首页|通道剪枝与知识蒸馏相结合的轻量化SAR目标检测

通道剪枝与知识蒸馏相结合的轻量化SAR目标检测

扫码查看
轻量化SAR目标检测方法对快速检测SAR影像中的地物目标具有重要意义.针对轻量化检测方法精度不高的问题,设计了一种通道剪枝与知识蒸馏相结合的轻量化SAR目标检测方法.该方法通过对复杂网络中批归一化层的缩放因子γ进行稀疏化训练,判别对应特征通道的重要程度,进而裁剪次要通道,并在微调训练后将其作为教师网络,构造知识蒸馏框架指导轻量模型训练,提高轻量模型的检测精度.采用YOLOv5-6.1算法搭建了检测框架,并在重组的MSAR和SSDD多类目标数据集上进行了训练和检测试验,结果表明该方法能够在保持模型体积仅3.73 MB的轻量化条件下,提升SAR目标检测精度,验证了本文方法的有效性.
Lightweight SAR target detection based on channel pruning and know-ledge distillation
Lightweight SAR target detection algorithm is of great significance for the rapid detection of ground objects in SAR images.Aiming at the low precision of lightweight detection algorithm,a lightweight SAR target detection method combining channel pruning and knowledge distillation was designed.In this method,the importance of the corresponding feature channels is identified by sparse training of the scaling factor y of the batch normalization layer in the complex network,and then the sec-ondary channels are cut.After fine-tuning training,it is used as a teacher network to construct a knowledge distillation frame-work to guide the training of lightweight model and improve the detection accuracy of light weight model.The YOLOv5-6.1 algorithm was used to build a detection framework,and the training and detection experiments were carried out on the reconsti-tuted MSAR and SSDD multi-class target datasets.The results show that the proposed method can improve the accuracy of SAR target detection while maintaining the lightweight model size of only 3.73 MB,which verifies the effectiveness of the pro-posed method.

SARtarget detectionlightweightchannel pruningknowledge distillation

黄启灏、靳国旺、熊新、王丽美、李佳豪

展开 >

信息工程大学地理空间信息学院,河南郑州 450001

河南城建学院测绘与城市空间信息学院,河南平顶山 467000

SAR 目标检测 轻量化 通道剪枝 知识蒸馏

国家自然科学基金国家自然科学基金国家自然科学基金

414740106140150942201492

2024

测绘学报
中国测绘学会

测绘学报

CSTPCD北大核心
影响因子:1.602
ISSN:1001-1595
年,卷(期):2024.53(4)
  • 28