Fusion Technology Research Of Visible And Millimeter-wave Imaging Based On Wavelet Theory

Passive millimeter-wave (PMMW) imaging technology is a method of formingimages through the differences of radiant energy in millimeter-wave band between ascene and an object, which has the ability to work during all-weather condition, throughclothing and military disguise. In accordance with visual observing habits aboutconventional visible light, it’s not appropriate to observe millimeter-wave imagescorresponding to radiometric temperature, with low resolution and noise caused by thelimitations of hardware implementation in normal conditions. Image fusion technologycan synthesize contour sharpness of optical sensor and detectivity about concealedweapons of millimeter-wave image, thus achieve mutual benefits of multi-sensors andimprove image quality effectively. It has become a frontier and hot issue researchtechnology.An important aspect of image fusion technology is image preprocessing andselection of fusion criterions. The thesis studied misalignment correction algorithm ofmillimeter-wave image based on frequency-domain phase, and fusion method based onwavelet theory about visible and millimeter-wave images. The main contents andconclusions are as follows:1. Phenomenon of misalignment caused by mechanical error of scanning servosystem is corrected by frequency-domain phase correction algorithm, and this thesispresents a solution of smooth transition to widen the fitting interval.2. Inaccurate estimation of individual dislocation leads to the incoherentphenomenon of dislocation overall. Kalman filter theory and LS estimation theory hadbeen used to smooth dislocation and eliminate bad value effectively.3. Improve the regional energy matching algorithm for low-frequency informationfusion of visible and millimeter-wave images, and it proves to be more suitable forobservation and the image texture more clearly.4. Extract edge using third-order spline wavelet, and utilize it in the selectionfusion of high-frequency information to overcome loud noise and invalid information inthe actual applications. Theoretical research above has passed the simulation experiments of visible andmillimeter-wave images fusion, and the results show that: estimation and optimizationalgorithm based on frequency-domain phase corrects the dislocation between lines inmillimeter-wave image; the improved regional energy matching method is more suitablefor the low-frequency information; the high-frequency information fusion algorithm ofedge localization can effectively solve blur edge in fusion image.