Feature extraction for target spacecraft in the final approaching phase of rendezvous and docking

The rendezvous and docking technology is the key issue to accomplish the spacecraft maintenance in-orbit, space station supply, and astronauts visiting. The technology of the vision measurement for spacecraft directly determines that its results are successful or fail in the final approaching phase of rendezvous and docking. Feature extraction for target spacecraft includes edges, lines, circles, and especially motion information. In this paper, image smoothing based on 2-Dimensional Adaptive Wiener Filtering is introduced, then Canny Edge Detector is used for edge detection, finally using Standard Hough Transform to extract lines and the characteristic circle. Experimental results show that the computation is obviously improved, meanwhile, the precision of detection is also improved. What’ s more, lines and circles detection as the fundamental step of extracting motion information, even the vision measurement for spacecraft position and attitude determination, is absolutely necessary.