A line extraction algorithm for synthetic aperture radar (SAR) image is presented.The algorithm is designed for the statistical characteristics of the speckle in SAR image.Three steps are involved.Firstly

a new edge detector

which combines the Canny operator and Ratio operator

is used to get the edge points and their directions

then the edge points are grouped based on the edge direction consistency to form the initial lines.Finally

the broken lines due to speckle noise are connected by the high level grouping.The proposed new edge operator is CFAR and prevents the line from cleaving.The high level grouping step analyses the initial lines

finds out the possible areas for expanding lines

and verifies the possible areas by statistical hypothesis determination.The line extraction algorithm is tested on the X-band airborne SAR images.The extracted line graph can be used for SAR image vectorization and comprehension etc.