Semi-Supervised Segmentation Framework Based on Spot-Divergence Supervoxelization of Multi-Sensor Fusion Data for Autonomous Forest Machine Applications

In this paper, a novel semi-supervised segmentation framework based on a spot-divergence supervoxelization of multi-sensor fusion data is proposed for autonomous forest machine (AFMs) applications in complex environments. Given the multi-sensor measuring system, our framework addresses three successive steps: firstly, the relationship of multi-sensor coordinates is jointly calibrated to form higher-dimensional fusion data. Then, spot-divergence supervoxels representing the size-change property are given to produce feature vectors covering comprehensive information of multi-sensors at a time. Finally, the Gaussian density peak clustering is proposed to segment supervoxels into sematic objects in the semi-supervised way, which non-requires parameters preset in manual. It is demonstrated that the proposed framework achieves a balancing act both for supervoxel generation and sematic segmentation. Comparative experiments show that the well performance of segmenting various objects in terms of segmentation accuracy (F-score up to 95.6%) and operation time, which would improve intelligent capability of AFMs.

[1]  Hui Liu,et al.  An Integrated GNSS/INS/LiDAR-SLAM Positioning Method for Highly Accurate Forest Stem Mapping , 2016, Remote. Sens..

[2]  Fan Wu,et al.  Rapid Localization and Extraction of Street Light Poles in Mobile LiDAR Point Clouds: A Supervoxel-Based Approach , 2017, IEEE Transactions on Intelligent Transportation Systems.

[3]  Jahng-Hyon Park,et al.  Weighted-graph-based supervoxel segmentation of 3D point clouds in complex urban environment , 2015 .

[4]  Akihiro Sugimoto,et al.  Fast 3D point cloud segmentation using supervoxels with geometry and color for 3D scene understanding , 2017, 2017 IEEE International Conference on Multimedia and Expo (ICME).

[5]  Pascal Fua,et al.  SLIC Superpixels Compared to State-of-the-Art Superpixel Methods , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  Zuyuan Wang,et al.  Towards automated forest mapping , 2017 .

[7]  Lei Yan,et al.  New Calibration Method of Two-Dimensional Laser Scanner and Camera Based on LM-BP Neural Network , 2016 .

[8]  Nathaniel Anderson,et al.  Use of real-time GNSS-RF data to characterize the swing movements of forestry equipment , 2017 .

[9]  Ahmad Kamal Aijazi,et al.  Segmentation Based Classification of 3D Urban Point Clouds: A Super-Voxel Based Approach with Evaluation , 2013, Remote. Sens..

[10]  Johannes Heinzel,et al.  Detecting Tree Stems from Volumetric TLS Data in Forest Environments with Rich Understory , 2016, Remote. Sens..

[11]  Jürgen Schmidhuber,et al.  A Machine Learning Approach to Visual Perception of Forest Trails for Mobile Robots , 2016, IEEE Robotics and Automation Letters.

[12]  Yusheng Xu,et al.  Voxel-based segmentation of 3D point clouds from construction sites using a probabilistic connectivity model , 2018, Pattern Recognit. Lett..

[13]  Manolis I. A. Lourakis.  A Brief Description of the Levenberg-Marquardt Algorithm Implemented by levmar , 2005 .

[14]  J. Trochta,et al.  3D Forest: An application for descriptions of three-dimensional forest structures using terrestrial LiDAR , 2017, PloS one.

[15]  Zhengyou Zhang,et al.  A Flexible New Technique for Camera Calibration , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[16]  Luc Van Gool,et al.  SEEDS: Superpixels Extracted Via Energy-Driven Sampling , 2012, International Journal of Computer Vision.

[17]  Rama Rao Nidamanuri,et al.  Segmentation based building detection approach from LiDAR point cloud , 2017 .

[18]  Michela Bertolotto,et al.  Octree-based region growing for point cloud segmentation , 2015 .

[19]  Michael Weinmann,et al.  A Classification-Segmentation Framework for the Detection of Individual Trees in Dense MMS Point Cloud Data Acquired in Urban Areas , 2017, Remote. Sens..

[20]  Florentin Wörgötter,et al.  Voxel Cloud Connectivity Segmentation - Supervoxels for Point Clouds , 2013, 2013 IEEE Conference on Computer Vision and Pattern Recognition.

[21]  Changming Sun,et al.  Refinement of LiDAR point clouds using a super voxel based approach , 2018, ISPRS Journal of Photogrammetry and Remote Sensing.

[22]  Hamid Hamraz,et al.  Forest understory trees can be segmented accurately within sufficiently dense airborne laser scanning point clouds , 2017, Scientific Reports.

[23]  Thomas Hellström,et al.  Autonomous Forest Vehicles: Historic, envisioned, and state-of-the-art , 2009 .

[24]  Sean Hughes,et al.  Clustering by Fast Search and Find of Density Peaks , 2016 .

[25]  Zhong Chen,et al.  Supervoxel Segmentation with Voxel-Related Gaussian Mixture Model , 2018, Sensors.

[26]  Andrea Pezzuolo,et al.  Determination of forest road surface roughness by Kinect depth imaging , 2017 .

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

[28]  Shuliang Wang,et al.  Clustering by Fast Search and Find of Density Peaks with Data Field , 2016 .

[29]  Yanming Chen,et al.  Segmentation of Individual Trees From TLS and MLS Data , 2017, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[30]  Bastian Leibe,et al.  Exploring Spatial Context for 3D Semantic Segmentation of Point Clouds , 2017, 2017 IEEE International Conference on Computer Vision Workshops (ICCVW).

[31]  Radu Bogdan Rusu,et al.  3D is here: Point Cloud Library (PCL) , 2011, 2011 IEEE International Conference on Robotics and Automation.

[32]  Yang Liu,et al.  Automatic Forest Mapping at Individual Tree Levels from Terrestrial Laser Scanning Point Clouds with a Hierarchical Minimum Cut Method , 2016, Remote. Sens..

[33]  Ning Ye,et al.  A supervoxel approach to the segmentation of individual trees from LiDAR point clouds , 2018 .

[34]  Lei Yan,et al.  New Hybrid Algorithms for Estimating Tree Stem Diameters at Breast Height Using a Two Dimensional Terrestrial Laser Scanner , 2015, Sensors.

[35]  Alessandro Laio,et al.  Clustering by fast search and find of density peaks , 2014, Science.

[36]  Jae-Young Sim,et al.  Supervoxel-based saliency detection for large-scale colored 3D point clouds , 2016, 2016 IEEE International Conference on Image Processing (ICIP).