基于改进YOLOv8的自然环境下板栗栗蓬检测方法

doi:10.13733/j.jcam.issn.20955553.2024.12.037

摘要/Abstract

摘要：

针对人工敲打式收获板栗带来高成本和安全风险问题，研究无人机板栗采摘方法至关重要。为达到在自然光照条件下迅速且精确地识别板栗栗蓬目标，提出一种基于YOLOv8的改进卷积网络模型板栗栗蓬检测方法。对YOLOv8主干网络的C2f模块添加CBAM注意力机制，增强卷积网络模型对板栗栗蓬特征提取能力。在YOLOv8的头部增加一个微小栗蓬目标检测头，与YOLOv8原有的3个检测头共同组成检测模块，使网络模型更好地捕捉小板栗栗蓬目标特征。经自建数据集上的训练和验证试验，改进后卷积网络YOLOv8-Vcj板栗栗蓬检测精确率比YOLOv8高1.3%，mAP@0.5和mAP@0.5∶0.95值比YOLOv8分别提高4.6%和3.4%。改进卷积网络板栗栗蓬检测误差主要来自光照条件和图像中板栗栗蓬目标的密集程度。研究结果表明:融合CBAM注意力机制和增加微小目标检测头的改进卷积神经网络YOLOv8-Vcj能够有效实现树上板栗栗蓬的检测。

关键词: 板栗栗蓬, YOLOv8, 目标检测, CBAM, 检测头

Abstract:

In view of the high cost and safety risk caused by artificial knock harvest chestnut, it is very important to study the unmanned aerial vehicle chestnut harvest methods. In order to rapidly and precisely identify chestnut targets under natural light conditions, a modified convolutional network model detection method based on YOLOv8 was proposed. The CBAM attention mechanism was added to the C2f module of the YOLOv8 backbone network to enhance the convolutional network model ability of extracting chestnut features. A small chestnut target detection head was added to the head of YOLOv8 which formed the detection module together with the original three detection heads of YOLOv8. This method (YOLOv8-Vcj) enabled the network model to better capture the target features of small chestnut. Through training and validation experiments on the self-built data set, the detection accuracy of YOLOv8-Vcj was 1.3% higher than YOLOv8 and the mAP@0.5 and mAP@0.5∶0.95 values were 4.6% and 3.4% higher than YOLOv8, respectively. The chestnut detection error of the improved convolution network mainly comes from the light conditions and the density of chestnut targets in the images. The research results show that the improved convolutional neural network YOLOv8-Vcj of combining the CBAM attention mechanism and a small target detection head can effectively detect chestnuts on the tree.

Key words: chestnut peng, YOLOv8, object detection, CBAM, detection head

中图分类号:

TS736+.4

李志臣, 罗卫平, 凌秀军, 李鸿秋. 基于改进YOLOv8的自然环境下板栗栗蓬检测方法[J]. 中国农机化学报, 2024, 45(12): 251-258.

Li Zhichen, Luo Weiping, Ling Xiujun, Li Hongqiu. Detection method of Chinese chestnut in natural environment based on improved YOLOv8[J]. Journal of Chinese Agricultural Mechanization, 2024, 45(12): 251-258.

参考文献

［1］　宗望远, 黄木昌, 肖洋轶, 等. 板栗收获拍打式落果装置设计与试验［J］. 农业工程学报, 2022, 37(18): 1-10.
Zong Wangyuan, Huang Muchang, Xiao Yangyi, et al. Design and experiment of the fruitbeating dropping device for chestnut harvesters ［J］. Transactions of the Chinese Society of Agricultural Engineering, 2022, 37(18): 1-10.
［2］　张桃桃. 基于可拓设计的无人机水果采摘的结构设计［D］. 广州：广东工业大学, 2016.
Zhang Taotao. The design of the structure about UAV fruit picker based on the extension design ［D］. Guangzhou： Guangdong University of Technology, 2016.
［3］　雒鹰, 程鸿远, 孟黎鹏, 等. 利用无人机采摘红松球果初探［J］. 林业机械与木工设备, 2022, 50(1): 64-66.
Luo Ying, Cheng Hongyuan, Meng Lipeng, et al. Preliminary study on picking pinus koraiensis cones using unmanned aerial vehicles ［J］. Forestry Machinery & Wood Working Equipment, 2022, 50(1): 64-66.
［4］　黄丽明, 王懿祥, 徐琪, 等. 采用YOLO算法和无人机影像的松材线虫病异常变色木识别［J］. 农业工程学报, 2021, 37(14): 197-203.
Huang Liming, Wang Yixiang, Xu Qi, et al. Recognition of abnormally discolored trees caused by pine wilt disease using YOLO algorithm and UAV images ［J］. Transactions of the Chinese Society of Agricultural Engineering, 2021, 37(14): 197-203.
［5］　鲍文霞, 谢文杰, 胡根生, 等. 基于TPH-YOLO的无人机图像麦穗计数方法［J］. 农业工程学报, 2023, 39(1): 155-161.
Bao Wenxia, Xie Wenjie, Hu Gensheng, et al. Wheat ear counting method in UAV images based on TPH-YOLO ［J］. Transactions of the Chinese Society of Agricultural Engineering, 2023, 39(1): 155-161.
［6］　王磊磊, 王斌, 李东晓, 等. 基于改进YOLOv5的菇房平菇目标检测与分类研究［J］. 农业工程学报, 2023, 39(17): 163-171.
Wang Leilei, Wang Bin, Li Dongxiao, et al. Object detection and classification of pleurotus ostreatus using improved YOLOv5 ［J］. Transactions of the Chinese Society of Agricultural Engineering, 2023, 39(17): 163-171.
［7］　肖德琴, 曾瑞麟，周敏, 等. 基于DH-YOLOX的群养马岗鹅关键行为监测［J］. 农业工程学报, 2023, 39(2): 142-149.
Xiao Deqin, Zeng Ruilin, Zhou Min, et al. Monitoring the vital behavior of Magang geese raised in flocks based on DH-YOLOX ［J］. Transactions of the Chinese Society of Agricultural Engineering, 2023, 39(2): 142-149.
［8］　王政, 许兴时, 华志新, 等. 融合YOLOv5n与通道剪枝算法的轻量化奶牛发情行为识别［J］. 农业工程学报, 2022, 38(23): 130-140.
Wang Zheng, Xu Xingshi, Hua Zhixin, et al. Lightweight recognition for the oestrus behavior of dairy cows combin ing YOLOv5n and channel pruning ［J］.Transactions of the Chinese Society of Agricultural Engineering, 2022, 38(23): 130-140.
［9］　金耀, 何秀文, 万世主, 等. 基于YOLOv3的生猪个体识别方法［J］. 中国农机化学报, 2021, 42(2): 178-183.
Jin Yao, He Xiuwen, Wan Shizhu, et al. Individual pig identification method based on YOLOv3 ［J］. Journal of Chinese Agricultural Mechanization, 2021, 42(2): 178-183.
［10］　张继成, 侯郁硕, 郑萍, 等. 低数据集下基于ASPP-YOLO V5模型的苋菜识别方法研究与试验［J］. 农业机械学报， 2023， 54（S2): 223-228.
Zhang Jicheng, Hou Yushuo, Zheng Ping, et al. Research and experiment of amaranth identification method based on ASPP-YOLO V5 model in low data set ［J］. Transactions of the Chinese Society for Agricultural Machinery， 2023， 54（S2): 223-228.
［11］　王秀红, 王庆杰, 李洪文, 等. 基于YOLOv5-Jexton TX2的秸秆覆盖农田杂草检测方法［J］. 农业机械学报, 2023, 54 (11): 39-48.
Wang Xiuhong, Wang Qingjie, Li Hongwen, et al. Weed detection method of strawcovered farmland based on YOLOv5-Jexton TX2 ［J］. Transactions of the Chinese Society for Agricultural Machinery, 2023, 54(11): 39-48.
［12］　乔琛, 韩梦瑶, 高苇, 等. 基于Faster-NAM-YOLO的黄瓜霜霉病菌孢子检测［J］. 农业机械学报, 2023, 54(12): 288-299.
Qiao Chen, Han Mengyao, Gao Wei, et al. Quantitative detection of cucumber downy mildew spores at multiscale based on Faster-NAM-YOLO ［J］. Transactions of the Chinese Society for Agricultural Machinery, 2023, 54(12): 288-299.
［13］　乔珠峰，赵秋菊，郭建鑫, 等. 基于改进YOLOv5的草莓病害智能识别终端设计［J］. 中国农机化学报, 2024, 45(3): 205-211.
Qiao Zhufeng, Zhao Qiuju, Guo Janxini, et al. Design of strawberry disease intelligent identification terminal based on improved YOLOv5［J］. Journal of Cinese Agricultural Mechanization, 2024, 45(3): 205-211.
［14］　刘子洋，徐慧英，朱信忠, 等. Bi-YOLO：一种基于YOLOv8改进的轻量化目标检测算法［J］. 计算机工程与科学, 2024, 46(8): 1444-1454.
Liu Ziyang, Xu Huiying, Zhu Xinzhong, et al. Bi-YOLO: An improved lightweight object detection algorithm based on YOLOv8 ［J］. Computer Engineering & Science, 2024, 46(8): 1444-1454.
［15］　王春梅，刘欢. YOLOv8-VSC:一种轻量级的带钢表面缺陷检测算法［J］.计算机科学与探索, 2024, 18(1): 151-161.
Wang Chunmei, Liu Huan. YOLOv8-VSC: A lightweight algorithm for strip surface defect detection ［J］. Journal of Frontiers of Computer Science and Technology, 2024, 18(1): 151-161.
［16］　崔克彬，焦静颐. 基于MCB-FAH-YOLOv8的钢材表面缺陷检测算法［J］. 图学学报, 2024, 45(1): 112-125.
Cui Kebin, Jiao Jingyi. Steel surface defect detection algorithm based on MCB-FAH-YOLOv8 ［J］. Journal of Graphics, 2024, 45(1): 112-125.
［17］　李松，史涛，井方科. 改进YOLOv8的道路损伤检测算法［J］. 计算机工程与应用, 2023, 59(23): 165-174.
Li Song, Shi Tao, Jing Fangke. Improved road damage detection algorithm of YOLOv8 ［J］.Computer Engineering and Applications, 2023, 59(23): 86-95.
［18］　左昊轩，黄祺成，杨佳昊, 等. 基于双目视觉和改进YOLOv8的玉米茎秆宽度原位识别方法［J］. 智慧农业(中英文)， 2023, 5(3): 165-174.
Zuo Haoxuan, Huang Qicheng, Yang Jiahao, et al. In situ identification method of maize stalk width based on improved YOLOv8 ［J］. Smart Agriculture， 2023, 5(3): 165-174.
［19］　李志臣, 凌秀军, 李鸿秋, 等. 基于改进ShuffleNet的板栗分级方法［J］. 山东农业大学学报(自然科学版), 2023, 54(2): 299-307.
Li Zhichen, Ling Xiujun, Li Hongqiu, et al. Chestnut grading method based on improved ShuffleNet ［J］. Journal of Shandong Agricultural University (Natural Science Edition), 2023, 54(2): 299-307.
［20］　张林鍹，巴音塔娜，曾庆松. 基于StyleGAN2-ADA和改进YOLOv7的葡萄叶片早期病害检测方法［J］. 农业机械学报, 2024, 55(1): 241-252.
Zhang Linxuan, Ba Yintana, Zeng Qingsong. Grape disease detection algorithm based on StyleGAN2-ADA and improved YOLOv7 ［J］. Transactions of the Chinese Society for Agricultural Machinery, 2024, 55(1): 241-252.
［21］　李斌，雷钧涵，郭毅. 基于CCA和Transformer的YOLOv8船舶目标检测算法［J］. 控制工程, 2024, 31(5): 901-911.
Li Bin, Lei Junhan, Guo Yi. YOLOv8 ship target detection algorithm based on CCA and Transformer ［J］. Control Engineering of China， 2024, 31(5): 901-911.
［22］　彭红星, 徐慧明, 高宗梅, 等. 基于改进YOLOF模型的田间农作物害虫检测方法［J］. 农业机械学报, 2023, 54(4): 285-294, 303.
Peng Hongxing, Xu Huiming, Gao Zongmei, et al. Insectpest detection of field crops based on improved YOLOF model ［J］. Transactions of the Chinese Society for Agricultural Machinery, 2023, 54(4): 285-294, 303.
［23］　苗荣慧, 李志伟, 武锦龙. 基于改进YOLO v7的轻量化樱桃番茄成熟度检测方法［J］. 农业机械学报, 2023, 54(10): 225-233.
Miao Ronghui, Li Zhiwei, Wu Jinlong. Lightweight maturity detection of cherry tomato based on improved YOLOv7 ［J］.Transactions of the Chinese Society for Agricultural Machinery, 2023, 54(10): 225-233.
［24］　倪云峰，霍洁，侯颖, 等. 基于YOLOv8-OCR的井下人员检测算法［J］.无线电工程， 2024， 54（8): 1847-1853.
Ni Yunfeng, Huo Jie, Hou Ying, et al. Underground personnel detection algorithm based on YOLOv8-OCR ［J］. Radio Engineering， 2024， 54（8): 1847-1853.
［25］　张秀再，沈涛，许岱. 基于改进YOLOv8算法的遥感图像目标检测［J］. 激光与光电子学进展, 2024, 61(10): 298-308.
Zhang Xiuzai, Shen Tao, Xu Dai. Improved YOLOv8 algorithm for remote sensing image target detection ［J］. Laser & Optoelectronics Progress, 2024, 61(10): 298-308.
［26］　赵辉, 黄镖, 王红君, 等. 基于改进YOLOv7的农田复杂环境下害虫识别算法研究［J］. 农业机械学报, 2023, 54(10): 245-254.
Zhao Hui, Huang Biao, Wang Hongjun, et al. Pest identification method in complex farmland environment based on improved YOLOv7［J］. Transactions of the Chinese Society for Agricultural Machinery, 2023, 54(10): 245-254.

[1]	张传栋, 亓璐, 丁华立. 基于改进 YOLOv8n模型的多品种葡萄簇检测方法[J]. 中国农机化学报, 2024, 45(9): 220-226.
[2]	李邦国, 王辉, 宋杨, 任志伟, 刘跃华, 徐乐程. 基于无人驾驶小麦收割机立体视觉感知系统[J]. 中国农机化学报, 2024, 45(9): 244-249.
[3]	杨巧梅, 崔婷婷, 袁永榜, 罗桦 . 轻量化 YOLO模型在农作物微小病虫害检测中的应用研究[J]. 中国农机化学报, 2024, 45(9): 265-270.
[4]	赵晓霞, 程曼, 袁洪波. 基于StrongSORT算法的羊只多目标跟踪方法[[J]. 中国农机化学报, 2024, 45(8): 180-188.
[5]	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.
[6]	周思捷, 刘天奇, 陈天华. 基于改进YOLOv5算法的水稻病害识别研究[J]. 中国农机化学报, 2024, 45(8): 246-253.
[7]	岳耀华, 张伟, 亓立强, . 基于改进YOLOv5的田间大豆花朵生长状态检测方法研究[J]. 中国农机化学报, 2024, 45(7): 188-193.
[8]	张延军, 赵建鑫, . 基于多路神经网络的黄花菜关节点检测算法研究[J]. 中国农机化学报, 2024, 45(7): 228-234.
[9]	张晨一, 张晓乾, 任振辉. 基于改进YOLOv3的自然场景中未成熟青苹果识别研究[J]. 中国农机化学报, 2024, 45(7): 243-248.
[10]	薛霞, 刘鹏, 周文. 基于改进YOLOv8算法的实时细粒度植物病害检测[J]. 中国农机化学报, 2024, 45(5): 188-194.
[11]	罗鑫, 李加强, 何超. 基于改进YOLOv4的澳洲坚果视觉监测方法[J]. 中国农机化学报, 2024, 45(5): 217-222.
[12]	霍正瑞, 孙铁波. 基于改进YOLOv5算法的瓜蒌分级方法[J]. 中国农机化学报, 2024, 45(4): 100-107.
[13]	林燕翔, 沈印, 李光林. 基于改进InceptionV3算法的小麦杂质识别研究[J]. 中国农机化学报, 2024, 45(4): 108-116.
[14]	黄威, 刘义亭, , 李佩娟, 陈光明, . 基于改进YOLOXS的苹果成熟度检测方法[J]. 中国农机化学报, 2024, 45(3): 226-232.
[15]	李嘉诚, 陈中举, 许浩然. 基于改进YOLOv8n的真实场景下草莓病害检测方法[J]. 中国农机化学报, 2024, 45(12): 267-274.