As a part of complex scene analysis system, this paper presents a line finder which can find a globally good line with respect to the knowledge available of the line to be sought. A variety of knowledge available of the line to be sought can be communicated to the line finding system simply in the form of descriptions so that the higher level of the program can easily use this line finding system. The optimization method is utilized to detect a globally good line with respect to the given knowledge of the line. In order to reduce the required storage space and computation time in the optimization process, we use the following techniques: 1) represent a line by a sequence of linear segments and find the good sequence of segments by the optimization method and 2) use a locus model of search to prune unpromising iines during the optimization process. Some experimental results applied to various scenes with different amount of noise and knowledge are given in the paper.
[1]
Saburo Tsuji,et al.
Detection of Ellipses by a Modified Hough Transformation
,
1978,
IEEE Transactions on Computers.
[2]
Ugo Montanari,et al.
On the optimal detection of curves in noisy pictures
,
1971,
CACM.
[3]
Alberto Martelli,et al.
Edge detection using heuristic search methods
,
1972,
Comput. Graph. Image Process..
[4]
Masahiko Yachida,et al.
A Versatile Machine Vision System for Complex Industrial Parts
,
1977,
IEEE Transactions on Computers.
[5]
Alberto Martelli,et al.
An application of heuristic search methods to edge and contour detection
,
1976,
CACM.
[6]
Arnold K. Griffith,et al.
Edge Detection in Simple Scenes Using a Priori Information
,
1973,
IEEE Transactions on Computers.
[7]
Raj Reddy,et al.
The LOCUS Model of Search and its Use in Image Interpretation
,
1977,
IJCAI.