基于YOLO11-SDP模型的饰面人造板表面缺陷检测研究

韩佳锴; 吕斌; 王晓欢; 张伟; 须恺; 刁兴良; 纪敏; 王禾

doi:10.12326/j.2096-9694.2025128

研究报告 | 浏览量 : 0 下载量: 12 CSCD: 0 CNKI被引量: 0 Crossref: 0 Wos: 0 Scopus: 0 更多指标

PDF
导出
分享
收藏
专辑

基于YOLO11-SDP模型的饰面人造板表面缺陷检测研究
Surface Defect Detection of Surface-Decorated Wood-Based Panels Based on YOLO11-SDP Model
1.中国林业科学研究院木材工业研究所，北京 100091
2.中国林业科学研究院木材工业研究所林草装备研究开发中心，北京 100091
3.苏州华翔木业机械有限公司，江苏苏州 215000
2026年40卷第2期页码：88-97
DOI： 10.12326/j.2096-9694.2025128
稿件说明：

移动端阅览

韩佳锴,吕斌,王晓欢等.基于YOLO11-SDP模型的饰面人造板表面缺陷检测研究[J].木材科学与技术,2026,40(02):88-97. DOI： 10.12326/j.2096-9694.2025128.

HAN Jiakai,LYU Bin,WANG Xiaohuan,et al.Surface Defect Detection of Surface-Decorated Wood-Based Panels Based on YOLO11-SDP Model[J].Chinese Journal of Wood Science and Technology,2026,40(02):88-97. DOI： 10.12326/j.2096-9694.2025128.

摘要

饰面人造板作为家居行业的核心原材料，其表面缺陷检测是质量管控的关键环节。当前基于机器视觉与YOLO系列深度学习检测方法已得到广泛应用，但在精确率、召回率等关键指标上仍有提升空间，且在设备资源受限场景下，难以兼顾推理速度与检测性能。为进一步优化模型综合性能，将其应用于饰面人造板表面缺陷检测场景，研究提出YOLO11-SDP检测模型，通过采用数据集Mosaic增强、BiFPN模块替换原有FPN模块、损失函数中引入尺度权重因子等3项组合策略对原有模型进行改进。试验结果表明，改进模型YOLO11-SDP精确率达98.0%、召回率达91.3%，较原始模型YOLO11s分别提升6.8%和6.1%。3项改进措施可有效增强模型对小目标的检测能力，提升模型泛化性与鲁棒性；在综合检测性能提升与资源消耗增加的权衡下，尺度权重因子的优化效果优于BiFPN模块，优于Mosaic增强。同时，YOLO11-SDP模型推理速度满足工业实时检测要求，可用于饰面人造板生产中干花、湿花、污斑、裂缝等常见缺陷的在线检测，为家居建材高质量发展提供技术支撑。

Abstract

Surface-decorated wood-based panels serve as a core raw material in the furniture industry

and surface defect detection is a critical step in quality control. Current detection methods commonly adopt YOLO-based machine vision

yet their performance in terms of precision and recall leave room for improvement. In resource-constrained deployment scenarios

the balance between inference speed and detection accuracy remains unresolved. To enhance the overall performance of existing detection models and apply the algorithm to surface defect detection of decorated wood-based panels

this study proposed an improved YOLO11-SDP (You Only Look Once11-Surface Decorated Panels) detection model. The original detection model was improved via three complementary measures: performing Mosaic augmentation on the dataset

replacing the original Feature Pyramid Network (FPN) module with a BiFPN module

and introducing a scale weight factor into the loss function. Experimental results showed that the improved model achieved a precision of 98.0% and a recall of 91.3%. Compared with the original model

the precision and recall rate were increased by 6.8% and 6.1%

respectively. These three enhancements significantly improved the model's capacity to detect small targets

yielding greater generalization and robustness. When balancing the performance gains against the increased computational cost

the scale weight factor was proved superior to BiFPN

followed by Mosaic augmentation. Furthermore

the detection speed satisfied real-time processing requirements. Consequently

the YOLO11-SDP model was well-suited for the online detection of common defects—including frosting mark

water mark

spots

and crack—in the manufacturing of surface decorated wood-based panels

thereby providing technical support for the high‑quality development of home furnishing and building materials.

关键词

Keywords

references

彭晓瑞 , 吕斌 , 李伟光 , 等 . 我国板式家具先进制造技术研发与应用进展 [J ] . 木材科学与技术 , 2023 , 37 ( 2 ): 1 - 7, 15 .

PENG X R , LYU B , LI W G , et al . Development and application progress of advanced manufacturing technology for China's panel furniture [J ] . Chinese Journal of Wood Science and Technology , 2023 , 37 ( 2 ): 1 - 7, 15 .

李博 , 吕斌 , 王兵 , 等 . 大数据技术在木材工业中的应用前景分析 [J ] . 中国人造板 , 2022 , 29 ( 6 ): 1 - 6 .

LI B , LYU B , WANG B , et al . Application prospect analysis of big data technology in wood industry [J ] . China Wood-Based Panels , 2022 , 29 ( 6 ): 1 - 6 .

张超 , 吴智慧 , 刘海波 . 机器视觉系统在家居木制品制造工业机器人中的应用 [J ] . 家具 , 2022 , 43 ( 4 ): 37 - 41, 50 .

ZHANG C , WU Z H , LIU H B . Application of machine vision system in the industrial robot for furnishing and wood products manufacturing [J ] . Furniture , 2022 , 43 ( 4 ): 37 - 41, 50 .

杨渊 . 机器视觉在刨花板表面缺陷检测中的应用现状与建议 [J ] . 中国人造板 , 2024 , 31 ( 5 ): 33 - 37 .

YANG Y . Application of machine vision in particle board surface defect detection: status and recommendations [J ] . China Wood-Based Panels , 2024 , 31 ( 5 ): 33 - 37 .

王雪 , 高雅 , 陶桂香 , 等 . 基于CBLP-YOLO 11n的无人机稻穗轻量化检测方法 [J ] . 农业机械学报 , 2025 , 56 ( 11 ): 461 - 470 .

WANG X , GAO Y , TAO G X , et al . Lightweigh detection method of rice panicles based on CBLP-YOLO 11n [J ] . Transactions of the Chinese Society for Agricultural Machinery , 2025 , 56 ( 11 ): 461 - 470 .

XU Y , WU H , LIU Y , et al . PCB Electronic component soldering defect detection using YOLO11 improved by retention block and neck structure [J ] . Sensors , 2025 , 25 ( 11 ). https://doi.org/10.3390/s25113550 https://doi.org/10.3390/s25113550 .

BAI J , ZHU W , NIE Z , et al . HFC-YOLO11:A lightweight model for the accurate recognition of tiny remote sensing targets [J ] . Computers , 2025 , 14 ( 5 ): 195 .

HE H L , ZHOU Z Y , LIU L , et al . Research on the directional bounding box algorithm of YOLO11 in tailings pond identification [J ] . Measurement , 2025 . DOI: 10.1016/j.measurement.2025.117674 http://dx.doi.org/10.1016/j.measurement.2025.117674 .

Ultralytics . YOLO11 [EB/OL ] . ( 2024-09-10 ). https://docs.ultralytics.com/zh/models/yolo11 https://docs.ultralytics.com/zh/models/yolo11 .

JIE C , JOO M E . Dynamic YOLO for small underwater object detection [J ] . Artificial Intelligence Review , 2024 , 57 ( 7 ). DOI: 10.1007/s10462-024-10788-1 http://dx.doi.org/10.1007/s10462-024-10788-1 .

王品博 , 王雨晨 , 王双永 , 等 . 基于改进YOLOv8的树木年轮实例分割及信息检测方法 [J ] . 木材科学与技术 , 2025 , 39 ( 3 ): 7 - 18 .

WANG P B , WANG Y C , WANG S Y , et al . Study on tree ring instance segmentation and information detection based on improved YOLOv8 [J ] . Chinese Journal of Wood Science and Technology , 2025 , 39 ( 3 ): 7 - 18 .

JI M , ZHANG W , WANG G , et al . Online measurement of outline size for Pinus densiflora dimension lumber: maximizing lumber recovery by minimizing enclosure rectangle fitting area [J ] . Forests , 2022 , 13 ( 10 ). https://doi.org/10.3390/f13101627 https://doi.org/10.3390/f13101627 .

乔澍祎 , 林秀仪 , 章伟伟 , 等 . 基于YOLOv5的人造板板材种类识别 [J ] . 林业工程学报 , 2023 , 8 ( 2 ): 138 - 143 .

QIAO S Y , LIN X Y , ZHANG W W , et al . Identification of the type of wood-based panels using YOLOv5 [J ] . Journal of Forestry Engineering , 2023 , 8 ( 2 ): 138 - 143 .

王勇 , 张伟 , 高锐 , 等 . 基于YOLOv4的结构用锯材表面缺陷识别 [J ] . 林业工程学报 , 2021 , 6 ( 4 ): 120 - 126 .

WANG Y , ZHANG W , GAO R , et al . Research on surface defect recognition of structural sawn timber using YOLOv4 [J ] . Journal of Forestry Engineering , 2021 , 6 ( 4 ): 120 - 126 .

惠永永 , 赵春雨 , 宋昭漾 , 等 . 基于BiFPN和Triplet注意力机制的YOLOv5s缺陷苹果识别算法 [J ] . 华南农业大学学报 , 2025 , 46 ( 3 ): 419 - 428 .

HUI Y Y , ZHAO C Y , SONG Z Y , et al . A YOLOv5s algorithm based on BiFPN and Triplet attention mechanism for identifing defective apple [J ] . Journal of South China Agricultural University , 2025 , 46 ( 3 ): 419 - 428 .

李文 , 叶坤涛 , 舒蕾蕾 , 等 . 基于高斯模糊逻辑和ADCSCM的红外与可见光图像融合算法 [J ] . 红外技术 , 2022 , 44 ( 7 ): 693 - 701 .

LI W , YE K T , SHU L L , et al . Infrared and visible image fusion algorithm based on gaussian fuzzy logic and adaptive dual-channel spiking cortical model [J ] . Infrared Technology , 2022 , 44 ( 7 ): 693 - 701 .

URBONAS A , RAUDONIS V , MASKELIŪNAS R , et al . Automated identification of wood veneer surface defects using faster region-based convolutional neural network with data augmentation and transfer learning [J ] . Applied Sciences , 2019 , 22 ( 9 ). https://doi.org/10.3390/app9224898 https://doi.org/10.3390/app9224898 .

ZHOU W J , LI Y , LIU L , et al . Cork classification based on multi-scale faster-RCNN with machine vision [J ] . Measurement , 2023 , 217 . DOI: 10.1016/j.measurement.2023.113089 http://dx.doi.org/10.1016/j.measurement.2023.113089 .

朱婷婷 , 李岩 , 杨茜 , 等 . 基于YOLOv9-DSConv桉树旋切单板缺陷识别方法 [J ] . 林业工程学报 , 2025 , 10 ( 5 ): 138 - 144 .

ZHU T T , LI Y , YANG Q , et al . Defect recognition method for eucalyptus rotary-cut veneer based on YOLOv9-DSConv [J ] . Journal of Forestry Engineering , 2025 , 10 ( 5 ): 138 - 144 .

江兴旺 , 赵兴强 . 改进YOLOv7的木材表面缺陷检测算法 [J ] . 计算机工程与应用 , 2024 , 60 ( 7 ): 175 - 182 .

JIANG X W , ZHAO X Q . Improved YOLOv7 algorithm for wood surface defect detection [J ] . Computer Engineering and Applications , 2024 , 60 ( 7 ): 175 - 182 .