基于深度学习的生物资产检测模型YOLOSC  

The Biological Asset Detection Model YOLOSC Based on Deep Learning

作　　者：关昆仑 朱思文 张仰森[1] 成琪昊 张学凯 GUAN Kun-lun;ZHU Si-wen;ZHANG Yang-sen;CHENG Qi-hao;ZHANG Xue-kai(Institute of Intelligent Information Processing,Beijing Information Science and Technology University,Beijing 100192,China)

机构地区：[1]北京信息科技大学智能信息处理研究所,北京100192

出　　处：《科学技术与工程》2025年第2期674-682,共9页Science Technology and Engineering

基　　金：北京市社会科学基金重点项目(21GLA007)。

摘　　要：为提高生物资产监盘审计过程中盘点准确性和盘点效率,提出了一种融入注意力机制和损失函数优化的生物资产检测模型YOLOSC。首先,将压缩-激励网络(squeeze-and-excitation networks,SENet)注意力机制引入YOLOv5s模型的主干网络中,以增强对生物资产图片中关键特征的提取能力;其次,采用完全交并比(complete intersection over union,CIoU)作为检测框回归的损失函数,以提升训练过程中检测框的回归速度与定位精度;最后,构建了一个生物资产数据集对所提模型进行针对性训练,以提升模型检测效果。实验结果表明,相较于YOLOv5模型,YOLOSC的精确率、召回率、F_(1)和AP分别提升了2.3%、2.1%、2.7%和1.6%,证明了所提出的生物资产检测模型YOLOSC的有效性。In order to improve the accuracy and efficiency of inventory counting in the process of monitoring and auditing biological assets,a biological asset detection model YOLOSC incorporating the attention mechanism and loss function optimization was proposed.Firstly,the SENet attention mechanism was introduced into the backbone network of the YOLOv5s model to enhance the ability of extracting the key features in the pictures of the biological assets.Secondly,the CIoU was adopted as the regression of the detection frames with the loss function to enhance the regression speed and localization accuracy of the detection frame during the training process.Finally,a biological asset datasets was constructed for targeted training of the proposed model to enhance the model detection effect.The experimental results show that compared with the YOLOv5model,the precision,recall,F_(1)value and AP of YOLOSC are improved by 2.3%,2.1%,2.7%and 1.6%,respectively,which proves the effectiveness of the proposed biological asset detection model YOLOSC.

关 键 词：目标检测模型 YOLOv5 注意力机制 损失函数 生物资产审计 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]