Eliminating Useless Object Detectors Evolved in Multiple-Objective Genetic Programming

  • Aaron Scoble
  • Mark Johnston
  • Mengjie Zhang
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7106)

Abstract

Object detection is the task of correctly identifying and locating objects of interest within a larger image. An ideal object detector would maximise the number of correctly located objects and minimise the number of false-alarms. Previous work, following the traditional multiple-objective paradigm of finding Pareto-optimal tradeoffs between these objectives, suffers from an abundance of useless detectors that either detect nothing (but with no false-alarms) or mark every pixel as an object (perfect detection performance with but a very large number of false-alarms); these are very often Pareto-optimal and hence inadvertently rewarded. We propose and compare a number of improvements to eliminate useless detectors during evolution. The most successful improvements are generally more inefficient than the benchmark MOGP approach due to the often vast numbers of additional crossover and mutation operations required, but as a result the archive populations generally include a much higher number of Pareto-fronts.

Keywords

Pareto Front False Alarm Rate Object Detector Mutation Operation Child Population 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Deb, K., Pratap, A., Agarwal, S.: A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE Transactions on Evolutionary Computation 6(2), 182–197 (2002)CrossRefGoogle Scholar
  2. 2.
    Knowles, J.: A summary-attainment-surface plotting method for visualizing the performance of stochastic multiobjective optimizers. In: Proceedings of the 5th International Conference on Intelligent Systems Design and Applications, pp. 552–557 (2005)Google Scholar
  3. 3.
    Liddle, T., Johnston, M., Zhang, M.: Multi-objective genetic programming for object detection. In: Proceedings of 2010 IEEE Congress on Evolutionary Computation, pp. 3345–3352 (2010)Google Scholar
  4. 4.
    Zhang, M.: Improving object detection performance with genetic programming. International Journal on Artificial Intelligence Tools 16(5), 849 (2007)CrossRefGoogle Scholar
  5. 5.
    Zhang, M., Andreae, P., Pritchard, M.: Pixel Statistics and False Alarm Area in Genetic Programming for Object Detection. In: Raidl, G.R., Cagnoni, S., Cardalda, J.J.R., Corne, D.W., Gottlieb, J., Guillot, A., Hart, E., Johnson, C.G., Marchiori, E., Meyer, J.-A., Middendorf, M. (eds.) EvoIASP 2003, EvoWorkshops 2003, EvoSTIM 2003, EvoROB/EvoRobot 2003, EvoCOP 2003, EvoBIO 2003, and EvoMUSART 2003. LNCS, vol. 2611, pp. 455–466. Springer, Heidelberg (2003)CrossRefGoogle Scholar
  6. 6.
    Zitzler, E., Deb, K., Thiele, L.: Comparison of multiobjective evolutionary algorithms: Empirical results. Evolutionary Computation 8(2), 173–195 (2000)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Aaron Scoble
    • 1
  • Mark Johnston
    • 1
  • Mengjie Zhang
    • 2
  1. 1.School of Mathematics, Statistics and Operations ResearchVictoria University of WellingtonWellingtonNew Zealand
  2. 2.School of Engineering and Computer ScienceVictoria University of WellingtonWellingtonNew Zealand

Personalised recommendations