改进轻量化的FCM-YOLOv8n钢材表面缺陷检测  

Improving the lightweight FCM-YOLOv8n for steel surface defect detection

作　　者：梁礼明 陈康泉 陈林俊 龙鹏威 Liang Liming;Chen Kangquan;Chen Linjun;Long Pengwei(School of Electrical Engineering and Automation,Jiangxi University of Science and Technology,Ganzhou,Jiangxi 341000,China)

机构地区：[1]江西理工大学电气工程与自动化学院,江西赣州341000

出　　处：《光电工程》2025年第2期108-119,共12页Opto-Electronic Engineering

基　　金：国家自然科学基金资助项目(51365017,61463018);江西省自然科学基金资助项目(20192BAB205084);江西省教育厅科学技术研究青年项目(GJJ2200848)。

摘　　要：针对现有钢材表面缺陷检测算法在资源消耗、检测精度和效率等方面存在的不足,提出一种基于YOLOv8n的轻量级钢材缺陷检测算法(FCM-YOLOv8n)。该方法一是采用频率感知特征融合网络,高效提取并融合高频信息,以降低计算成本并提升检测速度;二是重构轻量化特征交互模块(Cc-C2f),有效保留空间和通道依赖关系,减少特征冗余,以降低模型参数量和计算复杂度;三是利用多谱注意力机制,从频域维度减少特征信息缺失,以提升复杂缺陷的识别准确度。在Severstal和NEU-DET钢材缺陷数据集上的实验结果表明,相较于YOLOv8n算法,FCMYOLOv8n算法的mAP@0.5分别提高2.2%和1.5%;参数量和复杂度分别降低0.5 M和1.5 G;FPS分别达到143 f/s和154 f/s,展示优异的实时性。该算法在检测精度、计算成本和效率之间实现良好的平衡,为边缘终端设备应用提供有力的支持。在GC10-DET数据集上的进一步验证表明,FCM-YOLOv8n相较于基线模型mAP@0.5提升2.9%,充分佐证其卓越的泛化能力。In response to the deficiencies of existing steel surface defect detection algorithms in terms of resource consumption,detection accuracy,and efficiency,a lightweight steel defect detection algorithm based on YOLOv8n(FCM-YOLOv8n)is proposed.First,a frequency-aware feature fusion network is utilized to efficiently extract and integrate high-frequency information,reducing computational costs while enhancing detection speed.Second,a lightweight feature interaction module(Cc-C2f)is restructured to effectively preserve spatial and channel dependencies while reducing feature redundancy,thereby lowering model parameters and computational complexity.Finally,a multi-spectrum attention mechanism is applied to mitigate feature information loss in the frequency domain,improving the accuracy of detecting complex defects.Experimental results on the Severstal and NEU-DET steel defect datasets show that,compared to YOLOv8n,the FCM-YOLOv8n algorithm achieves a 2.2%and 1.5%improvement in mAP@0.5,respectively,with a 0.5 M and 1.5 G reduction in parameters and computational complexity.The FPS reaches 143 f/s and 154 f/s,respectively,demonstrating excellent real-time performance.The algorithm achieves an optimal balance between detection accuracy,computational cost,and efficiency,providing robust support for edge device applications.Further validation on the GC10-DET dataset shows a 2.9%improvement in mAP@0.5 compared to the baseline model,fully demonstrating the algorithm's exceptional generalization ability.

关 键 词：缺陷检测 YOLOv8n 频率感知特征融合网络 Cc-C2f 多谱注意力 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]