Abstract：In order to obtain precisely the features of moving vessels at the port and track the target quickly and efficiently while saving system resources, a target recognition algorithm based on invariant pseudo-line moment was proposed by combining the characteristics of traditional moment invariants and invariant line moment. The computational complexity was analyzed and the rapidity of the algorithm was verified. Meanwhile, the stability of invariant pseudo-line moments was also analyzed. The computational domain of each target for an image was obtained, and the pretreatment of edge detection was completed. Finally, the invariant pseudo-line moments were derived by calculating binary image. In addition, the performance of the invariant pseudo-line moments was evaluated, and simulation test and analysis were conducted. The results show that the algorithm can accurately obtain the features of the moving objects＇ regions while improving the computing speed, and the new method has been successfully applied in the intelligent monitoring system for moving vessels.
2010, Vol. 31 
