数字农科院2.0

基于改进YOLOv8的自然环境下柑橘果实识别

文献类型： 中文期刊

作者： 余圣新;韦莹莹;方辉;李敏;柴秀娟;曾志康;覃泽林

作者机构：

关键词： 柑橘果实识别;卷积神经网络;YOLOv8;ODConv全维动态卷积;MPDIoU损失函数

期刊名称：湖北农业科学

ISSN： 0439-8114

年卷期： 2024 年 63 卷 8 期

页码：

收录情况： 科技核心（2024版） ; ; 农林核心（2020版）

摘要： 为实现柑橘果实的精准快速识别,提出了一种改进YOLOv8网络模型。首先将YOLOv8网络模型中的部分传统卷积替换为ODConv全维动态卷积,以增强YOLOv8网络模型在复杂的自然环境下的适应能力,然后将YOLOv8的CIoU损失函数替换为MPDIoU损失函数,解决了CIoU损失函数在特殊情况下退化的问题,接着通过消融试验,分别验证了ODConv全维动态卷积与MPDIoU损失函数的有效性,改进后YOLOv8n、YOLOv8s、YOLOv8m、YOLOv8l、YOLOv8x的平均识别精度mAP分别从86.40%、88.92%、88.97%、88.99%、89.11%提高至88.25%、89.32%、89.57%、89.90%、90.12%。试验结果表明,ODConv全维动态卷积与MPDIoU损失函数能有效提高YOLOv8网络模型在自然环境下的柑橘果实识别能力。

分类号：

  • 相关文献

[1]基于改进YOLOv8模型的冬小麦穗识别技术. 宫志宏,闫锦涛,于红,刘涛,刘布春,李树岩. 2026

[2]基于改进实例分割算法的区域养殖生猪计数系统. 张岩琪,周硕,张凝,柴秀娟,孙坦. 2024

[3]End-to-end stereo matching network with two-stage partition filtering for full-resolution depth estimation and precise localization of kiwifruit for robotic harvesting. Xudong Jing,Hanhui Jiang,Shiao Niu,Haosen Zhang,Bryan Gilbert Murengami,Zhenchao Wu,Rui Li,Chengquan Zhou,Hongbao Ye,Jinyong Chen,Yaqoob Majeed,Longsheng Fu. 2024

[4]GVC-YOLO: A Lightweight Real-Time Detection Method for Cotton Aphid-Damaged Leaves Based on Edge Computing. Zhenyu Zhang,Yunfan Yang,Xin Xu,Liangliang Liu,Jibo Yue,Ruifeng Ding,Yanhui Lu,Jie Liu,Hongbo Qiao. 2024

[5]ACCURATE NON-DESTRUCTIVE TESTING METHOD FOR POTATO SPROUTS FOCUSING ON DEFORMABLE ATTENTION. Geng, Binxuan,Dai, Guowei,Zhang, Huan,Qi, Shengchun,Dewi, Christine. 2024

[6]Lightweight cotton diseases real-time detection model for resource-constrained devices in natural environments. Pan Pan,Mingyue Shao,Peitong He,Lin Hu,Sijian Zhao,Longyu Huang,Guomin Zhou,Jianhua Zhang. 2024

[7]Xoo-YOLO: a detection method for wild rice bacterial blight in the field from the perspective of unmanned aerial vehicles. Pan Pan,Wenlong Guo,Xiaoming Zheng,Lin Hu,Guomin Zhou,Jianhua Zhang. 2023

[8]Weed Recognition at Soybean Seedling Stage Based on YOLOV8nGP + NExG Algorithm. Tao Sun,Longfei Cui,Lixuan Zong,Songchao Zhang,Yuxuan Jiao,Xinyu Xue,Yongkui Jin. 2024

[9]A Method of Grape Cluster Target Detection and Picking Point Location Based on Improved YOLOv8. Liu, Huaiyang,Liu, Wanfu,Wang, Wenhao,Li, Huibin,Geng, Changxing. 2025

[10]Field Obstacle Detection and Location Method Based on Binocular Vision. Yuanyuan Zhang,Kunpeng Tian,Jicheng Huang,Zhenlong Wang,Bin Zhang,Qing Xie. 2024

[11]FEL-YoloV8: A New Algorithm for Accurate Monitoring Soybean Seedling Emergence Rates and Growth Uniformity. Yu, Xun,Jiang, Tiantian,Zhu, Yanqin,Li, Liming,Fan, Fan,Jin, Xiuliang. 2025

[12]YOLOv8-MSP-PD: A Lightweight YOLOv8-Based Detection Method for Jinxiu Malus Fruit in Field Conditions. Yi Liu,Xiang Han,Hongjian Zhang,Shuangxi Liu,Wei Ma,Yinfa Yan,Linlin Sun,Linlong Jing,Yongxian Wang,Jinxing Wang. 2025

[13]基于可见光图像和卷积神经网络的冬小麦苗期长势参数估算. 马浚诚,刘红杰,郑飞翔,杜克明,张领先,胡新,孙忠富. 2019

[14]基于卷积神经网络的温室黄瓜病害识别系统. 马浚诚,杜克明,郑飞翔,张领先,孙忠富. 2018

[15]基于卷积神经网络的冬小麦麦穗检测计数系统. 张领先,陈运强,李云霞,马浚诚,杜克明. 2019

[16]基于深度卷积神经网络的红树林物种无人机监测研究. 黄亦其,刘琪,赵建晔,黄文善,孙中宇,乔曦. 2020

[17]卷积神经网络在高分辨率影像分类中的应用. 李贤江,陈佑启,邹金秋,石淑芹,郭涛,蔡为民,陈浩. 2019

[18]基于卷积神经网络的水稻纹枯病图像识别. 刘婷婷,胡林,王婷. 2019

[19]一种基于密度估计和VGG-Two的大豆籽粒快速计数方法. 王莹,李越,武婷婷,孙石,王敏娟. 2021

[20]基于MSRA初始化卷积神经网络的草地牧草分类研究. 刘一磊,刘江平,赵烜赫,马玉宝,闫伟红,潘新. 2021

作者其他论文 更多>>