Research On Edge Exrtaction Of Optical Subaperture Based On Fractal Dimension

Optical synthetic aperture imaging technology is an effective approach to increase the aperture diameter of optical system for purpose of improving resolution. In optical synthetic aperture imaging system, the edge is more complex than that of traditional optical imaging system, and the relatively large size of the gaps between the subapertures makes cophasing a difficult problem. Hence, in order to avoid the loss of effective information and achieve phase stitching, it is significant to extract precise edge phase information of each segmented mirror surface from interference intensity measurements. As a measure feature of image surface irregularities, fractal dimension statistically evaluates the complexity and provides a powerful tool to do the edge extraction of optical subaperture.In our research, the edge extraction algorithm of optical subaperture based on differential box-counting dimension with window merge replication was proposed by studying traditional edge extraction algorithm based on fractal theory. The improved differential box-counting method was used to calculate fractal dimension. Specifically, window merge replication method was innovatively proposed to resolve the conflict of fractal dimension estimated in relatively small area. The size of scanning window and times of window merge replication were determined through the experimental analysis. By the evaluation experiment, our proposed algorithm has been proved to an effective approach with advantages in continuity, accuracy and no directional limit.According to the characteristics of interference pattern, the basic process of edge extraction based on our proposed algorithm was determined. And the subaperture region and interference fringe edge was accurately extracted by using two-step strategy of preprocessing and postprocessing. Through the optimization of process and speed, the software design of our algorithm has been completed which has a better practicality. Our research has laid the foundation for achieving cophasing of subaperture in the future work.