous lawn mowers are succesful market applications of gardening robotics. In this paper, we present a novel robot that is developed within the TrimBot2020 project, funded by the EU H2020 program. The project aims at prototyping the first outdoor robot for automatic bush trimming and rose pruning.
Prediction of slaughter age in pigs and assessment of the predictive value of phenotypic and genetic information using random forest
Alsahaf, Ahmad ; Azzopardi, George ; Ducro, Bart ; Hanenberg, Egiel ; Veerkamp, Roel F. ; Petkov, Nicolai - \ 2018
Journal of Animal Science 96 (2018)12. - ISSN 0021-8812 - p. 4935 - 4943.
Breeding - Grouping strategies - Machine learning - Pigs - Random forest - Regression
The weight of a pig and the rate of its growth are key elements in pig production. In particular, predicting future growth is extremely useful, since it can help in determining feed costs, pen space requirements, and the age at which a pig reaches a desired slaughter weight. However, making these predictions is challenging, due to the natural variation in how individual pigs grow, and the different causes of this variation. In this paper, we used machine learning, namely random forest (RF) regression, for predicting the age at which the slaughter weight of 120 kg is reached. Additionally, we used the variable importance score from RF to quantify the importance of different types of input data for that prediction. Data of 32,979 purebred Large White pigs were provided by Topigs Norsvin, consisting of phenotypic data, estimated breeding values (EBVs), along with pedigree and pedigree-genetic relationships. Moreover, we presented a 2-step data reduction procedure, based on random projections (RPs) and principal component analysis (PCA), to extract features from the pedigree and genetic similarity matrices for use as inputs in the prediction models. Our results showed that relevant phenotypic features were the most effective in predicting the output (age at 120 kg), explaining approximately 62% of its variance (i.e., R2 = 0.62). Estimated breeding value, pedigree, or pedigree-genetic features interchangeably explain 2% of additional variance when added to the phenotypic features, while explaining, respectively, 38%, 39%, and 34% of the variance when used separately.
Production and characterization of stable foams with fine bubbles from solutions of hydrophobin HFBII and its mixtures with other proteins
Dimitrova, Lydia M. ; Petkov, Plamen V. ; Kralchevsky, Peter A. ; Stoyanov, Simeon D. ; Pelan, Eddie G. - \ 2017
Colloids and Surfaces. A: Physicochemical and Engineering Aspects 521 (2017). - ISSN 0927-7757 - p. 92 - 104.
Hydrophobins are proteins that are excellent foam stabilizers. We investigated the effects of pH and addition of other proteins on the foaminess, bubble size, and stability of foams from aqueous solutions of the protein HFBII hydrophobin. The produced stable foams have bubbles of radii smaller than 40 μm that obey the lognormal distribution. The overrun of most foams is in the range from 5 to 8, which indicates a good foaminess. The foam longevity is characterized by the time dependences of the foam volume and weight. A combined quantitative criterion for stability, the degree of foam conservation, is proposed. The produced foams are stable for at least 12–17 days. The high foam stability can be explained with the formation of dense hydrophobin adsorption layers, which are impermeable to gas transfer and block the Ostwald ripening (foam disproportionation). In addition, the population of small bubbles formed in the HFBII solutions blocks the drainage of water through the Plateau borders in the foam. The variation of pH does not essentially affect the foaminess and foam stability. The addition of “regular” proteins, such as beta-lactoglobulin, ovalbumin and bovine serum albumin, to the HFBII solutions does not deteriorate the quality and stability of the produced foams up to 94% weight fraction of the added protein. The results and conclusions from the present study could be useful for the applications of hydrophobins as foam stabilizers.
Limited coalescence and Ostwald ripening in emulsions stabilized by hydrophobin HFBII and milk proteins
Dimitrova, Lydia M. ; Boneva, Mariana P. ; Danov, Krassimir D. ; Kralchevsky, Peter A. ; Basheva, Elka S. ; Marinova, Krastanka G. ; Petkov, Jordan T. ; Stoyanov, Simeon D. - \ 2016
Colloids and Surfaces. A: Physicochemical and Engineering Aspects 509 (2016). - ISSN 0927-7757 - p. 521 - 538.
Drop size distribution - Emulsification - Emulsion stability - HFBII hydrophobin - Ostwald ripening
Hydrophobins are proteins isolated from filamentous fungi, which are excellent foam stabilizers, unlike most of the proteins. In the present study, we demonstrate that hydrophobin HFBII can also serve as excellent emulsion stabilizer. The HFBII adsorption layers at the oil/water interface solidify similarly to those at the air/water interface. The thinning of aqueous films sandwiched between two oil phases ends with the formation of a 6 nm thick protein bilayer, just as in the case of foam films, which results in strong adhesive interactions between the emulsion drops. The drop-size distribution in hydrophobin stabilized oil-in-water emulsions is investigated at various protein concentrations and oil volume fractions. The data analysis indicates that the emulsification occurs in the Kolmogorov regime or in the regime of limited coalescence, depending on the experimental conditions. The emulsions with HFBII are very stable – no changes in the drop-size distributions are observed after storage for 50 days. However, these emulsions are unstable upon stirring, when they are subjected to the action of shear stresses. This instability can be removed by covering the drops with a second adsorption layer from a conventional protein, like β-lactoglobulin. The HFBII surface layer is able to suppress the Ostwald ripening in the case when the disperse phase is oil that exhibits a pronounced solubility in water. Hence, the hydrophobin can be used to stabilize microcapsules of fragrances, flavors, colors or preservatives due to its dense adsorption layers that block the transfer of oil molecules.
Shear rheology of hydrophobic adsorption layers at oil/water interfaces and data interpretation in terms of a viscoelastic thixotropic model
Radulova, G.M. ; Danov, K.D. ; Kralchevsky, P.A. ; Petkov, J.T. ; Stoyanov, S.D. - \ 2014
Soft Matter 10 (2014)31. - ISSN 1744-683X - p. 5777 - 5786.
oil-water interface - dependent relaxation-times - class-ii hydrophobin - protein hfbii - hexadecane/water interface - flexible proteins - bubble stability - beta-casein - surface - monolayers
Here, we investigate the surface shear rheology of class II HFBII hydrophobin layers at the oil/water interface. Experiments in two different dynamic regimes, at a fixed rate of strain and oscillations, have been carried out with a rotational rheometer. The rheological data obtained in both regimes comply with the same viscoelastic thixotropic model, which is used to determine the surface shear elasticity and viscosity, Esh and ¿sh. Their values for HFBII at oil/water interfaces are somewhat lower than those at the air/water interface. Moreover, Esh and ¿sh depend on the nature of oil, being smaller for hexadecane in comparison with soybean-oil. It is remarkable that Esh is independent of the rate of strain in the whole investigated range of shear rates. For oil/water interfaces, Esh and ¿sh determined for HFBII layers are considerably greater than for other proteins, like lysozyme and ß-casein. It is confirmed that the hydrophobin forms the most rigid surface layers among all investigated proteins not only for the air/water, but also for the oil/water interface. The wide applicability of the used viscoelastic thixotropic model is confirmed by analyzing data for adsorption layers at oil/water interfaces from lysozyme and ß-casein – both native and cross-linked by enzyme, as well as for films from asphaltene. This model turns out to be a versatile tool for determining the surface shear elasticity and viscosity, Esh and ¿sh, from experimental data for the surface storage and loss moduli, G' and G''.
Sound absorption properties of porous composites fabricated by a hydrogel templating technique
Rutkevicius, M. ; Mehl, G.H. ; Paunov, V.N. ; Qin, Q. ; Rubini, P.A. ; Stoyanov, S.D. ; Petkov, J. - \ 2013
Journal of Materials Research 28 (2013)17. - ISSN 0884-2914 - p. 2409 - 2414.
concrete - porosity
We have used a hydrogel templating technique followed by the subsequent evaporation of water present to fabricate porous cement and porous PDMS composites, and we have analyzed their sound absorption properties. All experiments were carried out with hydrogel slurries of broad bead size distributions. Porous PDMS and cement composites were produced with porosities of up to 80% and 70%, respectively. Scanning electron microscope analysis shows fibrous domains within the voids created by the hydrogel in the cement samples and open pore network in the PDMS composites of initial hydrogel content higher than 70 vol%. Sound absorption was improved with respect to control nonporous samples in all composites with porosities higher than 60 vol%, where an open pore structure was formed. The porous PDMS and porous cement produced by this method show better sound absorption at 200-400 Hz and 1200-1800 Hz frequency ranges when compared with the sound absorption in the intermediate frequencies range between 400 and 1000 Hz.
|Wetlands in the Sahel, taking Niger as an example: important for palearctic waterbirds, important for people
Brouwer, J. - \ 2003
In: Ferruginous Duck: From Research to Conservation / Petkov, N, Hughes, B, Gallo-Ursi, U, Sofia, Bulgarije, Slimbridge, UK : BridLife International (Conservation Series 6) - p. 130 - 137.
|On the attractiveness of working in a small firm
Mok, A.L. - \ 1985
In: Unattractive work / Kiuranovand, C., Petkov, K., - p. 33 - 45.