그림 1. 듀얼 카메라를 부착한 자율 주행 시스템 Fig. 1. Automotive driving system with multi-view camera
그림 3. 협각 카메라로부터 취득된 국내 도로 데이터 객체 별 종횡비(Width/Height) Fig. 3. Aspect ratios of domestic road data acquired from narrow-angle camera (Width/Height)
그림 2. 협각 카메라로 취득한 샘플 이미지 Fig. 2. Sample images acquired with a narrow camera
그림 4. 광각 카메라로부터 취득된 국내 도로 데이터 객체 별 종횡비(Width/Height) Fig. 4. Aspect ratios of domestic road data acquired from wide-angle camera (Width/Height)
그림 5. 협각 카메라 데이터 학습 후 검출 영상 Fig. 5. Detection image after learning the data from narrow-angle
그림 6. 협각, 광각 카메라 데이터 학습 후 검출 Fig. 6. Detection image after learning the data from wide, narrow-angle
그림 7. 광각 카메라 데이터 학습 후 검출 영상 Fig. 7. Detection image after learning the data from wide-angle
그림 8. 협각, 광각 카메라 데이터 학습 후 검출 영상 Fig. 8. Detection image after learning the data from wide, narrow-angle
표 1. 경량화 모델의 크기, 파라미터 수, 성능 비교 Table 1. Size, number of parameters and performance comparison of lightweight CNN models
표 2. 차량 전방 카메라로부터 취득한 객체 수 Table 2. Number of objects acquired from vehicle front camera
표 3. 제안된 방식의 성능분석 Table 3. Performance analysis of proposed method
References
- Sebastian Houben, Johannes Stallkamp. "Detection of traffic signs in real-world images: The German traffic sign detection benchmark" The 2013 International Joint Conference on Neural Networks, IJCNN.
- M. Cordts, M. Omran, "The Cityscapes Dataset for Semantic Urban Scene Understanding," in Proc. of the IEEE Conference on Computer Vision and Pattern Recognition CVPR, 2016
- A. Geiger, P. Lenz, and R. Urtasun, "Are we ready for autonomous driving? The KITTI vision benchmark suite" Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, pp.3354-3361, 2012.
- Son, K. Choi, N. Song, and D Lee, "Real-Time Dynamic Simulation of Vehicle and Occupant Using a Neural Network" Transactions of KSAE, Vol.10, No2, pp.132-140,2002.
- Wei Liu, Dragomir Anguelov, "SSD: Single Shot MultiBox Detector" In ECCV 2016.
- IIjoo Baek, Albert Davies, "Real-time Detection, Tracking, and Classification of Moving and Stationary Objects using Multiple Fisheye Images" arXiv:1803.06077.
- Dumitru Erhan, Christian Szegedy, "Scalable Object Detection using Deep Neural Networks" In CVPR 2014.
- Andrew G Howard, Menglong Zhu, "MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications" In CVPR 2017.
- Mark Sandler, Andrew Howard, "MobileNetV2: Inverted Residuals and Linear Bottlenecks", In CVPR 2018.
- Christian Szegedy, Vincent Vanhoucke, "Rethinking the Inception Architecture for Computer Vision" In CVPR 2015.
- Xiaozhi Chen, Kaustav Kundu, "3D Object Proposals for Accurate Object Class Detection" In NIPS 2015