Abstract
Imagery analysis represents a significant aspect of maritime domain awareness; however, the amount of imagery is exceeding human capability to process. Unfortunately, the maritime domain presents unique challenges for machine learning to automate such analysis. Indeed, when object recognition algorithms observe real-world data, they face hurdles not present in experimental situations. Imagery from such domains suffers from degradation, have limited examples, and vary greatly in format. These limitations are present satellite imagery because of the associated constraints in expense and capability. To this end, the Hypercube-based NeuroEvolution of Augmenting Topologies approach is investigated in addressing some such challenges for classifying maritime vessels from satellite imagery. Results show that HyperNEAT learns features from such imagery that allows better classification than Principal Component Analysis (PCA). Furthermore, HyperNEAT enables a unique capability to scale image sizes through the indirect encoding.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Aaltonen, T., et al.: Measurement of the top quark mass with dilepton events selected using neuroevolution at CDF. Phys. Rev. Lett. 102, 152001–152008 (2009)
Bengio, Y., Courville, A., Vincent, P.: Representation learning: a review and new perspectives. IEEE Trans. Pattern Anal. Mach. Intell. 35(8), 1798–1828 (2013)
Buck, H., Sharghi, E., Guilas, C., Stastny, J., Morgart, W., Schalcosky, B., Pifko, K.: Enhanced ship detection from overhead imagery. In: SPIE Defense and Security Symposium, vol. 6945. International Society for Optics and Photonics (2008)
Cardamone, L., Loiacono, D., Lanzi, P.L.: On-line neuroevolution applied to the open racing car simulator. In: Proceedings of the IEEE CEC., Piscataway, NJ, USA. IEEE Press (2009)
Ciresan, D., Meier, U., Masci, J., Schmidhuber, J.: Multi-column deep neural network for traffic sign classification. Neural Netw. 32, 333–338 (2012)
Coleman, O.J.: Evolving neural networks for visual processing. Ph.D. thesis, The University of New South Wales, Kensingtion, Austrailia (2010)
D’Ambroiso, D., Stanley, K.O.: Evolving policy geometry for scalable multiagent learning. In: Proceedings of the 9th AAMAS, New York, NY, USA, p. 8. ACM Press (2010)
D’Ambrosio, D.B., Stanley, K.O.: Generative encoding for multiagent learning. In: Proceedings of GECCO 2008, New York, NY. ACM Press (2008)
Eggers, J.: Plenary talk. In: ICCRTS (2013)
Gauci, J., Stanley, K.O.: Generating large-scale neural networks through discovering geometric regularities. In: Proceedings of GECCO 2007, New York, NY, p. 8. ACM (2007)
Gauci, J., Stanley, K.O.: A case study on the critical role of geometric regularity in machine learning. In: Proceedings of the 23rd AAAI Conference, Menlo Park, CA. AAAI Press (2008)
Gauci, J., Stanley, K.O.: Autonomous evolution of topographic regularities in artificial neural networks. Neural Comput. 22(7), 1860–1898 (2010)
Hall, D., McCool, C., Dayoub, F., Sunderhauf, N., Upcroft, B.: Evaluation of features for leaf classification in challenging conditions. In: IEEE WACV, Waikola, HI, January 2015
Harguess, J., Parameswaran, S., Rainey, K., Stastny, J.: Vessel classification in overhead satellite imagery using learned dictionaries. In: SPIE Optical Engineering+ Applications, p. 84992F–84992F. International Society for Optics and Photonics (2012)
Harguess, J., Rainey, K.: Are face recognition methods useful for classifying ships? In: IEEE AIPR Workshop, 2011, pp. 1–7. IEEE (2011)
Hausknecht, M., Khandelwal, P., Miikkulainen, R., Stone, P.: Hyperneat-gpp: A hyperneat-based atari general game player. In: Proceedings of GECCO 2012, Philadelphia, Pennsylvania, p. 8. ACM Press, July 2012
Hinton, G.E., Osindero, S., Teh, Y.-W.: A fast learning algorithm for deep belief nets. Neural Comput. 18, 1527–1554 (2006)
Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classfication with deep convolutional neural networks. In: Advances in NIPS (2012)
Rainey, K., Parameswaran, S., Harguess, J.: Maritime vessel recognition in degraded satellite imagery. In: Proceedings of SPIE Automatic Target Recognition XXIV, vol. 9090. International Society for Optics and Photonics, June 2014
Rainey, K., Stastny, J.: Object recognition in ocean imagery using feature selection and compressive sensing. In: IEEE AIPR Workshop, pp. 1–6. IEEE (2011)
Rifai, S., Dauphin, Y.N., Vincent, P., Bengio, Y., Muller, X.: The manifold tangent classifer. In: Advances in NIPS (2011)
Stanley, K.O.: Compositional pattern producing networks: a novel abstraction of development. Genet. Program. Evolvable Mach. Spec. Issue Dev. Syst. 8(2), 131–162 (2007)
Stanley, K.O., Bryant, B.D., Miikkulainen, R.: Real-time neuroevolution in the NERO video game. IEEE Trans. Evol. Comput. Spec. Issue Evol. Comput. Games 9(6), 653–668 (2005)
Stanley, K.O., D’Ambrosio, D.B., Gauci, J.: A hypercube-based indirect encoding for evolving large-scale neural networks. Artif. Life 15, 185–212 (2009)
Stanley, K.O., Miikkulainen, R.: A taxonomy for artificial embryogeny. Artif. Life 9(2), 93 (2003)
Stanley, K.O., Miikkulainen, R.: Competitive coevolution through evolutionary complexification. J. Artif. Intell. Res. 21, 63 (2004)
Szerlip, P.A., Morse, G., Pugh, J.K., Stanley, K.O.: Unsupervised feature learning through divergent discriminative feature accumulation (2014). arXiv:1406.1833v2, abs/1406.1833
Taylor, M.E., Whiteson, S., Stone, P.: Comparing evolutionary and temporal difference methods in a reinforcement learning domain. In: Proceedings of GECCO 2006, New York, NY, pp. 1321–1328. ACM Press, July 2006
Turing, A.M.: The chemical basis of Morphogenesis. R. Soc. Lond. Philos. Trans. Ser. B 237, 37–72 (1952)
Verbancsics, P., Harguess, J.: Deep learning through generative and developmental system. In: Proceedings of the Genetic and Evolutionary Computation (GECCO 2014) companion, p. 103. ACM (2014)
Verbancsics, P., Harguess, J.: Image classification using generative neuroevolution for deep learning. In: IEEE Winter Conference on Applications of Computer Vision (WACV), Waikola, HI, January 2015
Verbancsics, P., Stanley, K.O.: Evolving static representations for task transfer. J. Mach. Learn. Res. 11, 1737–1769 (2010)
Verbancsics, P., Stanley, K.O.: Constraining connectivity to encourage modularity in hyperneat. In: Proceedings of GECCO 2011, New York, NY. ACM Press (2011)
Whiteson, S.: Improving reinforcement learning function approximators via neuroevolution. In Proceedings of the 4th AAMAS, New York, NY, USA, pp. 1386–1386. ACM (2005)
Whiteson, S., Whiteson, D.: Stochastic optimization for collision selection in high energy physics. In: IAAI 2007, Vancouver, British Columbia, Canada. AAAI Press, July 2007
Acknowledgments
This work was supported and funded by the SSC Pacific Naval Innovative Science and Engineering (NISE) Program.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
A Result Standard Deviations
A Result Standard Deviations
Rights and permissions
Copyright information
© 2015 Springer International Publishing Switzerland
About this paper
Cite this paper
Verbancsics, P., Harguess, J. (2015). Feature Learning HyperNEAT: Evolving Neural Networks to Extract Features for Classification of Maritime Satellite Imagery. In: Lones, M., Tyrrell, A., Smith, S., Fogel, G. (eds) Information Processing in Cells and Tissues. IPCAT 2015. Lecture Notes in Computer Science(), vol 9303. Springer, Cham. https://doi.org/10.1007/978-3-319-23108-2_18
Download citation
DOI: https://doi.org/10.1007/978-3-319-23108-2_18
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-23107-5
Online ISBN: 978-3-319-23108-2
eBook Packages: Computer ScienceComputer Science (R0)