Pivotal techniques of cotton-harvest robots were summarized, including image segmentation, features generation, artificial classifiers and performances evaluation. Solutions based on machine vision and pattern recognition were analyzed to distinguish ripe from under-ripe/over-ripe cottons, and rank cottons according to government grading standards.


Cotton Harvest robot Classifier 


  1. Ling Wang, Changying Ji. Technical analysis and expectation for cotton harvesting based on agricultural robot. Cotton Science. Vol. 18(2). pp. 124-128, 2006.Google Scholar
  2. Pentland A. “Fractal based decomposition of natural scenes,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 6(6), pp. 661-674, 1984.CrossRefPubMedGoogle Scholar
  3. Searcy S.W., Beck A.D. Real time assessment of cotton plant height. Proceedings of Fifth International Conference on Precision Agriculture (CD), July 16 19, 2000. Bloomington, MN, USA.Google Scholar
  4. Wuncsh P. Laine A. “Wavelet descriptors for multiresolution recognition of handwritten characters,” Pattern Recognition, Vol. 28(8), pp. 1237-1249, 1995.Google Scholar

Copyright information

© IFIP International Federation for Information Processing 2008

Authors and Affiliations

  • Ling Wang
    • 1
  • Changying Ji
    • 1
  1. 1.Nanjing Agricultural UniversityChina

Personalised recommendations