海底声学底质分类的ELM-AdaBoost方法

基于自适应增强算法（AdaBoost）结合极限学习机（ELM），通过迭代、调整、优化ELM分类器之间的权值，从而构建了具有强鲁棒性、高精度的ELM-AdaBoost强分类器，增强了现有的ELM分类器的稳定性。以珠江口海区侧扫声呐图像为实验数据，对礁石、砂、泥3类典型底质进行分类识别，该方法的平均分类精度超过90%，优于单一ELM分类器的平均分类精度85.95%，也优于LVQ、BP等传统分类器，且在分类所耗时间上也远少于传统分类器。实验结果表明，本文构建的ELM-AdaBoost方法可有效应用于海底声学底质分类，可满足实时底质分类的需求。

ELM-AdaBoost method of acoustic seabed sediment classification

Based on the adaptive boosting algorithm (AdaBoost) combined with the extreme learning machine (ELM), the strong classifier of ELM-AdaBoost with strong robustness and high precision is thus constructed by iterating, adjusting, and optimizing the weights between each ELM classifier. ELM-AdaBoost method can enhance the stability of the existing ELM classifier. In this paper, the data collected by side scan sonar in the Zhujiang River Estuary was used to classify and identify three types of typical sediments as rock, sand, and mud. The average classification accuracy of new method exceeds 90%, which is better than the average classification accuracy of a single ELM classifier of 85.95%. It is also superior to other traditional classifiers (i.e. LVQ and BP) and it takes much less time to classify than traditional classifiers. The experimental result shows that the proposed ELM-AdaBoost method can be effectively applied to the classification and identification of seabed sediment and can meet the needs of real-time classification of seabed sediment.