[1]靳双燕*,李浩亮.doi: 10.3969/j.issn.1001-3849.2025.12.018锌基合金电镀层斑点缺陷多光谱成像检测方法[J].电镀与精饰,2025,(12):139-145.
　Jin Shuangyan*,Li Haoliang.Multi spectral imaging detection method for spot defects in zinc-based alloy electroplating coatings[J].Plating & Finishing,2025,(12):139-145.
点击复制

doi: 10.3969/j.issn.1001-3849.2025.12.018锌基合金电镀层斑点缺陷多光谱成像检测方法()

《电镀与精饰》[ISSN:1001-3849/CN:12-1096/TG]

卷:
期数:
2025年12
页码:
139-145
栏目:
出版日期:
2025-12-31

文章信息/Info

Title:
Multi spectral imaging detection method for spot defects in zinc-based alloy electroplating coatings
作者:
靳双燕1*; 李浩亮2
(1. 郑州工商学院 信息工程学院,河南 郑州 450000 ;2. 郑州大学 电气与信息工程学院,河南 郑州 45000)
Author(s):
Jin Shuangyan1*;  Li Haoliang2
(1. School of Information Engineering, Zhengzhou Technology and Business University, Zhengzhou 450000, China; 2. School of Electrical and Information Engineering, Zhengzhou University, Zhengzhou 450000, China)
关键词:
锌基合金电镀层斑点缺陷; 多光谱成像检测; 噪声数据; 光谱向量; 空间映射
Keywords:
spot defects in zinc-based alloy electroplating layer;  multi spectral imaging detection;  noise data;  spectral vector;  space mapping
分类号:
TN223；TQ153
文献标志码:
A
摘要:
由于电镀层表面复杂的光学特性，斑点缺陷的光谱向量与背景噪声存在多维耦合，导致传统方法难以有效分离光谱序列中的噪声数据。在噪声干扰下，斑点缺陷的尺度变化规律缺乏系统性表征，导致特征集构建不完整，无法覆盖多尺度缺陷形态，影响检测与定位的可靠性。为了解决这些问题，提出了一种基于多光谱成像的检测方法。通过分析多光谱成像数据的内在低秩特性，建立频谱间的噪声数据模型，输出正常与含噪信号的常量值，根据常量值减去光谱序列中每一行的对应数，完成去噪；在获得去噪数据后，针对锌基合金电镀层斑点的光谱向量与权重特征，采用多尺度融合求得斑点缺陷的尺度变化规律，整合形成全面特征集；基于这一规律构建全面特征集后，采用空间映射算法处理特征数据，生成反映缺陷特性的波频变化矩阵，通过分析矩阵参数锁定电镀层斑点缺陷在光谱维度的具体位置，实现缺陷的精准定位。实验数据表明，所提方法在信噪比、缺陷特征可视化及光谱匹配度3个关键指标上均显著优于对比方法，能够精准识别锌基合金电镀层中露铁斑、麻点斑等缺陷的轮廓、分布及光谱特征，且检测结果与实际值高度吻合。
Abstract:
Due to the complex optical properties of the surface of electroplated coatings, there is multidimensional coupling between the spectral vectors of spot defects and background noise, making it difficult for traditional methods to effectively separate noise data from spectral sequences. Under noise interference, the scale variation law of spot defects lacks systematic characterization, resulting in incomplete feature set construction and inability to cover multi-scale defect morphology, which affects the reliability of defect detection and localization results. To address these issues, a detection method based on multispectral imaging has been proposed. By analyzing the inherent low-rank characteristics of multispectral imaging data, a noise data model between spectra is established, and the constant values of normal and noisy signals are output. The noise is removed by subtracting the corresponding number in each row of the spectral sequence from the constant value. After obtaining the denoised data, the spectral vectors and weight features of the spots on the zinc-based alloy electroplating layer are analyzed. The multi-scale fusion method is adopted to obtain the scale variation law of the spot defects, and a comprehensive feature set is formed by integrating them. After constructing a comprehensive feature set based on this rule, the spatial mapping algorithm is used to process the feature data, generating a frequency variation matrix that reflects the defect characteristics. By analyzing the parameters of the matrix, the specific position of the spot defect in the electroplating layer in the spectral dimension is locked, achieving precise defect location. Experimental data shows that the proposed method is significantly superior to the comparative method in three key indicators: signal-to-noise ratio, defect feature visualization, and spectral matching. It can accurately identify the contour, distribution, and spectral characteristics of defects such as iron spots and pitting spots in zinc-based alloy electroplating coatings, and the detection results are highly consistent with the actual values

参考文献/References:

[1].Mi J R, Li Q W, Li B Y. Efficient direct x-ray detection and imaging based on a lead-free electron donor-acceptor MOF[J]. ACS applied materials & interfaces, 2024, 16(7): 9002-9011.
[2].黄志海, 邓耀华, 吴光栋. IC器件表面缺陷多光谱图像特征融合检测方法[J]. 仪器仪表学报, 2024, 45(9): 24-33.
[3].张毅, 范玉刚. 基于深度学习与域自适应的工件涡流热成像的缺陷检测[J]. 红外技术, 2024, 46(3): 347-353.
[4].化春键, 孙明春, 蒋毅, 等. 基于改进YOLOv7-tiny的多光谱苹果表层缺陷检测[J]. 激光与光电子学进展, 2024, 61(10): 236-244.
[5].张维召, 王静. 汽车后门外板电泳前基板团斑缺陷分析[J]. 电镀与涂饰, 2023, 42(8): 41-44.
[6].王麟博, 蔡长龙, 梁海锋, 等. 采用真空电弧离子镀技术制备厚铬膜的机械特性研究[J]. 光学与光电技术, 2023, 21(1): 64-71.
[7].胡广华, 涂千禧. 基于光度立体和双流特征融合网络的工业产品表面缺陷检测方法[J]. 华南理工大学学报(自然科学版), 2024, 52(10): 112-123.
[8].张斌, 郭宇会, 杨小平, 等. 辊径对Cu-Fe-P铜合金轧制带材表面微观缺陷及性能的影响[J]. 锻压技术, 2024, 49(9): 122-128.
[9].王楠, 李震, 李佳盟, 等. 融合多光谱成像与深度学习的作物植株叶绿素检测系统研究[J]. 农业机械学报, 2023, 54(S2): 260-269.
[10].刘俊泽,贺英彩,马秀琴,等.不同燕麦品种种子自然老化表型和多光谱变化特征[J]. 草地学报, 2024, 32(12): 3688-3697.
[11].方靖岳, 文之豪, 朱海涛, 等. 基于集成法布里-珀罗微腔阵列的16通道快照式多光谱成像[J]. 物理学报, 2024, 73(7): 164-173.
[12].狄荣育, 马晓珊, 李广运, 等. 空间目标多光谱成像仿真[J]. 计算机仿真, 2023, 40(8): 36-40, 397.
[13].李敏, 王慧琴, 王可, 等. 基于卷积神经网络的城墙多光谱成像病害检测方法[J]. 激光与光电子学进展, 2024, 61(4): 542-550.
[14].葛宏义, 郭春燕, 蒋玉英, 等. 基于可见光和先进成像技术的粮食不完善粒快速检测研究进展[J]. 食品工业科技, 2025, 46(6): 30-42.
[15].魏松岩, 孟祥强, 衣晓宾, 等. “一带一路”区域国产中高分辨率宽波段多光谱卫星数据获取技术研究[J]. 遥感技术与应用, 2023, 38(4): 776-782.
[16].吴鑫, 徐宝腾, 刘家林, 等. 基于RGB-NIR滤波阵列的引入权重系数的加权引导滤波去马赛克方法[J]. 光子学报, 2024, 53(9): 47-57.
[17].李广一, 朱涛, 杨欢红, 等. 火电厂区域分布式光伏支架表面锌铝合金镀层的腐蚀行为[J]. 电镀与涂饰, 2023, 42(4): 1-5.
[18].张俊林, 黄宁, 李冬晓, 等. 单电源VPPA-GTA双面穿孔焊缺陷机理分析与工艺优化[J]. 热加工工艺, 2023, 52(1): 147-151, 156.
[19].姜修成, 张正加, 王猛猛, 等. 基于特征值分解和自适应滤波的极化相位优化方法[J]. 遥感学报, 2024, 28(3): 781-792.
[20].游大朋, 杨静, 张露, 等. Re-YOLOv5: 一种基于结构重参数化的钢材缺陷检测方法[J]. 测控技术, 2024, 43(3): 9-21.

更新日期/Last Update: 2025-12-18