Research On Recognition And Matching For Marked Point In Photogrammetry

The photographic identification and matching of marked point is a key issue in measurement. With the rapid development of Photogrammetry, The photographic identification and matching has been made great progress, especially in the transformation from the traditional photogrammetry to digital photogrammetry.In order to increase the accuracy of image recognition and matching, people pay more attention on the various algorithms. A method of computer automatic identification marker points has been proposed in this paper. Each step has been discussed in detail, including image binarization, edge detection, scanning and testing standards. This method is an effective solution of the marker points in the photogrammetric problem of automatic identification. The Experimental results show that the method can identify the mark points efficiently thus improve the effectiveness of this method.A complete suit of decoding and matching methods have been presented for the characteristic of fan shape coding marked points. This method uses the characteristics of its shape fully avoiding the problem of excessive computing bought by the previous algorithm and the inaccurate of marked point matching. The Experimental data show that this simple method has high matching accuracy, and is easy to implement.When measuring a large work piece, they can not meet the actual demand, due to a limited number of coding points marked. So the use of non-coding marker points is needed. The matching is a difficult in marked points due to no non-coding marker point marker at the characteristics of coding shape. Based on the theory in projective geometry geometric invariants, by which the non-coding marker point-plane matching can be realized. The experimental results show that this method has a certain practicality and can also match effectively and efficiently in most non-coding markers point matching by using this method.