中文摘要：

根据人耳听觉模型以归一化响度主分量为特征，采用模糊直方图检索音频信息，并根据响度数据的统计分布对直方图编码进行优化。实验结果表明，该方法在低信噪比下具有较好的性能，对噪音具有良好的顽健性（Robustness）。为噪音下的固定音频检索问题提供了有效的解决办法。

英文摘要：

A novel audio retrieval method using fuzzy histogram based on principal loudness component is proposed. In the construction of histogram, statistical distribution of loudness is used to optimize the histogram coding. Experimental results show that the method has better performance when the SNR is relatively low compared with other histograms. It is an effective audio retrieval method under noisy environment.