摘要
红外热图具有高噪声、低对比度的特点,设计了图像增强环节解决该问题。对红外热图的频段分布和频段来源进行了分析,确定了有效频段为检测对象红外特性带来的高频段,并确定了增强该频段能量为图像增强环节的目标。针对红外热图频段的分布特点,基于处理目标,使用灰度值调整和巴特沃斯滤波对图像进行了处理,通过图像显示和大律法分割图像两种方法来检验图像增强效果。结果表明:经过设计增强环节,图像的对比度增强,噪声降低,对缺陷的分割更加准确,验证了增强方法的有效性,为更加准确地进行缺陷识别和尺寸定量计算打下了基础。
The contrast is low, and the noise is strong in infrared thermal images. In order to resolve these problems, image enhancement step was designed. At first, frequency bands distribution and their source were analyzed in infrared thermal image, and effective frequency bands was identified as high frequency bands, brought by the infrared characteristics of the tested object. The target of image enhancement step was to enhance energy of the high frequency bands. According to frequency distribution characteristics of infrared figure, based on treatment goals, gray value adjustments and Butterworth pass filtering were used to process heat image comprehensively. The effect of image enhancement was tested by means of displaying image and segmenting image by Otsu method. The result shows that image contrast is improved, noise is reduced, and segmenting defects is more accurate after image enhancing processes. These verify the effectiveness of the image enhancement step, which lays a foundation for defect recognition and quantitative calculation of defects sizes.
出处
《红外与激光工程》
EI
CSCD
北大核心
2012年第7期1922-1927,共6页
Infrared and Laser Engineering
关键词
图像增强
红外热波检测
灰度值调整
巴特沃斯滤波
有效频段
image enhancement
infrared waves NDT
gray adjustment
Butterworth pass filter
effective frequency bands
作者简介
刘涛（1984-），男，助教，硕士，主要从事红外热波检测中图像处理方面的研究。Email：liuta00418@126．com
