首页 | 杂志简介 | 征稿简则 | 投稿指南 | 常见问题 | 名词解释 | 刊物订阅 | 联系我们 | English
红外无损检测图像处理算法的研究
投稿时间:2009-01-21  最后修改时间:2009-02-15  点此下载全文
引用本文:
摘要点击次数: 3011
全文下载次数: 143
作者单位E-mail
冯贞 北京航空航天大学自动化科学与电气工程学院 sy0603414@yahoo.cn 
基金项目:国家自然科学基金项目(面上项目,重点项目,重大项目)
中文摘要:红外无损检测的原始热像通常存在信噪比低、温度对比度差的问题。为了滤除噪声的不良影响,增强缺陷的对比度,需要对原始热图像进行处理和重构。而算法的选择决定了图像处理的效果。本文通过机械加工模拟腐蚀件实验,对红外热图像处理中的4种典型算法进行了研究。对处理结果做出了比较,得出了各种算法的适用性和性能对比结论。该结论可作为实际应用中算法选择的依据和参考。
中文关键词:红外无损检测  图像处理  算法  模拟实验
 
Research on Image Processing Algorithms Applied in Infrared Non-destructive Testing
Abstract:The original images in Infrared Non-destructive Testing (NDT) generally have the problems of low Signal Noise Ratio (SNR) and poor temperature contrast. For filtering the harmful effects of noise and increasing the defect contrast, the processing and reconstruction of original thermal image is necessary in infrared NDT. The selection of algorithms determines the effectiveness of image processing. This paper works on researches of 4 typical algorithms in infrared thermal image processing on the basis of experiments on machining corrosion pieces. The applicability and performance conclusion of different algorithms is obtained by comparing the processing results. The conclusion can be referred as algorithm selection guidance in infrared image processing in practice.
keywords:infrared non-destructive testing  image processing  algorithms  experiments
查看全文  查看/发表评论  下载PDF阅读器

版权所有:《红外》编辑部

北京勤云科技发展有限公司