College of Information and Electrical Engineering/Key Laboratory of Standardization of Agricultural Informatization,Ministry of Agriculture and Rural Affairs/National Digital Fishery Innovation Center of the Ministry of Agriculture and Rural Affairs/Key Laboratory of Smart Breeding Technology,Ministry of Agriculture and Rural Affairs/Beijing Engineering and Technology Research Center for Internet of Things in Agriculture/National Digital Agricultural Product Circulation Innovation Sub center of the Ministry of Agriculture and Rural Affairs,China Agricultural University,Beijing 100083,China
To address the low efficiency and high subjectivity of manual ripeness assessment in wolfberry harvesting, as well as the high missed-detection rate and insufficient adaptability to varying Intersection over Union (IoU) thresholds of mainstream object detection algorithms in complex field environments, this study proposes an improved intelligent recognition method for wolfberry fruit maturity based on the YOLOv11 network. Focusing on specific challenges such as the visual similarity between immature fruits and leaves, class imbalance, and localization deviation due to fruit occlusion, three key modifications were implemented in the network architecture: the Convolutional Block Attention Module (CBAM) was embedded into the backbone network to enhance the model’s ability to extract discriminative fruit features; Focal Loss was adopted to replace the original classification loss, mitigating the impact of class imbalance on classification performance; and the EIoU loss function was introduced to substitute the original regression loss, improving bounding box localization accuracy under occlusion conditions. To validate the method’s effectiveness, a wolfberry field image dataset encompassing various illumination and occlusion levels was constructed, and data augmentation techniques were employed to enhance model generalization. Using precision (P), recall (R), mAP50, and mAP50-95 as evaluation metrics(where mAP50 denotes the mean average precision at IoU=0.50, and mAP50-95 represents the average precision over IoU thresholds of 0.50 to 0.95), the improved model, termed CEF-YOLOv11, was compared with YOLOv11, YOLOv8-s, YOLOv10-s, SSD, Faster R-CNN, YOLOv5n, and YOLOv5s. The results demonstrate that: 1) The improved model achieved the optimal recall of 90.8%, which is 4.7, 3.4, 4.9, 16.9, and 14.8 percentage points higher than YOLOv11, YOLOv8-s, YOLOv10-s, YOLOv5n, and YOLOv5s, respectively, significantly reducing the missed detection rate of mature fruits; 2) The model attained an mAP50 of 90.9%, indicating that the introduced attention mechanism and optimized loss functions synergistically enhance both classification and localization performance; 3) An mAP50-95 of 76.0% was achieved, representing a 9.5% improvement over the original YOLOv11, confirming the model's good robustness across different IoU thresholds. The proposed CEF-YOLOv11 algorithm can effectively improve the detection accuracy of ripe wolfberries in complex backgrounds, providing a technical basis for developing vision systems in picking robots.
[Author(id=1302637347968802864, tenantId=1045748351789510663, journalId=1250836471247831193, articleId=1302637346005869381, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=0009-0001-2262-6528, stid=null, country=null, authorPic=null, dead=0, email=gongdajike2101@163.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1302637348027523130, tenantId=1045748351789510663, journalId=1250836471247831193, articleId=1302637346005869381, authorId=1302637347968802864, language=EN, stringName=Ziang MAO, firstName=Ziang, middleName=null, lastName=MAO, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=College of Information and Electrical Engineering/Key Laboratory of Standardization of Agricultural Informatization,Ministry of Agriculture and Rural Affairs/National Digital Fishery Innovation Center of the Ministry of Agriculture and Rural Affairs/Key Laboratory of Smart Breeding Technology,Ministry of Agriculture and Rural Affairs/Beijing Engineering and Technology Research Center for Internet of Things in Agriculture/National Digital Agricultural Product Circulation Innovation Sub center of the Ministry of Agriculture and Rural Affairs,China Agricultural University,Beijing 100083,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1302637348073660479, tenantId=1045748351789510663, journalId=1250836471247831193, articleId=1302637346005869381, authorId=1302637347968802864, language=CN, stringName=毛子昂, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=中国农业大学 信息与电气工程学院/农业农村部农业信息化标准化重点实验室/农业农村部国家数字渔业 创新中心/农业农村部智慧养殖技术重点实验室/北京市农业物联网工程技术研究中心/ 农业农村部国家数字农产品流通创新分中心,北京 100083, bio={"img":"aTUBcHuatwYrd/1Hd7MoWg==","content":"
毛子昂,工学学士,中国农业大学计算机技术专业硕士在读。
"}, bioImg=aTUBcHuatwYrd/1Hd7MoWg==, bioContent=
毛子昂,工学学士,中国农业大学计算机技术专业硕士在读。
, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1302637347897499684, tenantId=1045748351789510663, journalId=1250836471247831193, articleId=1302637346005869381, xref=null, ext=[AuthorCompanyExt(id=1302637347910082598, tenantId=1045748351789510663, journalId=1250836471247831193, articleId=1302637346005869381, companyId=1302637347897499684, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=College of Information and Electrical Engineering/Key Laboratory of Standardization of Agricultural Informatization,Ministry of Agriculture and Rural Affairs/National Digital Fishery Innovation Center of the Ministry of Agriculture and Rural Affairs/Key Laboratory of Smart Breeding Technology,Ministry of Agriculture and Rural Affairs/Beijing Engineering and Technology Research Center for Internet of Things in Agriculture/National Digital Agricultural Product Circulation Innovation Sub center of the Ministry of Agriculture and Rural Affairs,China Agricultural University,Beijing 100083,China), AuthorCompanyExt(id=1302637347926859816, tenantId=1045748351789510663, journalId=1250836471247831193, articleId=1302637346005869381, companyId=1302637347897499684, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=中国农业大学 信息与电气工程学院/农业农村部农业信息化标准化重点实验室/农业农村部国家数字渔业 创新中心/农业农村部智慧养殖技术重点实验室/北京市农业物联网工程技术研究中心/ 农业农村部国家数字农产品流通创新分中心,北京 100083)])]), Author(id=1302637348119797829, tenantId=1045748351789510663, journalId=1250836471247831193, articleId=1302637346005869381, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1302637348174323788, tenantId=1045748351789510663, journalId=1250836471247831193, articleId=1302637346005869381, authorId=1302637348119797829, language=EN, stringName=Binlong WANG, firstName=Binlong, middleName=null, lastName=WANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=College of Information and Electrical Engineering/Key Laboratory of Standardization of Agricultural Informatization,Ministry of Agriculture and Rural Affairs/National Digital Fishery Innovation Center of the Ministry of Agriculture and Rural Affairs/Key Laboratory of Smart Breeding Technology,Ministry of Agriculture and Rural Affairs/Beijing Engineering and Technology Research Center for Internet of Things in Agriculture/National Digital Agricultural Product Circulation Innovation Sub center of the Ministry of Agriculture and Rural Affairs,China Agricultural University,Beijing 100083,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1302637348220461137, tenantId=1045748351789510663, journalId=1250836471247831193, articleId=1302637346005869381, authorId=1302637348119797829, language=CN, stringName=王彬茏, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=中国农业大学 信息与电气工程学院/农业农村部农业信息化标准化重点实验室/农业农村部国家数字渔业 创新中心/农业农村部智慧养殖技术重点实验室/北京市农业物联网工程技术研究中心/ 农业农村部国家数字农产品流通创新分中心,北京 100083, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1302637347897499684, tenantId=1045748351789510663, journalId=1250836471247831193, articleId=1302637346005869381, xref=null, ext=[AuthorCompanyExt(id=1302637347910082598, tenantId=1045748351789510663, journalId=1250836471247831193, articleId=1302637346005869381, companyId=1302637347897499684, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=College of Information and Electrical Engineering/Key Laboratory of Standardization of Agricultural Informatization,Ministry of Agriculture and Rural Affairs/National Digital Fishery Innovation Center of the Ministry of Agriculture and Rural Affairs/Key Laboratory of Smart Breeding Technology,Ministry of Agriculture and Rural Affairs/Beijing Engineering and Technology Research Center for Internet of Things in Agriculture/National Digital Agricultural Product Circulation Innovation Sub center of the Ministry of Agriculture and Rural Affairs,China Agricultural University,Beijing 100083,China), AuthorCompanyExt(id=1302637347926859816, tenantId=1045748351789510663, journalId=1250836471247831193, articleId=1302637346005869381, companyId=1302637347897499684, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=中国农业大学 信息与电气工程学院/农业农村部农业信息化标准化重点实验室/农业农村部国家数字渔业 创新中心/农业农村部智慧养殖技术重点实验室/北京市农业物联网工程技术研究中心/ 农业农村部国家数字农产品流通创新分中心,北京 100083)])]), Author(id=1302637348262404182, tenantId=1045748351789510663, journalId=1250836471247831193, articleId=1302637346005869381, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1302637348321124446, tenantId=1045748351789510663, journalId=1250836471247831193, articleId=1302637346005869381, authorId=1302637348262404182, language=EN, stringName=Hanyu ZHANG, firstName=Hanyu, middleName=null, lastName=ZHANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=College of Information and Electrical Engineering/Key Laboratory of Standardization of Agricultural Informatization,Ministry of Agriculture and Rural Affairs/National Digital Fishery Innovation Center of the Ministry of Agriculture and Rural Affairs/Key Laboratory of Smart Breeding Technology,Ministry of Agriculture and Rural Affairs/Beijing Engineering and Technology Research Center for Internet of Things in Agriculture/National Digital Agricultural Product Circulation Innovation Sub center of the Ministry of Agriculture and Rural Affairs,China Agricultural University,Beijing 100083,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1302637348363067491, tenantId=1045748351789510663, journalId=1250836471247831193, articleId=1302637346005869381, authorId=1302637348262404182, language=CN, stringName=张涵钰, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=中国农业大学 信息与电气工程学院/农业农村部农业信息化标准化重点实验室/农业农村部国家数字渔业 创新中心/农业农村部智慧养殖技术重点实验室/北京市农业物联网工程技术研究中心/ 农业农村部国家数字农产品流通创新分中心,北京 100083, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1302637347897499684, tenantId=1045748351789510663, journalId=1250836471247831193, articleId=1302637346005869381, xref=null, ext=[AuthorCompanyExt(id=1302637347910082598, tenantId=1045748351789510663, journalId=1250836471247831193, articleId=1302637346005869381, companyId=1302637347897499684, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=College of Information and Electrical Engineering/Key Laboratory of Standardization of Agricultural Informatization,Ministry of Agriculture and Rural Affairs/National Digital Fishery Innovation Center of the Ministry of Agriculture and Rural Affairs/Key Laboratory of Smart Breeding Technology,Ministry of Agriculture and Rural Affairs/Beijing Engineering and Technology Research Center for Internet of Things in Agriculture/National Digital Agricultural Product Circulation Innovation Sub center of the Ministry of Agriculture and Rural Affairs,China Agricultural University,Beijing 100083,China), AuthorCompanyExt(id=1302637347926859816, tenantId=1045748351789510663, journalId=1250836471247831193, articleId=1302637346005869381, companyId=1302637347897499684, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=中国农业大学 信息与电气工程学院/农业农村部农业信息化标准化重点实验室/农业农村部国家数字渔业 创新中心/农业农村部智慧养殖技术重点实验室/北京市农业物联网工程技术研究中心/ 农业农村部国家数字农产品流通创新分中心,北京 100083)])]), Author(id=1302637348405010539, tenantId=1045748351789510663, journalId=1250836471247831193, articleId=1302637346005869381, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=0000-0003-2914-1192, stid=null, country=null, authorPic=null, dead=0, email=lizb@cau.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1302637348463730805, tenantId=1045748351789510663, journalId=1250836471247831193, articleId=1302637346005869381, authorId=1302637348405010539, language=EN, stringName=Zhenbo LI, firstName=Zhenbo, middleName=null, lastName=LI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=College of Information and Electrical Engineering/Key Laboratory of Standardization of Agricultural Informatization,Ministry of Agriculture and Rural Affairs/National Digital Fishery Innovation Center of the Ministry of Agriculture and Rural Affairs/Key Laboratory of Smart Breeding Technology,Ministry of Agriculture and Rural Affairs/Beijing Engineering and Technology Research Center for Internet of Things in Agriculture/National Digital Agricultural Product Circulation Innovation Sub center of the Ministry of Agriculture and Rural Affairs,China Agricultural University,Beijing 100083,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1302637348505673851, tenantId=1045748351789510663, journalId=1250836471247831193, articleId=1302637346005869381, authorId=1302637348405010539, language=CN, stringName=李振波, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=中国农业大学 信息与电气工程学院/农业农村部农业信息化标准化重点实验室/农业农村部国家数字渔业 创新中心/农业农村部智慧养殖技术重点实验室/北京市农业物联网工程技术研究中心/ 农业农村部国家数字农产品流通创新分中心,北京 100083, bio={"img":"VWDLAMnYZRs8ZyqwX7P54A==","content":"
, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1302637347897499684, tenantId=1045748351789510663, journalId=1250836471247831193, articleId=1302637346005869381, xref=null, ext=[AuthorCompanyExt(id=1302637347910082598, tenantId=1045748351789510663, journalId=1250836471247831193, articleId=1302637346005869381, companyId=1302637347897499684, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=College of Information and Electrical Engineering/Key Laboratory of Standardization of Agricultural Informatization,Ministry of Agriculture and Rural Affairs/National Digital Fishery Innovation Center of the Ministry of Agriculture and Rural Affairs/Key Laboratory of Smart Breeding Technology,Ministry of Agriculture and Rural Affairs/Beijing Engineering and Technology Research Center for Internet of Things in Agriculture/National Digital Agricultural Product Circulation Innovation Sub center of the Ministry of Agriculture and Rural Affairs,China Agricultural University,Beijing 100083,China), AuthorCompanyExt(id=1302637347926859816, tenantId=1045748351789510663, journalId=1250836471247831193, articleId=1302637346005869381, companyId=1302637347897499684, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=中国农业大学 信息与电气工程学院/农业农村部农业信息化标准化重点实验室/农业农村部国家数字渔业 创新中心/农业农村部智慧养殖技术重点实验室/北京市农业物联网工程技术研究中心/ 农业农村部国家数字农产品流通创新分中心,北京 100083)])])]
毛子昂,王彬茏,张涵钰,李振波.
基于CEF-YOLOv11的智能采摘枸杞成熟度识别方法研究[J].
中国农业大学学报, 2026, 31(8): 172-182 DOI:10.11841/j.issn.1007-4333.2026.08.15
采用P、R以及mAP作为评价指标,对模型的检测性能进行评估。其中,mAP50表示在交并比(Intersection over union,IoU)阈值为0.50时的平均检测精度,mAP50-95表示IoU阈值分别从0.50至0.95的平均精度,用于衡量模型在不同重叠度条件下的目标检测能力。各指标计算公式如下:
RenH, WangJ L. Current situation and promotion path of Chinese wolfberry industry development[J]. Science & technology for development, 2019, 15(3): 310-317 (in Chinese)
Relevant departments of the national forestry and grassland administration. Advancing innovation and improving quality: high-quality development of China’s modern wolfberry industry in 2023 [EB/OL].[--].
[5]
ZhaoJ, ChenJ. Detecting maturity in fresh Lycium barbarum L. fruit using color information[J]. Horticulturae, 2021, 7(5): 108
KangC, ZhangK Z, CheJ. Online identification of wolfberry red fruit based on YOLOv5[J]. Journal of Ningxia Agriculture and Forestry Science and Technology, 2022, 63(6): 21-26 (in Chinese)
ZhangX R, WangX Z, ZhouK Z, MengH W, PengH J, LiY P. Daily behaviour detection of multi-target dairy cows based on improved YOLO11n in complex environment[J]. Transactions of the Chinese Society of Agricultural Engineering, 2025, 41(14): 155-164 (in Chinese)
MoH H, WeiL J. Improved YOLOv7 based apple target detection in complex environment[J]. Journal of Zhejiang University : Engineering Science, 2024, 58(12): 2447-2458 (in Chinese)
YeD P, JingJ, ZhangZ D, LiH H, WuH Y, XieL M. MSH-YOLOv8: Mushroom small object detection method with scale reconstruction and fusion [J]. Smart Agriculture: Chinese and English, 2024, 6(5): 139-152 (in Chinese)
[18]
ÖnlerE, KöycüN D. Wheat powdery mildew detection with YOLOv8 object detection model[J]. Applied Sciences, 2024, 14(16): 7073
[19]
ZhangJ, YangW Z, LuZ F, ChenD. HR-YOLOv8: A crop growth status object detection method based on YOLOv8 [J]. Electronics, 2024, 13(9): 1620
[20]
LiuQ, LvJ, ZhangC P. MAE-YOLOv8-based small object detection of green crisp plum in real complex orchard environments [J]. Computers and Electronics in Agriculture, 2024, 226: 109458
ZhangR H, BaiX, FanJ C. Crop pest target detection algorithm in complex scenes: YOLOv8-extend [J]. Smart Agriculture: Chinese and English, 2024, 6(2): 49-61 (in Chinese)
LiH B, TianX, RuanZ W, LiuS W, RenW Q, SuZ B, GaoR, KongQ M. Seedling stage corn line detection method based on improved YOLOv8[J]. Smart Agriculture: Chinese and English, 2024, 6(6): 72-84 (in Chinese)
[25]
FangC, YangX. Lightweight YOLOv8 for wheat head detection[J]. IEEE Access, 2024, 12: 66214-66222
[26]
YuY, ZhouQ, WangH, LvK, ZhangL J, LiJ, LiD M. LP-YOLO: A lightweight object detection network regarding insect pests for mobile terminal devices based on improved YOLOv8 [J]. Agriculture, 2024, 14(8): 1420
[27]
GuoA, SunK Q, ZhangZ Y. A lightweight YOLOv8 integrating FasterNet for real-time underwater object detection[J]. Journal of Real-Time Image Processing, 2024, 21(2): 49
WangJ H, JiangM D, AbbasT, ChenH, JiangY Y. YOLOv-MA: A high-precision foreign object detection algorithm for rice [J]. Agriculture, 2025, 15(13): 1354
[30]
SapkotaR, CheppallyR H, ShardaA, KarkeeM. YOLO26: Key architectural enhancements and performance benchmarking for real-time object detection [EB/OL].[--].
[31]
BadgujarC M, PouloseA, GanH. Agricultural object detection with You Only Look Once (YOLO) Algorithm: A bibliometric and systematic literature review [J]. Computers and Electronics in Agriculture, 2024, 223: 109090
[32]
KhanamR, HussainM. YOLOv11: An overview of the key architectural enhancements[EB/OL].[--].
[33]
BadekaE, KarapatzakE, KarampateaA, BouloumpasiE, KalathasI, LytridisC, TziolasE, TsakalidouV N, KaburlasosV G. A deep learning approach for precision viticulture, assessing grape maturity via YOLOv7[J]. Sensors, 2023, 23(19): 8126
[34]
AldabbaghL, ChangY J, ChungY C. Chili plant growth stage classification using deep learning[J]. Computers and Electronics in Agriculture, 2022, 199: 107194
[35]
SaadM, AbdullahH, RahimiM, MohdA S, HassanM A. Chili object detection with R-CNN model in natural field conditions[J]. Journal of Food Agriculture and Environment, 2021, 19(3): 54-60
[36]
KhanamR, HussainM. YOLOv11: An overview of the key architectural enhancements[EB/OL].[--].
[37]
SaI, GeZ Y, DayoubF, UpcroftB, PerezT, McCoolC. DeepFruits: A fruit detection system using deep neural networks[J]. Sensors, 2016, 16(8): 1222
[38]
ZhangY F, RenW Q, ZhangZ, JiaZ, WangL, TanT. Focal and efficient IOU loss for accurate bounding box regression[J]. Neurocomputing, 2022, 506: 146-157
[39]
LinT Y, GoyalP, GirshickR, HeK M, DollarP. Focal loss for dense object detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020, 42(2): 318-327