A Rapid Image Matching Approach for Auxiliary Navigation

To improve the velocity and accuracy of multi-sensor image matching in auxiliary navigation,after the edge feature of image is extracted,a 3-4DT method is applied to transform the edge binary image,then the transformed edge distance image is taken as the matching feature.A Hausdorff distance integrating points set coincidence numbers(I-HD) is proposed to overcome the limitation of traditional Hausdorff distance,and I-HD can be used as the similarity measure.On the basis of the human visual system,a far-near stratification search strategy is adopted.Meanwhile a genetic algorithm using real-coding is applied to accelerate the speed of matching at the bottom of image matching.Experimental results show that average matching time is 1283ms and average error value is 1.036,which shows that the method can meet the navigation requirement.