Evolution of a local boundary detector for natural images via genetic programming and texture cues

ABSTRACT

Boundary detection constitutes a crucial step in many computer vision tasks. We present a learning approach for automatically constructing high-performance local boundary detectors for natural images via genetic programming (GP). Our GP system is unique in that it combines filter kernels that were inspired by models of processing in the early stages of the primate visual system, but makes no assumptions about what constitutes a boundary, thus avoiding the need to make ad hoc intuitive definitions. By testing our evolved boundary detectors on a highly challenging benchmark set of natural images with associated human-marked boundaries, we show performance that outperforms most existing approaches.

REFERENCES

• 1: The berkeley segmentation dataset and benchmark www.cs.berkeley.edu/projects/vision/bsds/.
• 2: Piotr Dollar , Zhuowen Tu , Serge Belongie, Supervised Learning of Edges and Object Boundaries, Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, p.1964-1971, June 17-22, 2006 [doi>10.1109/CVPR.2006.298]
• 3: I. Kadar, O. Ben-Shahar, and M. Sipper. Evolving boundary detectors for natural images via genetic programming. In Proc. of the 19th International Conference on Pattern Recognition, Tampa, Florida, 2008.
• 4: John R. Koza, Genetic Programming IV: Routine Human-Competitive Machine Intelligence, Kluwer Academic Publishers, Norwell, MA, 2003
• 5: David R. Martin , Charless C. Fowlkes , Jitendra Malik, Learning to Detect Natural Image Boundaries Using Local Brightness, Color, and Texture Cues, IEEE Transactions on Pattern Analysis and Machine Intelligence, v.26 n.5, p.530-549, May 2004 [doi>10.1109/TPAMI.2004.1273918]
• 6: D. Matin, C. Fowlkes, D. Tal, and J. Malik. A database of human segmented natural images and its applications to evaluating segmentations algorithms and measuring ecological statistics. In Proc. of international Conference on Computer Vision, 2001.

INDEX TERMS

Primary Classification:
  I. Computing Methodologies
  I.4 IMAGE PROCESSING AND COMPUTER VISION
      I.4.6 Segmentation

General Terms:
Algorithms, Design, Measurement, Performance

Keywords:
boundary detection, computer vision, evolutionary algorithms, machine learning