A Method for Segmentation and Recognition of Mature Citrus and Branches-Leaves Based on Regional Features

  • Sa Liu
  • Changhui YangEmail author
  • Youcheng Hu
  • Lin Huang
  • Longye Xiong
Conference paper
Part of the Communications in Computer and Information Science book series (CCIS, volume 875)


A novel method based on the regional feature of mature citrus is proposed for segmentation and recognition in order to achieve the segmentation and identification of mature citrus and branches-leaves in Harvesting robots. Feature mapping table is used to reduce the dimension of the feature vector. The ROI (region of interest) size of the target object is determined by the size of picking up the working space of the manipulator, the binocular camera field and the citrus. According to the scores, the consistency of the primary ROI is sorted to a greater degree, and the ROI with the largest score is chosen as the optimal area. The experimental results show that the accuracy of recognition is 94% and the time required for single image segmentation is 0.24 s. This method is better than many existing methods.


Mature citrus Feature vector SVM segmentation Region of interest 


  1. 1.
    Yang, S.: China’s citrus industry status and development trend. Agric. Eng. Technol. (Agro-Ind.) 04, 13–17 (2014)Google Scholar
  2. 2.
    Yang, L.: Research on target recognition and positioning technology of citrus picking robot based on binocular vision. Chongqing University of Science and Technology (2017)Google Scholar
  3. 3.
    Xiaojun, Z.: Mathematical picking robot positioning and obstacle detection. Jiangsu University (2009)Google Scholar
  4. 4.
    Jianrong, C., Xiaojun, Z., Feng, W., Qiang, L.: Obstacle identification technology of citrus picking robot. Trans. Agric. Mach. 40(11), 171–175 (2009)Google Scholar
  5. 5.
    Juntao, X., Xiangjun, Z., Hongjun, W., et al.: Realization of ripe litchi under different light conditions based on Retinex image enhancement. J. Agric. Eng. China 29(12), 170–178 (2013)Google Scholar
  6. 6.
    Hongxing, P., Xiangjun, Z., Lijuan, C., et al.: Fast identification of field lychee multicolor targets based on double Otsu algorithm. J. Agric. Mechanization 45(4), 61–68 (2014)Google Scholar
  7. 7.
    Okamoto, H., Lee, W.S.: Green citrus detection using hyperspectral imaging. Comput. Electron. Agric. 66(2), 201–208 (2009)CrossRefGoogle Scholar
  8. 8.
    Stajnko, D., Lakota, M., Hočevar, M.: Estimation of number and diameter of apple fruits in an orchard during the growing season by thermal imaging. Comput. Electron. Agric. 42(1), 31–42 (2004)CrossRefGoogle Scholar
  9. 9.
    Amatya, S., et al.: Detection of cherry tree branches with full foliage in planar architecture for automated sweet-cherry harvesting. Biosyst. Eng. 146(Suppl. C), 3–15 (2016)CrossRefGoogle Scholar
  10. 10.
    Bac, C.W., Hemming, J., van Henten, E.J.: Robust pixel-based classification of obstacles for robotic harvesting of sweet-pepper. Comput. Electron. Agric. 96, 148–162 (2013)CrossRefGoogle Scholar
  11. 11.
    Juntao, X., Xiangjun, Z., Lijuan, C., Aixia, G.: Identification of ripe litchi in natural environment based on machine vision. Acta Autom. Mach. 42(09), 162–166 (2011)Google Scholar
  12. 12.
    Shun, F.: Design and research of end effector of citrus picking robot. Chongqing University of Science and Technology (2017)Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2018

Authors and Affiliations

  • Sa Liu
    • 1
  • Changhui Yang
    • 1
    • 2
    Email author
  • Youcheng Hu
    • 1
  • Lin Huang
    • 1
  • Longye Xiong
    • 1
  1. 1.College of Mechanical EngineeringChongqing University of TechnologyChongqingChina
  2. 2.College of Mechanical EngineeringXi’an Jiaotong UniversityXi’anChina

Personalised recommendations