基于独立分量分析的红外热波无损检测图像增强

基金项目:

国家自然科学基金资助项目(51075390)

详细信息

Image Enhancing for Infrared Thermal Wave Nondestructive Testing Based on Independent Component Analysis

摘要: 在红外热波无损检测过程中，热像图存在低对比度、高噪声的问题。提出一种基于独立分量分析的图像增强方法，该方法能够有效地去除热像图中噪声的干扰，提高缺陷的对比度。阐述了ICA的基本原理，介绍了基于负熵判据的Fast ICA算法，给出了该方法在热像图增强处理中的具体实现步骤及相应的试验处理结果。结果表明，该方法具有较大的峰值信噪比，可提高图像质量，改善图像的可视性，为后续的图像分割等研究奠定基础。
- 红外热波无损检测 /
- 独立分量分析 /
- 图像增强
Abstract: A method of image enhancing based on Independent Component Analysis(ICA) is presented for infrared(IR) thermal images that acquired by an IR camera in the test system and always have the problem of lowcontrast and highnoise, which can effectively eliminate the noise interference of the images and enhance defect contrast. The fundamental theory of ICA and Fast ICA algorithm based on negative entropy criterion is introduced. The detailed processes of the method and results are given. The results of the experiments show that the method has higher peak signal to noise ratio(PSNR) and can improve image quality, which establishes basis for future research of image segmentation.
- Infrared thermal wave nondestructive testing /
- Independent component analysis /
- Image enhancing

[1]	田裕鹏.红外检测与诊断技术［M］.北京：化学工业出版社,2006.
[2]	王迅，金万平，张存林，等.红外热波无损检测技术及其进展［J］.无损检测,2004,26(10):497-501.
[3]	戴景民，汪子君.红外热成像无损检测技术及其应用现状［J］.自动化技术与应用,2007,26(1):1-7.
[4]	杨正伟，张炜，田干，等.导弹发动机的热波无损检测［J］.无损检测,2009,31(1):7-9.
[5]	张布悦，王桂增，刘吉东，等.输油管线泄漏检测和定位技术综述［J］.上海海运学院学报,2001,22(3):13-16.
[6]	陆国良. NIM3长输管道泄漏实时监测系统［J］.石油规划设计,2004,15(2):1-3.
[7]	李鸿燕，郝润芳，马建芬，等.基于独立分量分析的图像增强［J］.弹箭与制导学报,2007,27(5):299-302.
[8]	郭武，张鹏，王润生.独立分量分析及其在图像处理中的应用现状［J］.计算机工程与应用,2008,44(23):172-177.
[9]	Hyvarinen A, Oja E. A fast fixed-point algorithm for independent component analysis［J］. Neural Computation,1997,9(7):1483-1492.
[10]	Hyvarinen A. Fast and robust fixed-point algorithms for independent component analysis［J］. IEEE Trans Neural Networks,1999,10(3):626-634.
[11]	DameraVenkata N, Thomas D Kite, Wilson S Geisler, et al. Image quality assessment based on a degradation model［J］. IEEE Trans on Image Process,2000,9(4):636-650.