Object Detection in Cluttered Environment Using 3D Map

  • Deepesh Jain
  • Renuka Ramachandran
  • Anuhya Vunnam
  • P. Vignesh
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 324)

Abstract

Autonomous mobile robot must act intelligently without external control by definition and require fundamental capabilities such as the awareness of its environment and of its location within the environment. These two problems are known, respectively, as mapping and localization. The ability to detect and identify mobile and fixed obstacles also plays an important role for achieving robots autonomy. The project is concerned with the problem of designing and implementing a robot system to recognize objects in cluttered environment using a 3D map generated by the system using efficient algorithms. For building dense 3D maps of the environment and to recognize objects, use RGB-D camera which accurately identifies objects as they take into consideration the shape and three-dimensional characteristics of the object.

Keywords

Kinect SURF RANSAC 

References

  1. 1.
    M. Magnabosco et al., Cross-spectral visual simultaneous localization and mapping (SLAM) with sensor handover. Robot. Auton. Syst. 61(2), 195–208 (2012)Google Scholar
  2. 2.
    P. Bongale et al., Implementation of 3D object recognition and tracking, in International Conference on Recent Advances in Computer and Software Systems (2012), pp. 77–79Google Scholar
  3. 3.
    J.J. Hernández-López et al., Detecting objects using color and depth segmentation, in IberoAmerican conference on electronics engineering and computer science (Elsevier, Amsterdam, 2012), pp. 196–204Google Scholar
  4. 4.
    K. Sanko et al, Practical 3D object detection using category and instance-level appearance model, in IEEE/RSJ International Conference on Intelligent Robots and Systems (2011), pp. 793–800Google Scholar
  5. 5.
    P. Henry et al., RGB-D Mapping, using Kinect-style depth cameras for dense 3-D modeling of an indoor environment. Int. J. Rob. Res., 1–17 (2012)Google Scholar
  6. 6.
    A. Janoch et al., A category-level 3D object dataset: putting the kinect to work, in IEEE International Conference on Computer Vision Workshops (2011), pp. 1168–1174Google Scholar

Copyright information

© Springer India 2015

Authors and Affiliations

  • Deepesh Jain
    • 1
  • Renuka Ramachandran
    • 1
  • Anuhya Vunnam
    • 1
  • P. Vignesh
    • 1
  1. 1.Department of Computer Science and EngineeringAmrita School of EngineeringCoimbatoreIndia

Personalised recommendations