Title :
Automatic ship classification by superstructure moment invariants and two-stage classifier
Author :
Zhongliang, Qian ; Wenjun, Wang
Author_Institution :
Dept. of Inf. & Electron. Eng., Zhejiang Univ., Hangzhou, China
Abstract :
Moment invariants have been used as descriptive feature in a variety of object recognition applications. A new descriptive scheme called superstructure moment invariants is presented. The authors calculate moment invariants only for the superstructure of a ship, which are different from the general moment invariants for the entire ship. The analysis of the theory and the statistics of the experimental results shows that the classification accuracy using superstructure moment invariants for the ship is higher than that of general ones. A modified 1-NN classifier, two stage classification is presented, which reduces the time expense and has a high classification accuracy
Keywords :
pattern recognition; ships; automatic ship classification; object recognition; superstructure moment invariants; two-stage classifier; Aircraft; Azimuth; Ear; Libraries; Marine vehicles; Neural networks; Object recognition; Sea measurements; Shape; Statistics;
Conference_Titel :
Singapore ICCS/ISITA '92. 'Communications on the Move'
Print_ISBN :
0-7803-0803-4
DOI :
10.1109/ICCS.1992.254892