水下目标特征提取与分类的C30硬件实验研究

Implementation and Experiment for Feature Extraction and Classification of Underwater Targets Using C30 Chip Hardware

本文针对水下目标识别问题，在Ｃ３０信号处理机上实现了目标特征提取、特征压缩和分类三个环节的几个典型方法，并实验研究了这些方法的性能。结果表明，Ｃ３０处理器能有效地实现复杂的目标识别方法，且各方法表现出良好的识别效果和实时性能，为目标识别技术实际应用提供了重要的依据和实验手段。

In this paper, some typical methods of feature extraction, feature compression and classification to underwater targets are implemented using C30 digital signal processor(DSP). The recognition effects and running time of these methods on C30 DSP are tested by a great number of simulation signal and real record data. Experimental results show that C30 DSP can completely implemented these complex targets recognition methods.The methods have good recognition effects and real time property.This work provides an important experimental method and evidence for targets recognition techniques.