A novel stumpage detection method for forest harvesting based on multi-sensor fusion

This paper describes a novel stumpage detection method for forest harvesting based on 2D laser scanner and infrared thermal imager. First, the stumpage information captured by the two sensors is fused via image fusion and laser matching. Then, rich and accurate stumpage features can be extracted from the fused information. Next, an SVM classifier model is constructed by sample training according to the feature data. Finally, in contrast to SVM with default parameters, three different optimization algorithms are proposed to optimize SVM parameters. Based on 400 stumpage samples, the test on the proposed algorithms is conducted. The results show that the SVM with GA has the best detection rate of 96.7 %. Finally, to verify the performance of the method in this paper, some comparative tests were carried out and the experimental results proved the feasibility and accuracy of the proposed method.

[1]  C. Hopkinson,et al.  Assessing forest metrics with a ground-based scanning lidar , 2004 .

[2]  N. Pfeifer,et al.  AUTOMATIC RECONSTRUCTION OF SINGLE TREES FROM TERRESTRIAL LASER SCANNER DATA , 2004 .

[3]  Jason M. Kinser,et al.  Image Processing using Pulse-Coupled Neural Networks , 1998, Perspectives in Neural Computing.

[4]  Martial Hebert,et al.  Pedestrian Detection and Tracking Using Three-dimensional LADAR Data , 2010, Int. J. Robotics Res..

[5]  Barbara Caputo,et al.  Recognizing human actions: a local SVM approach , 2004, ICPR 2004.

[6]  Bo Wu,et al.  Fast Pedestrian Detection with Laser and Image Data Fusion , 2011, 2011 Sixth International Conference on Image and Graphics.

[7]  Arto Visala,et al.  Tree Measurement in Forest by 2D Laser Scanning , 2007, 2007 International Symposium on Computational Intelligence in Robotics and Automation.

[8]  Varun Gupta,et al.  Occupancy grid mapping using artificial neural networks , 2010, 2010 International Conference on Industrial Electronics, Control and Robotics.

[9]  Shashi Bhushan,et al.  GAIT RECOGNITION OF HUMAN USING SVM AND BPNN CLASSIFIERS , 2014 .

[10]  Hailemariam Temesgen,et al.  A Comparison of the Spatial Linear Model to Nearest Neighbor (k-NN) Methods for Forestry Applications , 2013, PloS one.

[11]  N. Pfeifer,et al.  Three-dimensional reconstruction of stems for assessment of taper, sweep and lean based on laser scanning of standing trees , 2004 .

[12]  Paul A. Viola,et al.  Robust Real-Time Face Detection , 2001, International Journal of Computer Vision.

[13]  Arto Visala,et al.  Tree Measurement and Simultaneous Localization and Mapping System for Forest Harvesters , 2007, FSR.

[14]  Stefan Nilsson,et al.  Radar Detection of Moving Targets Behind Corners , 2011, IEEE Transactions on Geoscience and Remote Sensing.

[15]  Arto Visala,et al.  Simultaneous Localization and Mapping for Forest Harvesters , 2007, Proceedings 2007 IEEE International Conference on Robotics and Automation.

[16]  Pramod K. Varshney.  Multisensor data fusion , 1997 .

[17]  L.C. Bento,et al.  Multi-target detection and tracking with a laser scanner , 2004, IEEE Intelligent Vehicles Symposium, 2004.

[18]  Ma Zhen,et al.  Optical system design of visible camera for space target detection , 2010 .

[19]  Lynne E. Parker,et al.  Distributed Cooperative Outdoor Multirobot Localization and Mapping , 2004 .

[20]  Zhongliang Jing,et al.  Multi-focus image fusion using pulse coupled neural network , 2007, Pattern Recognit. Lett..

[21]  Minh N. Do,et al.  Contourlets: a directional multiresolution image representation , 2002, Proceedings. International Conference on Image Processing.