Title :
The Method of Pavement Image Splicing Based on SIFT Algorithm
Author :
Hou Xiang-dan ; Dong Yong-feng ; Guo Hai-jiao ; Xin, Yang
Author_Institution :
Coll. of Comput. Sci. & Software, Hebei Univ. of Technol., Tianjin, China
Abstract :
According to the characteristic of pavement image, a new splicing method for pavement image based on scale invariant feature transform was proposed. The SIFT algorithm is used to detect the feature points, FCM clustering is used to find the most similitude cluster relative to the number of feature point, and match the feature descriptors in the cluster ,then the false feature points pair is rejected by RANSAC algorithm. Finally the correct match feature point´s pair is used to realize image splicing in order to display the panoramic images of pavement images. There is a great reference value for improving the system of pavement image Automatic detection.
Keywords :
feature extraction; image processing; transforms; RANSAC algorithm; SIFT algorithm; pavement image automatic detection; pavement image splicing; scale invariant feature transform; Automated highways; Clustering algorithms; Computer science; Computer vision; Educational institutions; Image edge detection; Image segmentation; Intelligent systems; Software algorithms; Splicing; FCM clustering; eliminate mismatch; image splicing; scale invariant feature transform algorithm;
Conference_Titel :
Intelligent Systems, 2009. GCIS '09. WRI Global Congress on
Conference_Location :
Xiamen
Print_ISBN :
978-0-7695-3571-5
DOI :
10.1109/GCIS.2009.363
