Next Article in Journal
Precision Motion Control of a Linear Permanent Magnet Synchronous Machine Based on Linear Optical-Ruler Sensor and Hall Sensor
Next Article in Special Issue
A Deep Learning Approach on Building Detection from Unmanned Aerial Vehicle-Based Images in Riverbank Monitoring
Previous Article in Journal
LTCC Packaged Ring Oscillator Based Sensor for Evaluation of Cell Proliferation
Previous Article in Special Issue
Hyperspectral Image Classification with Capsule Network Using Limited Training Samples
Open AccessArticle

Segmentation and Multi-Scale Convolutional Neural Network-Based Classification of Airborne Laser Scanner Data

1
State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan 430010, China
2
Shenzhen Power Supply Co., Ltd., No. 2018 Cuizhu Road., Shenzhen 430079, China
*
Author to whom correspondence should be addressed.
Sensors 2018, 18(10), 3347; https://doi.org/10.3390/s18103347
Received: 18 August 2018 / Revised: 24 September 2018 / Accepted: 1 October 2018 / Published: 7 October 2018
(This article belongs to the Special Issue Deep Learning Remote Sensing Data)
The classification of point clouds is a basic task in airborne laser scanning (ALS) point cloud processing. It is quite a challenge when facing complex observed scenes and irregular point distributions. In order to reduce the computational burden of the point-based classification method and improve the classification accuracy, we present a segmentation and multi-scale convolutional neural network-based classification method. Firstly, a three-step region-growing segmentation method was proposed to reduce both under-segmentation and over-segmentation. Then, a feature image generation method was used to transform the 3D neighborhood features of a point into a 2D image. Finally, feature images were treated as the input of a multi-scale convolutional neural network for training and testing tasks. In order to obtain performance comparisons with existing approaches, we evaluated our framework using the International Society for Photogrammetry and Remote Sensing Working Groups II/4 (ISPRS WG II/4) 3D labeling benchmark tests. The experiment result, which achieved 84.9% overall accuracy and 69.2% of average F1 scores, has a satisfactory performance over all participating approaches analyzed. View Full-Text
Keywords: region growing segmentation; multi-scale convolutional neural network; ALS point clouds; semantic 3D labeling; feature image region growing segmentation; multi-scale convolutional neural network; ALS point clouds; semantic 3D labeling; feature image
Show Figures

Graphical abstract

MDPI and ACS Style

Yang, Z.; Tan, B.; Pei, H.; Jiang, W. Segmentation and Multi-Scale Convolutional Neural Network-Based Classification of Airborne Laser Scanner Data. Sensors 2018, 18, 3347.

Show more citation formats Show less citations formats
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Search more from Scilit
 
Search
Back to TopTop