论文信息 - Wide-Baseline Obstacle Mapping Using Monocular Camera for Unmanned Surface Vehicle

Wide-Baseline Obstacle Mapping Using Monocular Camera for Unmanned Surface Vehicle

In this paper, we applied a static obstacle mapping system for USV. This system consists of a monocular camera, GPS and compass. Obstacle map can be built while the USV is traveling. To increase the accuracy of the system, we apply feature matching after image transformation and introduce a threshold to filter motion parallax computation results. Moreover, we collect radar data as ground truth to evaluate the system performance and accuracy. According to the results of our experiments, this system can map the static obstacles with high accuracy and efficiency.

Han Wang | Jiaying Chen

[1] Tom Drummond,et al. Faster and Better: A Machine Learning Approach to Corner Detection , 2008, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2] H. C. Longuet-Higgins,et al. A computer algorithm for reconstructing a scene from two projections , 1981, Nature.

[3] Bernhard P. Wrobel,et al. Multiple View Geometry in Computer Vision , 2001 .

[4] Han Wang,et al. Vision-based navigation of an unmanned surface vehicle with object detection and tracking abilities , 2017, Machine Vision and Applications.

[5] Vincent Lepetit,et al. BRIEF: Binary Robust Independent Elementary Features , 2010, ECCV.

[6] Robert C. Bolles,et al. Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography , 1981, CACM.

[7] V. Lepetit,et al. EPnP: An Accurate O(n) Solution to the PnP Problem , 2009, International Journal of Computer Vision.

[8] C. Lawrence Zitnick,et al. Structured Forests for Fast Edge Detection , 2013, 2013 IEEE International Conference on Computer Vision.

[9] Han Wang,et al. Wide-Baseline Stereo-Based Obstacle Mapping for Unmanned Surface Vehicles , 2018, Sensors.

[10] Gary R. Bradski,et al. ORB: An efficient alternative to SIFT or SURF , 2011, 2011 International Conference on Computer Vision.