首页 > 分享 > 基于Inception

基于Inception

萌宠菠菠乐园
2024-12-09 14:55

基于Inception-CSA深度学习模型的鸟鸣分类

作者单位:

中南林业科技大学计算机与信息工程学院/人工智能应用研究所，长沙 410004

基金项目:

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

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

Author:

LI Huaicheng

College of Computer and Information Engineering/Institute of Applied Artificial Intelligence， Central South University of Forestry and Technology，Changsha 410004，China
在期刊界中查找
在百度中查找
在本站中查找

YANG Daowu

WEN Zhifang

WANG Ya’nan

CHEN Aibin

Affiliation:

College of Computer and Information Engineering/Institute of Applied Artificial Intelligence， Central South University of Forestry and Technology，Changsha 410004，China

摘要 | 图/表 | 访问统计 | 参考文献 [22] | | || 文章评论

摘要:

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模块获取鸟鸣声特征图的全局注意力权重，将二者的输出结合得到更强的特征图，再次利用最大池化层对特征图进行下采样；最后利用全连接层进行分类，得到最终的分类结果。以采集的华南地区自然环境中的10种野生鸟类的鸣叫声构建数据集，用于实验部分以验证方法的有效性。结果表明，本研究提出的方法在自建数据集上准确率达到了93.11%，相比于基于其他经典模型的分类方法，基于Inception-CSA模型的分类方法在拥有较少模型参数量的同时达到了更高的准确率。

Abstract:

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-CSA deep learning model， which consists of three steps including bird audio sample preprocessing， feature extraction， and classifier classification. First， the samples of bird sounds were preprocessed into Mel spectrum maps with the same size as the feature maps of bird sounds. Then the feature of bird sounds was extracted with the Inception-CSA model including the Inception module extracting the multi-scale local time-frequency domain features in the feature map of bird sounds and the CSA module obtaining the global attention weights of the feature map of bird sounds. The output of both was combined to obtain a stronger feature map. The feature maps were downsampled with the maximum pooling layer. Finally， the results of final classification were obtained with the fully connected layer. The calls of 10 wild bird species in the natural environment of south China were collected and the dataset was constructed to verify the effectiveness of the method. The results showed that the proposed method achieved 93.11% accuracy on the self-built dataset. The classification method based on the Inception-CSA model had higher accuracy with fewer model parameters compared with the classification methods based on other classical models.

引用本文

李怀城,杨道武,温治芳,王亚楠,陈爱斌.基于Inception-CSA深度学习模型的鸟鸣分类[J].华中农业大学学报,2023,42(3):97-104

复制

分享文章指标点击次数:289 下载次数: 770 HTML阅读次数: 113 历史收稿日期:2022-09-19 在线发布日期: 2023-06-20 文章二维码