Research Of Near-field Optical Image Processing

Optical imaging is always a main and important method in the research fields of physics, chemistry, biology, medicine and so on for studying the characteristics and structures of the sample. Nanometer technique is coming into more and more fields as the developing of the technique, while the traditional microscope can not come with it for the limit of diffraction in detecting the nanometer characteristics. Near-field scanning optical microscope satisfies the requirement for nanometer observation, which can break through the limit of the diffraction and get the nanometer characteristic and structure of an object by scanning the surface of it using a nanometer probe, and so NOSM imaging techniques become a hot research point all over the world.After lots of researches about the characteristics of NOSM imaging, several image processing methods of NOSM imaging are presented in this paper. There is a special kind of noise existing in NOSM image as the result of scanning imaging method, which named line noise. In this paper new arithmetic is brought out to denoise the line noises by using the noises compensation factor. Then random noises are denoised by wavelet transform filter further. It is trouble to detect the edges of the NOSM image for its weak contrast, which is due to weak imaging signal. So an edge detection method based on phase information is presented, which is more efficiency in detecting the NOSM image edges than the method based on intensity. Considering the models of NOSM imaging, image fusion processing is used in transmission and reflection model images of NOSM, which makes useful information from different images into one and enhance the image quality.