세그멘테이션 기반 차선 인식 네트워크를 위한 적응형 키포인트 추출 알고리즘
Adaptive Key-point Extraction Algorithm for Segmentation-based Lane Detection Network
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
딥러닝 기반의 이미지 세그멘테이션은 차선 인식을 위해 널리 사용되는 접근 방식 중 하나로, 차선의 키포인트를 추출하기 위한 후처리 과정이 필요하다. 일반적으로 키포인트는 사용자가 지정한 임계값을 기준으로 추출한다. 하지만 최적의 임계값을 찾는 과정은 큰 노력을 요구하며, 데이터 세트(또는 이미지)마다 최적의 값이다를 수 있다. 본 연구는 사용자의 직접 임계값 지정 대신, 대상의 이미지에 맞추어 적절한 임계값을 자동으로 설정하는 키포인트 추출 알고리즘을 제안한다. 본 논문의 키포인트 추출 알고리즘은 차선 영역과 배경의 명확한 구분을 위해 줄 단위 정규화를 사용한다. 그리고 커널 밀도 추정을 사용하여, 각 줄에서 각 차선의 키포인트를 추출한다. 제안하는 알고리즘은 TuSimple과 CULane 데이터 세트에 적용되었으며, 고정된 임계값 사용대비 정확도 및 거리오차 측면에서 1.80%p와 17.27% 향상된 결과를 얻는 것을 확인하였다.
Abstract
Deep-learning-based image segmentation is one of the most widely employed lane detection approaches, and it requires a post-process for extracting the key points on the lanes. A general approach for key-point extraction is using a fixed threshold defined by a user. However, finding the best threshold is a manual process requiring much effort, and the best one can differ depending on the target data set (or an image). We propose a novel key-point extraction algorithm that automatically adapts to the target image without any manual threshold setting. In our adaptive key-point extraction algorithm, we propose a line-level normalization method to distinguish the lane region from the background clearly. Then, we extract a representative key point for each lane at a line (row of an image) using a kernel density estimation. To check the benefits of our approach, we applied our method to two lane-detection data sets, including TuSimple and CULane. As a result, our method achieved up to 1.80%p and 17.27% better results than using a fixed threshold in the perspectives of accuracy and distance error between the ground truth key-point and the predicted point.
참고문헌 (28)
- 이민철, "무인차량 적용을 위한 차선강조기법 기반의 차선 인식", 제어.로봇.시스템학회 논문지, 16(7), pp.625~631 (2010), 제어·로봇·시스템학회
- 이강수, "영상의 채도와 도로 폭 정보를 사용한 차선 검출 및 추적 방법", 제어.로봇.시스템학회 논문지, 25(5), pp.476~483 (2019), 제어·로봇·시스템학회[DOI]
- Z. Kim, "Robust lane detection and tracking in challenging scenarios", IEEE Transactions on intelligent trans-portation systems, 9(1), pp.16~26 (2008)
- M. Aly, "Real time detection of lane markers in urban streets", IEEE Intelligent Vehicles Symposium, pp.7~12 (2008)
- G. Liu, "Combining statistical hough ´ transform and particle filter for robust lane detection and tracking", IEEE Intelligent Vehicles Symposium, pp.993~997 (2010)
- 김광백, "Real Time Road Lane Detection with RANSAC and HSV Color Transformation", Journal of Information and Communication Convergence Engineering, 15(3), pp.187~192 (2017), 한국정보통신학회[DOI]
- T. Zheng, "Resa : Recurrent feature-shift aggregator for lane detection", Proc. of the AAAI Conference on Artificial Intelligence, 35(4), pp.3547~3554 (2021)
- X. Pan, "Spatial as deep: Spatial cnn for traffic scene understanding", Proc. of the AAAI Conference on Artificial Intelligence, 32(1) (2018)
- D. Neven, "Towards end-to-end lane detection : an instance segmenta tion approach", IEEE intelligent vehicles symposium (IV), pp.286~291 (2018)
- L. Tabelini, "Keep your eyes on the lane : Attention-guided lane detection", arXiv e-prints, pp. arXiv–2010, ~ (2020)
- H. Xu, "Curvelane-nas : Unifying lane-sensitive architecture search and adaptive point blending", European Conference on Computer Vision, pp.689~704 (2020), Springer
- L. Tabelini, "Polylanenet : Lane estimation via deep polynomial regression", 2020 25th International Conference on Pattern Recognition (ICPR), pp.6150~6156 (2021)
- R. Liu, "End-to-end lane shape prediction with transformers", Proc. of the IEEE, 2021, 3694~3702
- K. He, "Deep residual learning for image recognition", Proc. of the IEEE conference on computer vision and pattern recognition, pp.770~778 (2016)
- [URL]
- Brody Huval, "An empirical evaluation of deep learning on highway driving", arXiv pre- print arXiv:1504.01716, ~ (2015)
- Z. Qin, "Ultra fast structure-aware deep lane detec tion", European Conference on Computer Vision, pp.276~291 (2020), Springer
- S. Yoo, "End-to-end lane marker detection via row-wise classification", Proc. of the IEEE, 2020, 1006~1007
- X. Li, "Line-cnn : End-to-end traffic line detection with line proposal unit", IEEE Transactions on Intelligent Transportation Systems, 21(1), pp.248~258 (2019)
- S. Ren, "Faster R-CNN : Towards real time object detection with region proposal networks", Advances in neural information processing systems, 28 (2015)
- Y. Dong, "A hy-brid spatial–temporal deep learning architecture for lane detection", Computer-Aided Civil and Infrastructure Engineering (2022)
- O. Ronneberger, "U-Net : Convolutional networks for biomedical image segmentation", International Conference on Medical image computing and computer-assisted intervention, pp.234~241 (2015), Springer
- E. Parzen, "On estimation of a probability density function and mode", The annals of mathematical statistics, 33(3), pp.1065~1076 (1962)
- [URL]
- Zhan Qu, "Focus on local: Detecting lane marker from bottom up via key point", Proceedings of the IEEE, 2021, ~
- Tu Zheng, "CLRNet : Cross Layer Refinement Network for Lane Detection", Proceedings of the IEEE, 2022, ~
- 민승기, "빛의내부산란의 효과적인 표현을위한밀도 추정기법", 한국컴퓨터그래픽스학회논문지, 16(1), pp.9~20 (2010), (사)한국컴퓨터그래픽스학회[DOI]
- De Gelder, E., "Constrained sampling from a kernel density estimator to generate scenarios for the assessment of automated vehicles", 2021IEEE Intelligent Vehicles Symposium Workshops, pp.203~208 (2021)