Vehicle Detection from Aerial Images Using Local Shape Information

Authors:
Jae-Young Choi;Young-Kyu Yang
Affiliations:
College of IT, Kyungwon University, Seongnam, Republic of Korea 461-701;College of IT, Kyungwon University, Seongnam, Republic of Korea 461-701
Venue:
PSIVT '09 Proceedings of the 3rd Pacific Rim Symposium on Advances in Image and Video Technology
Year:
2009

Citing 5
Cited 0

Shape Matching and Object Recognition Using Shape Contexts

IEEE Transactions on Pattern Analysis and Machine Intelligence
Robust analysis of feature spaces: color image segmentation

CVPR '97 Proceedings of the 1997 Conference on Computer Vision and Pattern Recognition (CVPR '97)
Vehicle Detection on Aerial Images: A Structural Approach

ICPR '96 Proceedings of the International Conference on Pattern Recognition (ICPR '96) Volume III-Volume 7276 - Volume 7276
Distinctive Image Features from Scale-Invariant Keypoints

International Journal of Computer Vision
SURF: speeded up robust features

ECCV'06 Proceedings of the 9th European conference on Computer Vision - Volume Part I

Quantified Score

Hi-index	0.00

Visualization

Abstract

Detection and extraction of vehicle objects in high resolution satellite imagery are required in many transportation applications. This paper presents an approach to automatic vehicle detection from aerial images. The initial extraction of candidate vehicle is based on Mean-shift algorithm with symmetric character of blob-like car structure. By fusing the density and the symmetry, the method can remove the ambiguous blobs and reduce the cost of the detected ROI processing in the subsequent stage. To verify the blob as a vehicle, log-polar shape descriptor is used for measuring similarity. The edge strengths are obtained and represented as its spatial histogram by the orientation and distance from the center of blob. The proposed algorithm is able to successfully detect the vehicle and very useful for the traffic surveillance system.