3차원 점군 데이터를 이용한 수정 구형 특징 표현기 개발 및 도시 구조물 분류를 위한 컨볼루션 신경망의 응용
Development of Modified Spherical Signature Descriptor Using 3D Point Cloud Data and Application to Convolutional Neural Network for Urban Structure Classification
2
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
This paper presents the novel observation model, called Modified Spherical Signature Descriptor(MSSD), capable of representing 2D image generated from 3D point cloud data. The Modified Spherical Signature Descriptor has a uniform mesh grid to accumulate the occupancy evidence caused by neighbor point cloud data. According to a kind of area such as wall, road, tree, car, and so on, the evidence pattern of 2D image looks so different each other. For the parameter learning of Convolutional Neural Network(CNN) layers, these 2D images were applied as the input layer. The Convolutional Neural Network, one of the deep learning methods and familiar with the image analysis, was utilized for the urban structure classification. The case study on CNN practice was introduced in detail in this paper. The simulation results shows that the classification accuracy of CNN with 2D images of the proposed MSSD was improved more than the traditional methods' one.
Abstract
This paper presents the novel observation model, called Modified Spherical Signature Descriptor(MSSD), capable of representing 2D image generated from 3D point cloud data. The Modified Spherical Signature Descriptor has a uniform mesh grid to accumulate the occupancy evidence caused by neighbor point cloud data. According to a kind of area such as wall, road, tree, car, and so on, the evidence pattern of 2D image looks so different each other. For the parameter learning of Convolutional Neural Network(CNN) layers, these 2D images were applied as the input layer. The Convolutional Neural Network, one of the deep learning methods and familiar with the image analysis, was utilized for the urban structure classification. The case study on CNN practice was introduced in detail in this paper. The simulation results shows that the classification accuracy of CNN with 2D images of the proposed MSSD was improved more than the traditional methods' one.
참고문헌 (21)
- Alex Krizhevsky, "ImageNet Classification with Deep Convolutional Neural Networks", NIPS 2012, pp.1~9 (2012)
- Yungeun Choe, "Urban structure classification using the 3D normal distribution transform for practical robot applications", Advanced Robotics, 27(5), pp.351~371 (2013)[DOI]
- Sungho Kim, "Object Recognition Using a Generalized Robust Invariant Feature and Gestalt's Law of Proximity and Similarity", Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPRW? 06), 41(2), pp.726~741 (2008)
- Daniel Maturana, "VoxNet: A 3D Convolutional Neural Network for Real-Time Object Recognition", IEEE, 2015, ~
- Z. Wu, "3D ShapeNets: A Deep Representation for Volumetric Shapes", Proceedings of 28th IEEE Conference on Computer Vision and Pattern Recognition (CVPR2015) (2015)
- 이세진, "이동 로봇의 확장 칼만 필터 기반 위치 추정을 위한 새로운 형태의 초음파 센서 배열", 한국기계기술학회지, 19(4), pp.518~524 (2017), 한국기계기술학회[DOI]
- "First Geodesic Dome: Planetarium in Jena 1922 incl. patent information Archived March 19, 2013, at the Wayback Machine", ~
- 이세진, "도시 구조물 분류를 위한 3차원 점 군의 구형 특징 표현과심층 신뢰 신경망 기반의 환경 형상 학습", 로봇학회 논문지, 11(3), pp.115~126 (2016), 한국로봇학회[DOI]
- S. Thrun, "Probabilistic Robotics" (2005), The MIT Press
- I.Bernard Cohen, "Proposition 75, Theorem 35 : The Principia: Mathematical Principles of Natural Philosophy. Preceded by A Guide to Newton's Principia", pp.956~ (1999), University of California Press
- Yann Lecun, "Gradient-Based Learning Applied to Document Recognition", Proc of the IEEE (1998)
- Dr. Roman V Belavkin, "Lecture 11: Feed-Forward Neural Networks, BIS3226", ~
- Vinod Nair, "Rectified linear units improve restricted Boltzmann machines", ICML (2010)
- Bishop, Christopher M., "Pattern Recognition and Machine Learning" (2006), Springer
- 안승욱, "무인 자동차의 2차원 레이저 거리 센서를 이용한 도시 환경에서의 빠른 주변 환경 인식 방법", 로봇학회 논문지, 7(2), pp.92~100 (2012), 한국로봇학회[DOI]
- Chulheen Bae, "Modified Spherical Signature Description based of 3D point Cloud for Improvement of Performance in DBN", Proceedings of the 12th Korea Robotics Society Annual Conference (2017)
- Christian Szegedy, "Going Deeper with Convolutions", Computer Vision and Pattern Recognition (2014)
- Kaiming He, "Deep Residual Learning for Image Recognition", Computer Vision and Pattern Recognition (2015)
- S. Ioffe, "Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift", International Conference on Machine Learning (2015)
- W. Ouyang, "DeepID-Net: Deformable Deep Convolutional Neural Networks for Object Detection", Computer Vision and Pattern Recognition (2015)