基于小波变换条纹修补和条纹自相似性的干涉条纹极值提取

Fringe Mending Based on Wavelet Transformation and Maximum Extraction on Self-resemble of Interference Fringes

提出了基于小波变换的条纹修补方法和利用干涉条纹自相似性的条纹灰度极值自动判读方法.基于小波变换的条纹修补方法包括滤除噪音、多尺度小波变换、模极大值检测滤除奇异区域、近似信号自动修补,在此基础上进行多项式拟合,首先实现最外层条纹灰度极值的自动提取.基于干涉条纹自相似性的条纹灰度极值自动判读方法是通过逐渐平移拟合区域,准确地提取了内层条纹的灰度极值位置,从而可以处理整幅图片的干涉条纹.此方法的特点是自适应、无需人为设定参量、处理速度快.实验结果表明,该方法有很好的可靠性和准确性,并且处理区域大、用时少,仅需对图片扫描一次即可提取全部条纹的灰度极值.

The methods to mend fringes based on wavelet transformation and to extract grey maximum on fringe self-resemble were presented. The mending method using wavelet transformation includes in filtering noise,multiscale wavelet transformation, filtering singularity areas and auto mending fringes. The grey maximum of the external fringe was first auto extracted and polynomial fitting was given after the fringe mending. By gradual translating the fitting area,the grey maximum extraction of the internal fringes based on the fringe self-resemble was realized. The whole interference fringe pattern can be processed by the present methods. These methods are characteristic for self-adaptation, no personal parameters needing design and quick processing speed. The experimental results show that the methods have good reliability and precision. What＇s more,it can process large area and save time. It only needs to scan the fringe pattern one time then can extract the whole grey maximum of the fringes.