다시점 영상 집합을 활용한 선체 블록 분류를 위한 CNN 모델 성능 비교 연구
Comparison Study of the Performance of CNN Models with Multi-view Image Set on the Classification of Ship Hull Blocks
1
조회
0
다운로드
3
인용
0
참고문헌
키워드
초록
It is important to identify the location of ship hull blocks with exact block identification number when scheduling the shipbuilding process. The wrong information on the location and identification number of some hull block can cause low productivity by spending time to find where the exact hull block is. In order to solve this problem, it is necessary to equip the system to track the location of the blocks and to identify the identification numbers of the blocks automatically. There were a lot of researches of location tracking system for the hull blocks on the stockyard. However there has been no research to identify the hull blocks on the stockyard. This study compares the performance of 5 Convolutional Neural Network (CNN) models with multi-view image set on the classification of the hull blocks to identify the blocks on the stockyard. The CNN models are open algorithms of ImageNet Large-Scale Visual Recognition Competition (ILSVRC). Four scaled hull block models are used to acquire the images of ship hull blocks. Learning and transfer learning of the CNN models with original training data and augmented data of the original training data were done. 20 tests and predictions in consideration of five CNN models and four cases of training conditions are performed. In order to compare the classification performance of the CNN models, accuracy and average F1-Score from confusion matrix are adopted as the performance measures. As a result of the comparison, Resnet-152v2 model shows the highest accuracy and average F1-Score with full block prediction image set and with cropped block prediction image set.
Abstract
It is important to identify the location of ship hull blocks with exact block identification number when scheduling the shipbuilding process. The wrong information on the location and identification number of some hull block can cause low productivity by spending time to find where the exact hull block is. In order to solve this problem, it is necessary to equip the system to track the location of the blocks and to identify the identification numbers of the blocks automatically. There were a lot of researches of location tracking system for the hull blocks on the stockyard. However there has been no research to identify the hull blocks on the stockyard. This study compares the performance of 5 Convolutional Neural Network (CNN) models with multi-view image set on the classification of the hull blocks to identify the blocks on the stockyard. The CNN models are open algorithms of ImageNet Large-Scale Visual Recognition Competition (ILSVRC). Four scaled hull block models are used to acquire the images of ship hull blocks. Learning and transfer learning of the CNN models with original training data and augmented data of the original training data were done. 20 tests and predictions in consideration of five CNN models and four cases of training conditions are performed. In order to compare the classification performance of the CNN models, accuracy and average F1-Score from confusion matrix are adopted as the performance measures. As a result of the comparison, Resnet-152v2 model shows the highest accuracy and average F1-Score with full block prediction image set and with cropped block prediction image set.
참고문헌 (29)
- 변영현, "자동차 주행환경에서 보행자 분류를 위한 딥러닝 모델의 전이학습 및 성능비교", 한국정보기술학회논문지, 16(10), pp.83~92 (2018), 한국정보기술학회[DOI]
- Cho, D. Y., "Block and logistics simulation", Bulletin of the Society of Naval Architects of Korea, 48(4), pp.24~29 (2011)
- Donahue, J., "Decaf: A deep convolutional activation feature for generic visual recognition", International conference on machine learning, pp.647~655 (2014)
- Francois, C., "Deep learning with Python" (2018), Gilbut Publichin Co,. Ltd
- He, K., "Deep residual learning for image recognition", Proceedings of the IEEE conference on computer vision and pattern recognition, pp.770~778 (2016)
- Huang, G., "Densely connected convolutional networks", Proceedings of the IEEE conference on computer vision and pattern recognition, pp.4700~4708 (2017)
- Kang, J. H., "A study on mobile block logistics system for shipyard", Master, pp.2014
- 김병철, "합성곱 신경망을 이용한 3차원 스캐닝 점군 데이터의 플랜트 기자재 형상 인식", 대한기계학회논문집 A, 42(9), pp.863~869 (2018), 대한기계학회
- Kim, J.O., "Development of real time location measuring and logistics system for assembled block in shipbuilding", Korean Institute of Industrial Engineers, pp.834~839 (2009)
- Kim, M. S., "Determination of arrangement and take-out path in ship block stockyard considering available space and obstructive block", Society for Computational Design and Engineering, pp.433~438 (2013)
- Krizhevsky, A., "Imagenet classification with deep convolutional neural networks", Advances in neural information processing systems, pp.1097~1105 (2012)
- LeCun, Y., "Gradient-based learning applied to document recognition", Proceedings of the IEEE, 86(11), pp.2278~2324 (1998)
- Lee, Y. H., "Study on the positioning system for logistics of ship-block", Special Issue of the Society of Naval Architects of Korea, pp.68~75 (2008)
- Mun, S. H., "Real time block locating system for shipbuilding through GNSS and IMU fusion", Ph.D., pp.2019
- 남병욱, "의사결정트리 학습을 적용한 조선소 블록 적치 위치 선정에 관한 연구", 대한조선학회 논문집, 54(5), pp.421~429 (2017), 대한조선학회[DOI]
- Pan, S. J., "A survey on transfer learning", IEEE Transactions on Knowledge and Data Engineering, 22(10), pp.1345~1359 (2009)
- 박성욱, "전이학습에 방법에 따른 컨벌루션 신경망의영상 분류 성능 비교", 멀티미디어학회논문지, 21(12), pp.1387~1395 (2018), 한국멀티미디어학회[DOI]
- Saito, G., "Deep learning from scratch" (2017), Hanbit Media, lnc
- 이장현, "PDA 및 GPS를 이용한 옥외 작업장 블록 위치 추적 시스템 개발", 대한조선학회 논문집, 43(1), pp.803~811 (2006), 대한조선학회
- Simonyan, K., "Very deep convolutional networks for large-scale image recognition", conference paper at ICLR 2015 (2014)
- Srivastava, N., "Dropout : a simple way to prevent neural networks from overfitting", The journal of machine learning research, 15(1), pp.1929~1958 (2014)
- "ILSVRC" (2018)[URL]
- Szegedy, C., "Going deeper with convolutions", Proceedings of the IEEE conference on computer vision and pattern recognition, pp.1~9 (2014)
- Szegedy, C., "Rethinking the inception architecture for computer vision", Proceedings of the IEEE conference on computer vision and pattern recognition, pp.2818~2826 (2016)
- Yosinski, J., "How transferable are features in deep neural networks?", Advances in neural information processing systems, pp.3320~3328 (2014)
- Zeiler, M. D., "Visualizing and understanding convolutional networks", European conference on computer vision, pp.818~833 (2014)
- Zhou, B., "Learning deep features for discriminative localization", Proceedings of the IEEE conference on computer vision and pattern recognition, pp.2921~2929 (2016)
- Zhou, B., "Object detectors emerge in deep scene cnns", International Conference on Learning Representations (2014)
- Zoph, B., "Learning transferable architectures for scalable image recognition", Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition 2018, pp.8697~8710 (2018)