摘要
误差反传神经网络在被动声呐目标分类中得到了广泛的应用,但该算法在搜索过程中容易陷入局部最小值,同时使用赢者独活的识别决策策略,导致识别率的下降。针对BP网络的弱点,提出了基于模糊非对称贴近度的舰船辐射噪声分类方法,该方法能在对BP网络不作大改动的前提下,解决神经网络陷入局部最小值的问题和使用赢者独活决策策略导致误识的弱点。实验结果证明了该方法是有效的。
Though Back-Propagation neural network is widely used in passive sonar target classification, it tends to come into the local minimum value. Being accompanied with Winner Takes All (WTA) strategy taken in classification, then the recognition is led to be low. Against the disadvantages of BP nets, ship radiated-noise classification based on fuzzy degree of nearness is proposed in this paper. The experimental result confirms that this method is more effective in passive sonar target classification.
出处
《电声技术》
北大核心
2004年第6期49-52,共4页
Audio Engineering
关键词
模糊贴近度
舰船
辐射噪声
被动声呐目标
fuzzy degree of nearness
back-propagation algorithm
classification
