秦乾坤 杨慧 岳威 李宝磊



摘 要: 针对红外热波无损检测技术在应用过程中红外图像信噪比较低、细节不明了的问题,分析了小波阈值降噪方法的理论原理和实现步骤,针对传统的小波硬阈值、软阈值函数在去噪中的不足之处,提出了改进的阈值函数。改进的阈值函数通过引入调节因子,改善了硬阈值去噪在小波系数重构时可能会产生振荡的现象,软阈值去噪函数存在固定差异问题,且具有一定的灵活性和自适应性。实验结果表明:采用改进阈值去噪方法在有效的减少图像噪声的同时,能保留更多图像细节信息,去噪后的红外图像具有更高的信噪比。
关键词: 图像去噪; 红外热波无损检测; 红外热图; 小波阈值去噪
中图分类号:TP751.1 文献标志码:A 文章编号:1006-8228(2019)01-75-04
Abstract: In view of the low signal-to-noise ratio of infrared image in the application process, the theoretical principle and implementation steps of wavelet threshold de-noising method are analyzed. The traditional wavelet hard threshold and soft threshold are analyzed. An improved threshold function is proposed for the inadequacy of functions in de-noising. The improved threshold function improves the phenomenon that the hard threshold de-noising may produce oscillation when the wavelet coefficients are reconstructed by introducing the adjustment factor, and the soft threshold de-noising function has a fixed difference problem, and has certain flexibility and adaptability. The experimental results show that the improved threshold de-noising method can effectively reduce image noise while retaining more image detail information and the de-noised infrared image has higher signal-to-noise ratio.
Key words: image de-noising; infrared thermal wave nondestructive testing; infrared thermal image; wavelet threshold de-noising
0 引言
今年我国汽车、铁路、船舶、航空航天等现代工业快速发展,为了保证产品质量、安全生产,提高产品在使用过程中可靠性,无损检测技术越来越受到相关学者和工程技术人员的重视。其中红外热波无损检测技术(Infrared Thermography NDT)是一种新型的无损检测技术[1],其多采用主动热源对试件表面加热,在热流向试件内部传导的过程中,通过红外热像仪采集试件表面的温度变化信息,将结构中的损伤信息通过热图序列的形式表现出来。其具有检测速度快、单次检测面积大、和试件不接触、检测结果以图像形式直观显示等优点,目前已普遍应用于轨道交通、航天航空、国防军工和新能源等领域[2]。……