Hidden feature extraction for unstructured agricultural environment based on supervised kernel locally linear embedding modeling

Abstract

An online hidden feature extraction algorithm is proposed for unknown and unstructured agricultural environments based on a supervised kernel locally linear embedding (SKLLE) algorithm. Firstly, an online obtaining method for scene training samples is given to obtain original feature data. Secondly, Bayesian estimation of the a posteriori probability of a cluster center is performed. Thirdly, nonlinear kernel mapping function construction is employed to map the original feature data to hyper-high-dimensional kernel space. Fourthly, the automatic determination of hidden feature dimensions is performed using a local manifold learning algorithm. Then, a low-level manifold computation in hidden space is completed. Finally, long-range scene perception is realized using a 1-NN classifier. Experiments are conducted to show the effectiveness and the influence of parameter selection for the proposed algorithm. The kernel principal component analysis (KPCA), locally linear embedding (LLE), and supervised locally linear embedding (SLLE) methods are compared under the same experimental unstructured agricultural environment scene. Test results show that the proposed algorithm is more suitable for unstructured agricultural environments than other existing methods, and that the computational load is significantly reduced.

Keywords

Long-range scene perception / Hidden feature extraction / Agricultural vehicle / Unstructured agricultural environment

References

[1] Ahuja S, Iles P, Waslander SL. Three-dimensional scan registration using curvelet features in planetary environments. J Field Robot, 2016, 33(2): 243-259.

[2] Bietresato M, Carabin G, Vidoni R, et al. Evaluation of a LiDAR-based 3D-stereoscopic vision system for crop-monitoring applications. Comput Electron Agric, 2016, 124: 1-13.

[3] de Ridder D, Kouropteva O, Okun O et al (2003) Supervised locally linear embedding. Artificial neural networks and neural information processing—ICANN/ICONIP 2003. Springer, Berlin, Heidelberg, pp 333–341

[4] Fang HC, Ong SK, Nee AYC. Novel AR-based interface for human-robot interaction and visualization. Adv Manuf, 2014, 2(4): 275-288.

[5] Fjeldaas S. Computer vision supported by 3D geometric modelling. Adv Manuf, 2014, 2(1): 22-31.

[6] Hadsell R, Sermanet P, Ben J, et al. Learning long-range vision for autonomous off-road driving. J Field Robot, 2009, 26(2): 120-144.

[7] Jackel LD, Krotkov E, Perschbacher M, et al. The DARPA LAGR program: goals, challenges, methodology, and phase I results. J Field Robot, 2006, 23(11–12): 945-973.

[8] Jamali A, Rahman AA, Boguslawski P, et al. An automated 3D modeling of topological indoor navigation network. GeoJournal, 2017, 82(1): 157-170.

[9] Jensen MAF, Bochtis D, Sørensen CG, et al. In-field and inter-field path planning for agricultural transport units. Comput Ind Eng, 2012, 63(4): 1054-1061.

[10] Kraus T, Ferreau HJ, Kayacan E, et al. Moving horizon estimation and nonlinear model predictive control for autonomous agricultural vehicles. Comput Electron Agric, 2013, 98: 25-33.

[11] Liu H, Taniguchi T, Tanaka Y, et al. Visualization of driving behavior based on hidden feature extraction by using deep learning. IEEE Trans Intell Transp Syst, 2017, 18(9): 2477-2489.

[12] Lunga D, Prasad S, Crawford MM, et al. Manifold-learning-based feature extraction for classification of hyperspectral data: a review of advances in manifold learning. IEEE Signal Process Mag, 2014, 31(1): 55-66.

[13] Matveev AS, Hoy M, Katupitiya J, et al. Nonlinear sliding mode control of an unmanned agricultural tractor in the presence of sliding and control saturation. Rob Auton Syst, 2013, 61(9): 973-987.

[14] Mehta SS, Burks TF, Dixon WE. Vision-based localization of a wheeled mobile robot for greenhouse applications: a daisy-chaining approach. Comput Electron Agric, 2008, 63(1): 28-37.

[15] Milella A, Reina G. 3D reconstruction and classification of natural environments by an autonomous vehicle using multi-baseline stereo. Intel Serv Robot, 2014, 7(2): 79-92.

[16] Mousazadeh H. A technical review on navigation systems of agricultural autonomous off-road vehicles. J Terrramech, 2013, 50(3): 211-232.

[17] Niesterowicz J, Stepinski TF. Regionalization of multi-categorical landscapes using machine vision methods. Appl Geogr, 2013, 45: 250-258.

[18] Noguchi N, Will J, Reid J, et al. Development of a master–slave robot system for farm operations. Comput Electron Agric, 2004, 44(1): 1-19.

[19] Procopio MJ, Mulligan J, Grudic G. Learning terrain segmentation with classifier ensembles for autonomous robot navigation in unstructured environments. J Field Robot, 2009, 26(2): 145-175.

[20] Roweis ST, Saul LK. Nonlinear dimensionality reduction by locally linear embedding. Science, 2000, 290(5500): 2323-2326.

[21] Schölkopf B, Smola A, Müller KR (1997) Kernel principal component analysis. In: International conference on artificial neural networks. Springer, pp 583–588

[22] Schölkopf B, Smola A, Müller KR. Nonlinear component analysis as a kernel eigenvalue problem. Neural Comput, 1998, 10(5): 1299-1319.

[23] Si YS, Liu G, Feng J. Location of apples in trees using stereoscopic vision. Comput Electron Agric, 2015, 112: 68-74.

[24] Zhang W, Chen Q, Zhang W, et al. Long-range terrain perception using convolutional neural networks. Neurocomputing, 2018, 275: 781-787.

Funding

National Natural Science Foundation of China(51375293)

key program for Basic Research of the Science and Technology Commission of Shanghai Municipality(12JC1404100)