期刊论文

Inception-CSA deep learning model-based classification of bird sounds

中文

华中农业大学学报

1000-2421

2023

42

3

97-104

10.13300/j.cnki.hnlkxb.2023.03.012

62276276:国家自然科学基金 2019TP1015:智慧物流技术湖南省重点实验室项目 CX20210879:湖南省研究生科研创新项目

本校为第一机构

为进一步提高通过声音识别鸟类的精确度,本研究提出基于Inception-CSA深度学习模型的鸟鸣声分类方法,包含鸟鸣声音频样本预处理、特征提取、分类器分类等步骤。首先将鸟鸣声样本预处理成尺寸相同的梅尔频谱图,作为鸟鸣声特...展开更多 为进一步提高通过声音识别鸟类的精确度,本研究提出基于Inception-CSA深度学习模型的鸟鸣声分类方法,包含鸟鸣声音频样本预处理、特征提取、分类器分类等步骤。首先将鸟鸣声样本预处理成尺寸相同的梅尔频谱图,作为鸟鸣声特征图;其次利用Inception-CSA模型对鸟鸣声特征图进行特征提取,其中Inception模块提取鸟鸣声特征图中的多尺度局部时频域特征,CSA模块获取鸟鸣...

Bird sounds have diverse features,and most of the current convolutional neural network models based on a single receptive field are difficult to learn the diversity of bird sound features from audio containing complex background noise.In this article,we proposed a metho...MORE Bird sounds have diverse features,and most of the current convolutional neural network models based on a single receptive field are difficult to learn the diversity of bird sound features from audio containing complex background noise.In this article,we proposed a method of classifying bird sounds based on the Inception-...