文献类型：期刊文献

英文题名：Small target detection algorithm based on SAHI-Improved-YOLOv8 for UAV imagery: A case study of tree pit detection

作者：Liang, Xiuhao[1] Xiang, Jun[1] Qin, Sheng[2,3] Xiao, Yundan[4] Chen, Lifen[5] Zou, Dongxia[1] Ma, Honglun[6] Huang, Dong[7] Huang, Yongxin[8] Wei, Wei[1]

第一作者：Liang, Xiuhao

通信作者：Wei, W[1]

机构：[1]Guangxi Forestry Res Inst, Guangxi Oil tea Super Species Cultivat Res Ctr Eng, Guangxi Lab Forestry, Nanning Eucalypt Plantat Ecosyst Observat & Res St, Nanning 530002, Peoples R China;[2]Guangxi Normal Univ, Sch Elect & Informat Engn, Guangxi Key Lab Brain inspired Comp & Intelligent, Guilin 541000, Peoples R China;[3]Guangxi Normal Univ, Educ Dept Guangxi Zhuang Autonomous Reg, Key Lab Nonlinear Circuits & Opt Commun, Guilin 541000, Peoples R China;[4]Chinese Acad Forestry, Inst Forest Resource Informat Tech, Beijing 100091, Peoples R China;[5]Yulin Normal Univ, Sch Phys & Telecommun Engn, Yulin 537006, Peoples R China;[6]Guangxi Zhuang Autonomous Reg State Owned Dongmen, Chongzuo 532108, Peoples R China;[7]Guilin Forestry Sci Inst, Guilin 541004, Peoples R China;[8]Guangxi State Owned Beijianghe Forest Farm Rongshu, Liuzhou 545000, Peoples R China

年份：2025

卷号：12

外文期刊名：SMART AGRICULTURAL TECHNOLOGY

收录：Scopus(收录号：2-s2.0-105010680331);WOS:【ESCI(收录号:WOS:001548287800001)】；

基金：This research was funded by Guangxi Forestry Research Institute Basic Research Operating Expenses "Research on intelligent monitoring method of forest survival rate based on UAV imagery" under grant Linke202312; In part by Guangxi Science and Technology Plan Project under grant GuikeAB21076012; In part by National Natural Science Regional Foundation under Grant 32160362; In part by the sub-project of the National Key Research and Development Plan under grant 2023YED2201705-4; In part by Guangxi Self-Funded Forestry Science and Technology Project under Grant GXLK [2022ZC] 99. Thanks to the planting areas and forest farm staff who have assisted in the implementation of the project.

语种：英文

外文关键词：Tree pit detection; Small target; SAHI-improved-YOLOv8; UAV imagery; Deep learning

摘要：The application of deep learning in tree pit detection of unmanned aerial vehicle (UAV) images has problems such as dense distribution, high density, small size, false detections, missed detections, and high localization error. To address these challenges, this paper proposes a small target detection algorithm based on SAHIImprovedYOLOv8 for detecting tree pits, which includes Slicing Aided Hyper Inference (SAHI), Focal loss, Spatial Pyramid Pooling Concurrent Spatial Pyramid Convolution (SPPCSPC), and Convolutional Block Attention Module (CBAM). The accuracy of identification and positioning can be improved by using the SAHI via cutting high-resolution UAV imagery into slices that match the detection model, avoiding the loss of small target detail caused by direct downsampling. The identification accuracy is improved by using the Focal loss and CBAM, SPPCSPC to mitigate the data imbalance, strengthen key semantic features, and realize fine-grained information enhancement. The experimental results show that the SAHI-Improved-YOLOv8 model outperforms YOLOv3, YOLOv5, YOLOv8, YOLOv10, YOLOv11 and SAHI-YOLOv8 with a Precision of 85.17 %, a Recall of 85.07 %, a AP50-90 of 78.63 % and a F1 score of 85.12 %. In conclusion, the SAHI-Improved-YOLOv8 has the capability of efficiently processing high-resolution images, which alleviates the problems of high density of small targets, false detections, missed detections, and high localization error. In practical applications, the SAHI-Improved-YOLOv8 model performs excellently in tree pit detection in UAV imagery, significantly reducing false detections and missed detections, and providing reliable technology support for large-scale forest management.