论文部分内容阅读
在红外热波无损检测中获取的热像序列存在着背景噪声大、缺陷边缘模糊、对比度低等特点。为了提高由红外热像序列重构的数字图像的缺陷显示能力,以小波变换为热像处理工具,采用基于像素级和特征级的图像融合算法对热像序列进行了处理,并采用基于统计学的图像评估标准对处理效果作了定量评价。通过对铝合金试件的检测实验说明该方法可用于材料内部缺陷的红外热波无损检测。研究结果表明,此种图像融合算法可对不同深度缺陷所对应的两幅最佳热像进行有效地融合,在一幅融合图像中直观地反映出全部缺陷,并能有效地减少加热不均和背景噪声对缺陷识别的不利影响。
The thermal imaging sequences acquired in the nondestructive testing of infrared thermal waves have the characteristics of large background noise, blurred edges and low contrast. In order to improve the defect display ability of the digital image reconstructed by infrared thermal imaging sequence, the thermal imaging sequence is processed by using image fusion algorithm based on pixel level and feature level based on wavelet transform and thermal image processing. Based on the statistics The image evaluation criteria for the treatment effect were quantitatively evaluated. Experiments on aluminum alloy specimens show that this method can be applied to the nondestructive testing of infrared thermal waves in materials with internal defects. The results show that this image fusion algorithm can effectively fuse the two best thermal images corresponding to different depth defects, visually reveal all the defects in a single fusion image, and effectively reduce the uneven heating and Adverse effects of background noise on defect recognition.