桥梁钢构件损伤的红外图像识别研究

Infrared Image Recognition of Damage of Bridge Steel Members

本研究针对热波图像在采集过程中会存在的热场不均,设备自身红外热波辐射损失消耗严重,以及实验构件的内部材质分布不均等诸多不利因素的影响,导致后期热波图像序列存在非均匀性、背景噪声较大、目标部位对比度低、损伤或缺陷部位显示效果差等问题提出了一套完整的红外图像处理的改进算法,弥补现有方法不足,在热波检测原理的基础上利用MATLAB处理软件对红外热波图像序列进行增强、分割及重构。从关键性技术环节展开深入和系统的研究,提高红外热波图像序列处理的质量及速度,增强热波图像算法的实用性和稳定性。

This study is aimed at the uneven thermal field in the process of heat wave image acquisition,the serious loss of infrared heat wave radiation loss of the device itself,and the uneven distribution of internal materials of the experimental components,which leads to the late heat wave image. A series of improved infrared image processing algorithms are proposed to compensate for the inhomogeneity of the sequence,the background noise is large,the contrast of the target is low,and the damage of the damaged or defective parts is poor. Based on the MATLAB processing software,the infrared heat wave image sequence is enhanced,segmented and reconstructed. In-depth and systematic research from key technical aspects,improve the quality and speed of infrared thermal wave image sequence processing,and enhance the practicality and stability of thermal wave image algorithm.