Paper Info
Title
Determination Of Length And Width Of A Line-Segment By Using A Hough Transform
Abstract
The standard Hough transform does not provide length and width of a line-segment detected in an image; it just detects the normal parameters of the line. We present a novel method for determining also length and width of a line segment by using the Hough transform. Our method uses statistical analysis of voting cells around a peak in the Hough space. In image space, the voting cells and voting values are analysed. The functional relationship between voting variance and voting angle is deduced. We approximate this relationship by a quadratic polynomial curve. In Hough space, the statistical variances of columns around a peak are computed and used to fit a quadratic polynomial function. The length and width of a line segment are determined simultaneously by resolving the equations generated by comparing the corresponding coefficients of two functions. We tested and verified the proposed method on simulated and real-world images. Obtained experimental results demonstrate the accuracy of our novel method for determining length and width of detected line segments.
Year
DOI
Venue
2014
10.1007/978-3-319-09955-2_16
DISCRETE GEOMETRY FOR COMPUTER IMAGERY, DGCI 2014
Keywords
Field
DocType
Hough transform, length, width, curve fitting
Discrete mathematics,Line segment,Curve fitting,Voting,Computer science,Hough space,Hough transform,Algorithm,Quadratic function,Statistical analysis
Conference
Volume
ISSN
Citations 
8668
0302-9743
1
PageRank 
References 
Authors
0.35
16
3
Name
Order
Citations
PageRank
Zezhong Xu1667.12
Bok-Suk Shin2689.27
Klette, R.3645.29