Volume 11 - Issue 1
Improved vehicle occlusion segmentation method based on feature points combination
Abstract
Based on the algorithm introduction and the defect analysis for the original method that using the combinations of four kinds of feature points, this paper proposes an improved method that appended another four kinds of feature points for the segmentation of vehicle occlusion. A rectangle region is assumed as the shape template of a vehicle. After the feature points on the contour of the connected region were extracted, sorted and merged, every vehicle could be separated from each other by comparing every four adjacent feature points with the specific feature points combination. Simulation results show that the improved method is more robust, accurate, easily achieved and has good practical application merit, comparing with the original one.
Paper Details
PaperID: 84872827949
Author's Name: Ma, Z., Yin, S., Li, C., Zheng, Y.
Volume: Volume 11
Issues: Issue 1
Keywords: Feature points combination, Improved segmentation method, Occluded vehicles separation
Year: 2015
Month: January
Pages: 373-380