[1]刘 婷*,刘成涛.doi: 10.3969/j.issn.1001-3849.2025.09.012红外热像技术下金属镀层表面缺陷无损检测方法[J].电镀与精饰,2025,(09):83-89.
　Liu Ting*,Liu Chengtao.Non-destructive testing method for surface defects of metal coatings using infrared thermal imaging technology[J].Plating & Finishing,2025,(09):83-89.
点击复制

doi: 10.3969/j.issn.1001-3849.2025.09.012红外热像技术下金属镀层表面缺陷无损检测方法()

《电镀与精饰》[ISSN:1001-3849/CN:12-1096/TG]

卷:
期数:
2025年09
页码:
83-89
栏目:
出版日期:
2025-09-30

文章信息/Info

Title:
Non-destructive testing method for surface defects of metal coatings using infrared thermal imaging technology
作者:
刘 婷1*; 刘成涛2
(1. 商丘职业技术学院 计算机工程学院,河南 商丘 476000 ;2. 西安工程大学 电子信息学院,陕西 西安 710048)
Author(s):
Liu Ting1*;  Liu Chengtao2
(1. Shangqiu Polytechnic, College of Computer Engineering, Shangqiu 476000, China; 2. Xian Polytechnic University, School of Electronics and Information, Xian 710048, China)
关键词:
红外热像技术; 金属镀层; 缺陷识别; 无损检测; 卷积神经网络
Keywords:
infrared thermal imaging technology;  metal coating;  defect identification;  non-destructive testing;  convolutional neural network
文献标志码:
A
摘要:
金属镀层具有复杂的形状和表面结构，非接触式检测可以更容易地适应这些复杂形状，实现全面、准确的检测。但当前非接触式检测方法主要基于图像的低级特征进行匹配和识别，缺乏对图像内容的深入理解和分析，难以捕捉图像中的高层语义信息，导致检测精度低。为精确高效地识别金属镀层表面缺陷，提出红外热像技术下金属镀层表面缺陷无损检测方法。采用红外测温热成像仪采集待检测目标红外图像样本，通过小波变换消除图像噪声，利用去除噪声后的图像样本创建红外图像温度场模型，观察镀层样本表面温度分布状况，推算缺陷部位和正常部位之间的温度差。通过卷积神经网络训练红外图像样本，提取金属镀层表面红外图像温度分布图中的高层语义信息，识别出金属镀层表面的缺陷类型，并求出预测值与真实值的均方误差，以均方误差最小为前提完成网络收敛，获得金属镀层表面缺陷无损检测结果。实验结果证明：所提方法缺陷精准度高，具有良好的适应性和可扩展性，能完成不同种类和规格的金属镀层表面缺陷无损检测任务。
Abstract:
Metal coatings have complex shapes and surface structures, and non-contact detection can more easily adapt to these complex shapes, achieving comprehensive and accurate detection. However, current non-contact detection methods mainly rely on low-level features of images for matching and recognition, lacking in-depth understanding and analysis of image content, making it difficult to capture high-level semantic information in images, resulting in low detection accuracy. To accurately and efficiently identify surface defects of metal coatings, a non-destructive testing method for surface defects of metal coatings using infrared thermography technology is proposed. Using an infrared thermography instrument to collect infrared image samples of the target to be detected, eliminating image noise through wavelet transform, creating an infrared image temperature field model using the denoised image samples, observing the temperature distribution on the coating surface, and calculating the temperature difference between the defective and normal areas. By training infrared image samples through convolutional neural networks, high-level semantic information in the temperature distribution map of the infrared image of the metal coating surface is captured to identify the defect types on the metal coating surface. The mean square error between the predicted value and the true value is calculated, and the network convergence is completed on the premise of minimizing the mean square error to obtain the non-destructive testing results of the metal coating surface defects. The experimental results demonstrate that the proposed method has high defect accuracy, good adaptability and scalability, and can complete non-destructive testing tasks for surface defects of different types and specifications of metal coatings

参考文献/References:

[1].张颖, 李雪, 刘彦伯, 等. 贵金属超薄镀层厚度的超声显微测量 [J]. 贵金属, 2022, 43(4): 56-63.
[2].代锋琪, 刘双杰, 郝永平, 等. 不同金属镀层单晶硅靶板的冲击受力特性 [J]. 探测与控制学报, 2024, 46(1): 114-120.
[3].李海鹏, 孙邦兴, 李嘉烨. 双碳目标下绿色制氢技术的进展 [J]. 电池, 2024, 54(2): 271-275.
[4].梁林, 郭昭蔚, 许俊, 等. 热应力对身管铬-钢结合面裂纹扩展的影响 [J]. 计算机仿真, 2020, 37(9): 6-10.
[5].师恬恬, 焦新生, 孔袁莉, 等. 卫星天线复合材料的模具金属镀层影响因素研究 [J]. 热加工工艺, 2024, 53(10): 37-41.
[6].江子航, 翟荃, 张继承, 等. 基于压电传感器的纤维陶粒混凝土冻融损伤检测 [J]. 科学技术与工程, 2024, 24(2): 576-583.
[7].张继阳, 郑秀, 赵斌, 等. 电网级大规模储能的电池技术进展 [J]. 电池, 2024, 54(5): 745-750.
[8].王单单, 高岩, 李伟仙, 等. 数字剪切散斑干涉玻璃幕墙缺陷无损检测 [J]. 中国测试, 2022, 48(4): 29-34.
[9].顾健, 任姣姣, 张丹丹, 等. 基于多次回波数值模拟的涂层缺陷太赫兹无损检测 [J]. 光学学报, 2022, 42(15): 111-118.
[10].徐杰, 刘畅. 基于改进ELM和计算机视觉的核桃缺陷检测 [J]. 食品与机械, 2024, 40(5): 122-127.
[11].Zhang H W, Chen X W, Lu S, et al. A contrastive learning-based attention generative adversarial network for defect detection in colour-patterned fabric [J]. Coloration technology, 2023, 139(3): 248-264.
[12].范福强, 邢素霞, 张俊举. 基于湍流模型的飞行器温度场数值仿真研究 [J]. 激光与红外, 2024, 54(5): 766-773.
[13].童志鹏, 邱志斌, 吴睿雯, 等. 基于多尺度模板匹配的配电线路劣化绝缘子红外热像检测 [J]. 红外技术, 2023, 45(12): 1337-1345.
[14].段进涛, 董毓利, 欧曙光, 等. 基于红外热成像技术烃类火灾下沉管隧道开裂分析 [J]. 中国安全生产科学技术, 2022, 18(8): 170-177.
[15].韩梦, 尹嘉雯, 黄军科, 等. 铝合金板疲劳微裂纹超声红外成像检测的数值及实验研究 [J]. 应用声学, 2022, 41(5): 727-734.
[16].孙晨, 王昕, 蒋国臻. 基于Levy-SOA自适应阈值分割和改进引导滤波的NSST图像增强 [J]. 控制工程, 2024, 31(7): 1297-1304.
[17].程杰. 基于红外图像的飞机主体合金温度检测 [J]. 兵器材料科学与工程, 2022, 45(1): 137-140.
[18].陈昭宇, 范洪博, 马美燕, 等. 基于结构重参数化的红外与可见光图像融合 [J]. 控制与决策, 2024, 39(7): 2275-2283.
[19].王洪君. 镀层性能检测技术研究 [J]. 冶金与材料, 2023, 43(10): 96-98.
[20].杨彬, 亚森江·木沙, 安波. 改进Mask RCNN的焊缝缺陷检测 [J]. 机械设计与制造, 2023(6): 157-161.
[21].唐茂俊, 黄海松, 张松松, 等. 改进的Faster-RCNN在焊缝缺陷检测中的应用 [J]. 组合机床与自动化加工技术, 2021(12): 83-86.

更新日期/Last Update: 2025-09-11