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

Abstract

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.

Deep‐Stacking Network Approach by Multisource Data Mining for Hazardous Risk Identification in IoT‐Based Intelligent Food Management Systems

“…The proposed methods in this paper can combine other identification schemes for studying new modeling and prediction of dynamical systems [ 40 – 42 ] and can be applied to other fields [ 43 – 46 ] such as signal modeling, tracking, and control systems. On the contrary, our network has high requirements for data and does not perform well for unstructured data.…”

Utilization of Image, LiDAR and Gamma-Ray Information to Improve Environmental Sustainability of Cut-to-Length Wood Harvesting Operations in Peatlands: A Management Systems Perspective

“…The most likely reason for the use of different indicators in the literature is the difficulty of measuring the maximum depth of soil rutting by LiDAR in deep ruts which may be full of water. Terrestrial LiDAR also has several other problems when measuring the maximum depth of soil rutting, which also negatively affect the accuracy when calculating the soil rutting percentage [67].…”

“…Regarding the development of planning systems, the use of terrestrial LiDAR to produce ground rutting data has been tested, e.g., by using harvester data for the needs of authorities at the forest and stand levels [8]. Although standard techniques have been developed for laser technology to identify objects, the separation of individual objects and backgrounds from forest environments raises problems that have not yet been satisfactorily resolved and are the main challenge in planning systems [66,67]. Based on the results from an experimental field study, a forest machine-mounted LiDAR sensor can be used to collect rut depth data at a spatial resolution of 10-20 m with careful treatment of raw point cloud data [67].…”

“…Although standard techniques have been developed for laser technology to identify objects, the separation of individual objects and backgrounds from forest environments raises problems that have not yet been satisfactorily resolved and are the main challenge in planning systems [66,67]. Based on the results from an experimental field study, a forest machine-mounted LiDAR sensor can be used to collect rut depth data at a spatial resolution of 10-20 m with careful treatment of raw point cloud data [67]. Studies have shown that the spatial data produced could be used to develop nationwide forest trafficability maps of wood harvesting for administration [68].…”

Streaming Data Fusion for the Internet of Things

“…Sensor fusion is helpful for improving certain functionalities and model accuracy in various domains , i.e., in positioning and navigation [15,16,17], in activity recognition [18,19], in system monitoring and fault diagnosis [20,21,22,23,24,25], in transport [26], in health care [27] and in others.…”