Abstract: This series of papers deal with ship-target recognition. The project is conducted by using fuzzy neural networks and basing on the spectra of ship radiated--noise. Based on the studies of a large amount of ship radiated-noise data, which has been collected from actual ships on the sea, we extracted effectively recognizable features. Such features include lin^pectrum features, stationary and nonstationary spectrum features as well as rhythm features. Finally the categorization are tested by unknown samples on the sea, including, 33 surface ships, 8 underwater targets in 30 operating conditions.Methods for memorization and classification are also explored in the project. Paper (Ⅱ) is the second of the series paper. It focuses on how to memorize the stable features of line spectrum of specific ship targets by using line spectrum pattern plate and on related problems. This paper examines the analyzing parameters of line spectrum: average times, time length and their impact on the occurrence of stable lines. It compares the impact of two different average times on the occurrence of stable lines (occurrence ratio > 70 %) and unstable lines, and shows that it takes longer time span for average when stable lines for recognition are used. Moreover, the paper discusses the statistic methods of establishing line spectrum pattern plate using stable lines, including the definition of stability and related parameters.the stability of line spectrum and the uniqueness of stable lines are investigated in over 1000 samples gathered from 43 ships in 65 operating conditions (with an original recording time of 3.5 hours). The results demonstrate the statistical implication of such uniqueness. The average overlapping ratio is 5 %,the proportion of ships without stable lines is 8spectrums is not an identifying feature, distinguishing 'A' type ships from 'B' kind ships.