MOVING OBJECT DETECTION ON A RUNWAY PRIOR TO

Зарегистрируйтесь, чтобы просмотреть полный документ!

MOVING OBJECT DETECTION ON A RUNWAY PRIOR TO LANDING USING AN ONBOARD INFRARED CAMERA Dr. Gerard Medioni Cheng Hua Pai Yu Ping Lin

Introduction n Input: Infrared runway sequence n Goal: Detect moving objects on runway

Approach n We do it in two steps: 1. Stabilize the sequence 2. Detect motion on the stabilized sequence

Flow chart of the system Video In Reference Frame Runway Identification Update Reference frame Image Stabilization Motion Detection Yes Update Reference frame? Locally Stabilized Image Sequence and Homographies Blobs in motion

Stabilization n Issues: n n n Planar region containing the runway Feature choice and matching Transformation between consecutive frames

Stabilization n Approach n n n Manually Label planar region SIFT provides sufficient and descriptive features RANSAC to estimate best transformation

Stabilization n Result: Stabilized runway sequence

Adaptive Reference Frame n Issues: n For longer sequence Small errors accumulate n Big scale difference n Beginning of a Sequence End of a Sequence

Adaptive Reference Frame n When to change reference frame? n Check the lower edge length ratio

Stabilization algorithm Landing UAV image sequence Manually labeled planar region input 1. Extract SIFT features 4. Use RANSAC to remove outliers and estimate homography 2. Region of Interest 5. Update reference frame if necessary 3. Match features to previous frame to establish correspondence 6. Warp to the reference frame output Locally stabilized image sequence and for all s

Adaptive Reference Frame n Result: Original Sequence Locally Stabilized Sequence

Detection module n Issues: n n n Detection method Global intensity variation Noise Moire in the sequence n Poor stabilization n Local intensity variation n Random noise n

Detection n Approach: n Use simple Gaussian background model t = (1 - ) * ( t-1) + * (It) t 2 = (1 - ) * ( t-1) 2 + * (It- t) n Foreground: More than 4 t 2 from mean Foreground Background 4 t 2 t Intensity distribution of an image

Global intensity variation n Approach: n Compensate gain with affine transformation [Yalcin 05] Before compensation After compensation

Noise reduction n Approach: n Moire in the sequence n n Compare 8 neighbouring background pixels Poor stabilization n Restabilize with gradient map (also SIFT) To Gradient

Noise reduction n Approach: n Local intensity variation n n Intensity normalization on the foreground pixels Random noise n Compare consecutive foreground masks With random noise Without random noise

Detailed flow chart of Motion Detection Module

Detection Result: Locally Stabilized Sequence Foreground mask

Evaluation 1. 2 1 0. 8 0. 6 0. 4 0. 2 0 Speed vs. Detection Rate Obj. size 2 x 2 2. 8 2. 5 2. 2 1. 9 1. 6 1. 3 1 4 x 4 5 x 5 6 x 6 00 0 0. 1 00 00 0. 00 4 00. . . 3 x 3 0. 7 n Tested on 150 synthesized and 18 realworld sequences Results (synthetic data): Detection rate n Speed (pixel/frame)

Conclusion n Detection affected by: n n n Program limitation: n n n Object speed and size Threshold parameters Moving objects fade in and out Bad result near the end of the sequence Future work: n n More test on larger dataset Speed improvement

Reference n n n n n W. G. Chris Stauffer. Adaptive background mixture models for real-time tracking. 1999 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'99), 2: 2246, 1999. M. Fischler and R. Bolles. Random sample consensus: A paradigm for model fitting with applications to image analysis and automated cartography. Communications of the ACM, 24(6): 381 -395, 1981. C. Harris and M. Stephens. A combined corner and edge detector. Proceedings of The Fourth Alvey Vision Conference, (1): 147 -152, 1988. T. G. R. Kasturi, O. Camps and S. Devadiga. Detection of obstacles on runway using ego-motion compensation and tracking of signicant features. Proceedings 3 rd IEEE Workshop on Applications of Computer Vision, 1996 (WACV'96), pages 168 -173, 1996. D. G. Lowe. Distinctive image features from scale invariant keypoints. International Journal of Computer Vision, 60(2): 91 -110, 2004. R. S. B. Sridhar and B. Hussien. Passive range estimation for rotor-craft low-altitude flight. Machine Vision and Applications, 6(1): 10 -24, 1993. S. Sull and B. Sridhar. Runway obstacle detection by controlled spatiotemporal image Low disparity. IEEE Transactions on Robotics and Automation, 15(3): 537 -547, 1999. R. C. H. Yalcin and M. Hebert. Background estimation under rapid gain change in thermal imagery. Second IEEE Workshop on Object Tracking and Classification in and Beyond the Visible Spectrum (OTCBVS'05), 2005. Q. Zheng and R. Chellappa. Motion detection in image sequences acquired from a moving platform. Proc. Int. Conf. Acoustics, Speech, and Signal Processing, Minneapolis, 5: 201 -204, 1993.

Скачать презентацию MOVING OBJECT DETECTION ON A RUNWAY PRIOR TO

261f5030ca8270e72f5585dd80839b9a.ppt

Количество слайдов: 21