Paper Info
Title
An Algorithm for Automatic Recognition of Manhole Covers Based on MMS Images.
Abstract
Considering difficulties in performing automatic recognition for manhole covers under complex background, this paper proposed a Hough transform algorithm to locating the possible positions of the manhole covers based on features of ellipse geometry extracting from images by Mobile Mapping Systems (MMS). Firstly, the original images need to be preprocessed by image enhancement, edge detection, and morphological closing operation. Then the processed image was filled, the image noises with very small areas were removed, and the remaining blocks were identified. A ratio of the ellipse area to its outer rectangle area was calculated as the threshold value to eliminate large amount of image noises so as to obtain the maximum likely positions of manhole covers. Finally, the Hough transform was used to identify the accurate locations of the manhole covers. Experiments with fifty images have showed that the proposed algorithm can achieve a 88 % accuracy of identifying the manhole covers correctly, automatically and quickly.
Year
DOI
Venue
2016
10.1007/978-981-10-2260-9_4
ADVANCES IN IMAGE AND GRAPHICS TECHNOLOGIES, IGTA 2016
Keywords
DocType
Volume
Manhole cover,Automatic recognition,Geometric properties,Coarse positioning,Hough transform
Conference
634
ISSN
Citations 
PageRank 
1865-0929
0
0.34
References 
Authors
0
2
Name
Order
Citations
PageRank
Zhang Chong100.34
Yang Liu214.15