基于无人机航拍与改进YOLOv5s的油茶果实检测

doi:10.13733/j.jcam.issn.20955553.2024.12.035

摘要/Abstract

摘要：

针对无人机航拍图像中油茶果实小且互相遮挡的问题，提出改进YOLOv5s模型。首先，使用SPD-Conv代替YOLOv5s模型中池化操作，使模型在执行下采样操作时能够保留更多细粒度信息。然后，在YOLOv5s模型的颈部网络末端引入坐标注意力机制CA，提高模型对遮挡目标的鲁棒性。另外，改进YOLOv5s模型使用NWD边界框损失函数替换YOLOv5s中的CIOU边界框损失函数，以提升模型对无人机航拍图像中小油茶果实的检测能力。改进YOLOv5s模型的精确率、召回率、F1分数和平均精度均值分别达到93.1%、90.5%、91.78%和91.2%，与YOLOv5s模型相比，平均精度均值提升3.6个百分点。试验表明，改进YOLOv5s对航拍图像中较小的油茶果实和遮挡果实有更强的检测能力。可为利用无人机进行油茶果实的产量估计研究提供参考。

关键词: 油茶果实, 无人机航拍, YOLOv5s, 坐标注意力机制, 边界框损失函数

Abstract:

Aiming at the problem that the fruit of Camellia oleifera is small and occluding each other in UAV aerial images, an improved YOLOv5s model is proposed. Firstly, SPD-Conv is used to replace the pooling operation in the YOLOv5s model, so that the model can retain more fine-grained information during the down-sampling operation. Then, Coordinate Attention (CA) is introduced at the end of the neck network of YOLOv5s model to improve the robustness of the model to occluding targets. Additionally, the improved YOLOv5s model replaces the YOLOv5s CIOU bounding box loss function with the NWD (Normalized Gaussian Wasserstein) bounding box loss function to improve its ability to detect small Camellia oleifera fruits in drone aerial images. The precision, recall, F1 score, and mean average precision of the improved YOLOv5s model are 93.1%, 90.5%, 91.78% and 91.2%, respectively. Compared to the YOLOv5s model, the improved YOLOv5s model's mean average precision has increased by 3.6 percentage points. The experiments indicate that the improved YOLOv5s has stronger detection capabilities for smaller and occluded Camellia oleifera fruits in aerial images. This research can provide a reference for the estimation of Camellia oleifera fruit yield by using drones.

Key words: Camellia oleifera, UAV aerial photography, YOLOv5s, coordinate attention mechanism, bounding box loss function

中图分类号:

沈德宇, , 陈锋军, , 朱学岩, , 张新伟, 陈闯. 基于无人机航拍与改进YOLOv5s的油茶果实检测[J]. 中国农机化学报, 2024, 45(12): 238-244.

Shen Deyu, , Chen Fengjun, , Zhu Xueyan, , Zhang Xinwei, Chen Chuang. Camellia oleifera fruit detection based on UAV aerial photography and improved YOLOv5s[J]. Journal of Chinese Agricultural Mechanization, 2024, 45(12): 238-244.

参考文献

［1］　罗凡, 费学谦, 郭少海. 油茶果采收及干燥方式对油茶籽油品质的影响［J］. 中国油脂, 2015, 40(11): 69-73.
Luo Fan, Fei Xueqian, Guo Shaohai. Effects of harvest time and drying methods on quality of oiltea camellia seed oil ［J］. China Oils and Fats, 2015, 40(11): 69-73.
［2］　闫锋欣, 李许杰, 杨永霞, 等. 手持冲击梳刷式油茶果采摘装置设计与试验［J］. 农业机械学报, 2023, 54(12):129-140.
Yan Fengxin, Li Xujie, Yang Yongxia, et al. Design and experiment of handheld impacting combtype Camellia oleifera fruit harvester ［J］. Transactions of the Chinese Society for Agricultural Machinery, 2023, 54(12): 129-140.
［3］　伍德林, 杨俊华, 刘芸, 等. 我国油茶果采摘装备研究进展与趋势［J］. 中国农机化学报, 2022, 43(1): 186-194.
Wu Delin, Yang Junhua, Liu Yun,et al. Research progress and trend of camellia fruit picking equipment in China ［J］. Journal of Chinese Agricultural Mechanization, 2022, 43(1): 186-194.
［4］　Yuan H, Huang K, Ren C, et al. Pomelo tree detection method based on attention mechanism and crosslayer feature fusion ［J］. Remote Sensing, 2022, 14(16): 3902.
［5］　Koirala A, Walsh K B, Wang Z, et al. Deep learning for realtime fruit detection and orchard fruit load estimation: Benchmarking of ‘MangoYOLO’ ［J］. Precision Agriculture, 2019, 20(6): 1107-1135.
［6］　赵辉, 乔艳军, 王红君, 等. 基于改进YOLOv3的果园复杂环境下苹果果实识别［J］. 农业工程学报, 2021, 37(16): 127-135.
Zhao Hui, Qiao Yanjun, Wang Hongjun, et al. Apple fruit recognition in complex orchard environment based on improved YOLOv3 ［J］. Transactions of the Chinese Society of Agricultural Engineering, 2021, 37(16): 127-135.
［7］　Wu D, Jiang S, Zhao E, et al. Detection of Camellia oleifera fruit in complex scenes by using YOLOv7 and data augmentation ［J］. Applied Sciences, 2022, 12(22): 11318.
［8］　Tang Y, Zhou H, Wang H, et al. Fruit detection and positioning technology for a Camellia oleifera C. Abel orchard based on improved YOLOv4-tiny model and binocular stereo vision ［J］. Expert Systems with Applications, 2023, 211: 118573.
［9］　吕帅朝, 马宝玲, 宋磊, 等. 基于YOLON网络的多形态油茶果实夜间检测方法研究［J］. 西北农林科技大学学报(自然科学版), 2023(8): 141-154.
Lü Shuaichao, Ma Baoling, Song Lei, et al. Nighttime detection method of polymorphic Camellia oleifera fruits based on YOLON network ［J］. Journal of Northwest A & F University (Natural Science Edition), 2023(8): 141-154.
［10］　宋怀波, 王亚男, 王云飞, 等. 基于YOLO v5s的自然场景油茶果识别方法［J］. 农业机械学报, 2022, 53(7): 234-242.
Song Huaibo, Wang Yanan, Wang Yunfei, et al. Camellia oleofera fruit detection in natural scene based on YOLOv5s ［J］. Transactions of the Chinese Society for Agricultural Machinery, 2022, 53(7): 234-242.
［11］　Chen S, Zou X, Zhou X, et al. Study on fusion clustering and improved YOLOv5 algorithm based on multiple occlusion of Camellia oleifera fruit ［J］. Computers and Electronics in Agriculture, 2023, 206: 107706.
［12］　严恩萍, 棘玉, 尹显明, 等. 基于无人机影像自动检测冠层果的油茶快速估产方法［J］. 农业工程学报, 2021, 37(16): 39-46.
Yan Enping, Ji Yu, Yin Xianming, et al. Rapid estimation of camellia oleifera yield based on automatic detection of canopy fruits using UAV images ［J］. Transactions of the Chinese Society of Agricultural Engineering, 2021, 37(16): 39-46.
［13］　尹川, 苏议辉, 潘勉, 等. 基于改进YOLOv5s的名优绿茶品质检测算法［J］. 农业工程学报, 2023, 39(8): 179-187.
Yin Chuang, Su Yihui, Pan Mian, et al. Detection of the quality of famous green tea based on improved YOLOv5s ［J］. Transactions of the Chinese Society of Agricultural Engineering, 2023, 39(8): 179-187.
［14］　Chen C, Liu M Y, Tuzel O, et al. R-CNN for small object detection ［C］. Computer VisionACCV 2016: 13th Asian Conference on Computer Vision, 2017: 214-230.
［15］　Sunkara R, Luo T. No more strided convolutions or pooling: A new CNN building block for lowresolution images and small objects ［C］. Joint European Conference on Machine Learning and Knowledge Discovery in Databases， 2022: 443-459.
［16］　Hou Q, Zhou D, Feng J. Coordinate attention for efficient mobile network design ［C］. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition， 2021: 13713-13722.
［17］　Xu C, Wang J, Yang W, et al. Detecting tiny objects in aerial images: A normalized Wasserstein distance and a new benchmark ［J］. ISPRS Journal of Photogrammetry and Remote Sensing, 2022, 190: 79-93.
［18］　Sajjadi M S M, Vemulapalli R, Brown M. Framerecurrent video superresolution ［C］. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition， 2018: 6626-6634.

[1]	宋立航, 张屹, 时寅豪. 基于改进 YOLOv5s的自然环境下油桃成熟度检测方法[J]. 中国农机化学报, 2024, 45(9): 250-257.
[2]	李洁, 高尚兵, 余骥远, 陈新, 李士丛, 袁星星. 基于PMMS-Net和叶绿素荧光成像的绿豆叶斑病抗性鉴定方法[J]. 中国农机化学报, 2024, 45(8): 210-216.
[3]	冯晓, 张辉, 刘运超, 张微, 李小红, 马中杰, . 基于改进YOLOv5s的硬核期葡萄簇检测[J]. 中国农机化学报, 2024, 45(8): 240-245.
[4]	王克涛, 陈世锋, 陈贵, 韦锦, 蒙丽雯, 陈泉成. 基于YOLOv5s的园林球形绿篱检测方法[J]. 中国农机化学报, 2024, 45(8): 262-269.
[5]	黄尧, 何敬, 付饶, 刘刚, , 林远杨. YOLOv5s-CBAM算法在福寿螺虫卵识别中的应用分析[J]. 中国农机化学报, 2024, 45(6): 223-228.
[6]	严蓓蓓, 纪元浩, 曲凤凤, 许金普. 基于改进YOLOv5s的茶叶嫩芽检测[J]. 中国农机化学报, 2024, 45(4): 168-174.
[7]	席光泽, 周建平, 许燕, 彭炫, 崔超. 基于改进YOLOv5s的复杂环境下棉花顶芽识别[J]. 中国农机化学报, 2024, 45(12): 275-280.
[8]	叶荣, 马自飞, 高泉, 李彤, 邵郭奇, 王白娟. 基于改进YOLOv5sECAASFF算法的茶叶病害目标检测[J]. 中国农机化学报, 2024, 45(1): 244-251.
[9]	李光明, 弓皓斌, 袁凯. 基于轻量化YOLOv5s的花椒簇检测研究[J]. 中国农机化学报, 2023, 44(4): 153-158.