| 申小虎,李冠宇,史洪飞,王传之.网络剪枝与知识蒸馏相结合的轻量级鸟声识别方法*[J].,2025,44(2):350-361 |
| 网络剪枝与知识蒸馏相结合的轻量级鸟声识别方法* |
| A light-weight bird sound recognition method combining network pruning and knowledge distillation |
| 投稿时间：2023-10-27 修订日期：2025-02-28 |
| 中文摘要: |
| 在鸟声识别应用中，算法模型多数采用参数密集型，缺少能够搭载至被动声学监测设备的高效网络。针对EfficientNet网络结构特点，将结构化剪枝与知识蒸馏方法相结合，确保剪枝后的网络保持良好的泛化能力，能够满足不同资源配置条件下的网络需求。一方面，通过逆背包准则建立了剪枝通道与资源间的信息表述，在保留网络框架条件下完成通道剪枝。另一方面，在知识蒸馏方法中通过加入MBConv模块内部蒸馏损失分量并完成训练，确保跨组信息交换保留了剪枝前后特征映射之间的距离。通过对南京浦口区老山森林中收集的10类鸟声检测分类实验，在压缩后网络参数量仅3.0 M的条件下，分类精度可达到91.64%。该文所提方法在完成网络规模压缩的同时，较好地保留了分类精度，与相同规模主流轻量级网络相比较，能更好地适应鸟声识别被动声学监测的设备需求。 |
| 英文摘要: |
| In bird sound recognition, most algorithm models heavily rely on parameters and lack efficient networks compatible with passive acoustic monitoring (PAM) equipment. The deficit becomes pronounced given the typically complex of EfficientNet network structure. This challenge is addressed by synergistically combining structured pruning and knowledge distillation techniques, which upholds the generalization capability of the pruned EfficientNet network while accommodating diverse resource allocation conditions. The network pruning was reconceptualized using an inverse knapsack criterion. This strategic approach facilitates channel pruning while maintaining the network’s foundational architecture. Simultaneously, in the knowledge distillation method, by adding internal distillation loss components between the MBConv modules and conducting subsequent training, the preservation of the disparity between feature maps before and after pruning in cross-group information exchange was ensured. Through experiments involving the classification of ten distinct bird sound types recorded within the Laoshan Forest in Pukou District, Nanjing, a classification accuracy of 91.64% with a compressed network parameter of merely 3.0 M was obtained. This approach achieves network scale compression while preserving classification accuracy. Compared to mainstream methods of equivalent scale, the proposed technique more adeptly meets the requirements of PAM equipment tailored for bird sound recognition. |
| DOI：10.11684/j.issn.1000-310X.2025.02.009 |
| 中文关键词: 网络剪枝 知识蒸馏 鸟声识别 轻量级网络 被动声学监测 |
| 英文关键词: network pruning knowledge distillation bird sound recognition light-weight network passive acoustic monitoring |
| 基金项目:国家自然科学基金项目（面上项目，重点项目，重大项目），野生动植物物证技术国家林业和草原局重点实验室开放课题 |
|
| 摘要点击次数: 13 |
| 全文下载次数: 12 |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
