基于样本扩充网络的水声目标分类模型优化算法

摘要:

水声目标识别是近年来各国的研发热点，但是由于水声目标难以采集而导致样本数据不足，严重影响了神经网络的识别效率以及自动化识别装备的水平和性能的发挥。为此，提出了一种基于样本扩充网络的水声目标分类模型优化方法，通过搭建掩模重建的样本扩充网络，充分利用无标注数据进行训练，使模型学习到样本的全局高维特征，再生成样本加入到后续的识别模型训练中，在两次试验过程中，平均识别准确率从76%提升至到80%，最佳识别准确率从88%提升至96%。基于实测数据的实验表明，该方法提升了分类器的准确率、收敛速度以及稳定性。

Abstract:

Underwater acoustic target recognition is a research and development hot spot in many countries in recent years. However, due to the difficulty of collecting underwater acoustic targets, the sample data is insufficient, which seriously affects the recognition efficiency of neural network and the level and performance of automatic recognition equipment. Therefore, an optimization method of underwater acoustic target classification model based on the sample expansion network is proposed. By building the sample expansion network reconstructed by the mask, the model is trained by making full use of the unlabeled data, so that the model can learn the global high-dimensional features of the samples, and then generate the samples to be added to the subsequent recognition model training. Based on the result of two experiments, the average accuracy of target classification model improves from 76% to 80% while the maximum accuracy of target classification model improves from 88% to 96%. Experiments show that this method improves the accuracy, convergence speed and stability of the classifier.