ABSTRACT
Recent adaptive image interpretation systems can reach optimal performance for a given domain via machine learning, without human intervention. The policies are learned over an extensive generic image processing operator library. One of the principal weaknesses of the method lies with the large size of such libraries, which can make the machine learning process intractable. We demonstrate how evolutionary algorithms can be used to reduce the size of the operator library, thereby speeding up learning of the policy while still keeping human experts out of the development loop. Experiments in a challenging domain of forestry image interpretation exhibited a 95% reduction in the average time required to interpret an image, while maintaining the image interpretation accuracy of the full library.
- D. Culvenor. Tida: an alg. for the delineation of tree crowns in high spatial resolution remotely sensed imagery. Computers & Geosciences, 28(1):33--44, 2002. Google ScholarDigital Library
- B. Draper, J. Bins, and K. Baek. ADORE: adaptive object recognition. Videre, (4):86--99, 2000.Google Scholar
- B. A. Draper. From knowledge bases to Markov models to PCA. In Proceedings of Workshop on Computer Vision System Control Architectures, Austria, 2003.Google Scholar
- C. Emmanouilidis, A. Hunter, J. MacIntyre, and C. Cox. Multiple Criteria Genetic Algorithms for Feature Selection in Neurofuzzy Modeling. In In Proceedings of IJCNN, Washington, D.C., 1999.Google Scholar
- F. Gougeon and D. Leckie. Forest information extraction from high spatial resolution images using an individual tree crown approach. Technical report, Pacific Forestry Centre, 2003.Google Scholar
- J. Jarmulak and S. Craw. Genetic algorithms for feature selection and weighting. In In Proceedings of the IJCAI'99 workshop on Automating the Construction of Case Based Reasoners, pages 28--33, 1999.Google Scholar
- Y. Jin, M. Olhofer, and B. Sendhoff. Managing approximate models in evolutionary aerodynamic design optimization. In Proceedings of the 2001 Congress on Evolutionary Computation CEC2001, pages 592--599, COEX, World Trade Center, 159 Samseong-dong, Gangnam-gu, Seoul, Korea, 27-30 May 2001. IEEE Press.Google Scholar
- K. Kira and L. Rendell. The feature selection problem: Traditional methods and a new algorithm. In Proceedings of the Tenth National Conference on Artificial Intelligence (AAAI-92), pages 129--134, 1992.Google Scholar
- R. Kohavi and G. H. John. Wrappers for feature subset selection. Artificial Intelligence, 97(1-2):273--324, 1997. Google ScholarDigital Library
- G. Lee. Automated action set selection in Markov decision processes. Master's thesis, Department of Computing Science, University of Alberta, 2004.Google Scholar
- I. Levner and V. Bulitko. Machine learning for adaptive image interpretation. In Proceedings of the 16th Innovative Applications of Artificial Intelligence'04 conference, 2004. Google ScholarDigital Library
- L. S. Oliveira, N. Benahmed, R. Sabourin, F. Bortolozzi, and C. Y. Suen. Feature subset selection using gas for handwritten digit recognition. In In Proceedings of the 14th Brazilian Symposium on Computer Graphics and Image Processing, pages 362--369, Florianópolis-Brazil, 2001. IEEE Computer Society. Google ScholarDigital Library
- R. Pollock. A model-based approach to automatically locating tree crowns in high spatial resolution images. In J. Desachy, editor, Image and Signal Processing for Remote Sensing, 1994.Google Scholar
- P. Pudil, J. Novovicova, and J. Kittler. Floating search methods in feature-selection. PRL, 15(11):1119--1125, November 1994. Google ScholarDigital Library
- Z. Sun, X. Yuan, G. Bebis, and S. Louis. Neural-network-based gender classification using genetic eigen-feature extraction. In In Proceedings of IEEE International Joint Conference on Neural Networks, Honoloulu, Hawaii, 2002.Google Scholar
- R. Sutton and A. Barto. Reinforcement Learning: An Intro. MIT Press, 1998. Google ScholarDigital Library
- H. Vafaie and K. D. Jong. Genetic algorithms as a tool for feature selection in machine learning. In In Proceeding of the 4th International Conference on Tools with Artificial Intelligence, pages 200--204, Arlington, VA, 1992.Google ScholarCross Ref
- H. Vafaie and K. D. Jong. Robust feature selection algorithms. In In Proceedings of the Fifth Conference on Tools for Artificial Intelligence, pages 356--363, Boston, MA, 1993. IEEE Computer Society Press.Google ScholarCross Ref
- P. Viola and M. Jones. Fast and robust classification using asymmetric adaboost and a detector cascade. In T. G. Dietterich, S. Becker, and Z. Ghahramani, editors, Advances in Neural Information Processing Systems 14, MA, 2002. MIT Press.Google Scholar
- C. Watkins. Learning from Delayed Rewards. PhD thesis, King's College, University of Cambridge, UK, 1989.Google Scholar
Index Terms
- GAMM: genetic algorithms with meta-models for vision
Recommendations
Genetic algorithms for action set selection across domains: a demonstration
GECCO '06: Proceedings of the 8th annual conference on Genetic and evolutionary computationAction set selection in Markov Decision Processes (MDPs) is an area of research that has received little attention. On the other hand, the set of actions available to an MDP agent can have a significant impact on the ability of the agent to gain optimal ...
A Knowledge-Intensive Genetic Algorithm for Supervised Learning
Special issue on genetic algorithmsSupervised learning in attribute-based spaces is one of the most popular machine learning problems studied and, consequently, has attracted considerable attention of the genetic algorithm community. The full-memory approach developed here uses the same ...
Comparison of Machine Learned Image Interpretation Systems in the Domain of Forestry
WACV-MOTION '05: Proceedings of the Seventh IEEE Workshops on Application of Computer Vision (WACV/MOTION'05) - Volume 1 - Volume 01Automated image interpretation is an important task with numerous applications. Until recently, designing such systems required extensive subject matter and computer vision expertise resulting in poor cross-domain portability and expensive maintenance. ...
Comments