Currently

synthetic aperture radar（SAR） is a hot research topic in the field of Microwave Remote Sensing.It possesses many incomparable advantages such as the capability to work at all time and under all weather

high spatial resolution and strong penetrability through the ground surface and so on.Therefore

it is significant to extract information from SAR imagery.It can not only compensate the deficiency in optical imagery

but also help constructing the spatial database.Road information is a kind of most important spatial basic information

which has a lot of significant applications

including military defense

map matching

database updating etc.Ratio of average（RoA） algorithm is a classical method for road extraction from SAR imagery developed in recent 20 years

however

because of SAR speckle noise and other factors

it still has some disadvantages.For example :low locality accuracy

thickened edges

higher false alarming rate and non-continuous road segments.This paper is mainly divided into two sections to resolve these problems.In the first section

it introduces a local ratio detector for road extraction.But it has defects on the accuracy of position and widening road width.In order to tacle these problems

a pruning window is added into the process

which is used to remove effects of rivers and trees.Therefore

a thinned and more accurate road map can be gained.Based on the above method

in the second section

we firstly describe the road linking method using chain codes

define the chain code energy network and import the prior knowledge into processing course.Then

we analyze linking probability of line element in simulated road map and propose an approach to optimize the chain codes.Because the process is programmed with classed computation

and need no search for all line segments so that it improves efficiency highly.At last

this method is applied to extracting the road network from real SAR image and its validity has been proved.