Font Size: a A A

Study Infrared Polarization Imaging Techniques And Methods

Posted on:2015-01-02Degree:MasterType:Thesis
Country:ChinaCandidate:W J YuFull Text:PDF
GTID:2268330425487719Subject:Optical Engineering
Abstract/Summary:
This article focuses on setting up image acquisition system, researching polarization image fusion algorithm,polarization image fusion algorithm, researching polarization image fusion algorithm researching polarization image simulation algorithm to improve the quality of the infrared image, improve the recognition rate and resolve infrared polarization image can not be good for target classification problem and getting polarization images before getting real polarization images.The paper uses visible light polarization theory to get the model of infrared polarization images. Since polarization degree images can outstand man-made objects very well, the polarization angle can describe the different surface orientations, intensity images can reflect intensity information of background.Using the infrared polarization image fusion algorithm based on wavelet transform to improve the image quality and identifiable rate of artifacts. In order to facilitate the observation of the human eye,using the pseudo-color algorithm to improve the image quality.But infrared polarization images can not be a good classification. By analyzing the infrared polarization image at each polarization angle of redundancy and complementarity, Fuzzy C-means clustering algorithm based on edge detection algorithm and propose a suitable under infrared polarization.image pattern recognition method for classification of objects in the image. The result is that it can detect man-made object under the background of mixed natural and obtain highly accurate results effectively and quickly. In order to obtain the true polarization image obtained before polarization image,finally using simulation methods to obtain polarization images. The resulting effect is obvious polarization image simulation can well distinguish between artifacts and natural objects in order to verify the feasibility of the simulation algorithm. Image simulation research for better and more convenient polarization properties of objects and the establishment of a database of polarization has a very important significance.
Keywords/Search Tags:Infrared polarization images, Wavelet transform, Fuzzy C-means clustering, Image simulation
Related items