The Super-Resolution Algorithm Of MMW Image And The Study Of Calibration Method

With the development of science progress, applications of the infrared and microwave technology are quite widely in civil life and military technology. But the millimeter wave technology development lagging far behind them. Millimeter wave frequency spectrum between infrared and microwave between, the characteristics of the millimeter wave is different from the characteristics of the infrared and microwave. Therefore, millimeter wave technology and passive millimeter wave imaging technology have become a new hot point of the research in recent years.At present, passive millimeter wave imaging technology is still in the development stage. Due to the parts of an apparatus or appliance and related technology factors, practical obtained millimeter wave imaging its resolution is lower. In order to improve the imaging quality, in addition to the restricted to the current technology and processing technology level conditions outside, the other way is through the research all kinds of algorithm, from the low resolution images to improve the resolution of the images, called the super-resolution image restoration and reconstruction. So the all kinds of algorithm is proposed to realize from the lower resolution image for high resolution image, has become a hot spot in the study of millimeter wave (MMW) passive imaging one.This paper is based on the subject in the algorithm, puts forward a new millimeter wave imaging processing algorithm, and making the processed images the resolution was improved. This article main research results and the innovation place has the following several aspects:1. In order to improve the passive millimeter wave imaging resolution, a new adaptive super-resolution algorithm was proposed. This article through the improvement of wiener filtering algorithm, restore image low frequency component, and make full use of millimeter wave image information to get the low frequency component instead of the pass-band with frequency spectrum component that will be able to adaptive. Then combined with regular MAP algorithm and improved wiener filtering algorithm, using characteristics of the regular MAP algorithm that lap extrapolation high frequency component, the iterative algorithm and the images were Fourier transform for correction of frequency domain, and then inverse transformation. Finally, make further correction. The effectiveness of the new algorithm was verified through many experiments.2. Richardson-Lucy (R-L) algorithm is a kind of effective recovery method. But R-L algorithm easy to lead. The disadvantages such as noise amplification, when the noise is very large, restoring is often not ideal. Anisotropic diffusion process (P-M process) has strong denoising, edge protection ability. This paper puts forward a kind of based on anisotropic expansion process of passive millimeter wave R-L algorithm. First, image denoising to be done by using P-M process, and then restore image using R-L algorithm. The new algorithm is effective to reduce the noise on the influence of R-L algorithm.3. Passive millimeter wave detection and imaging technology is used through the radiometer receive from the outside world to the object of millimeter wave radiation energy. In the millimeter wave radiometer detection, scaling is one of indispensable a step. And calibration precision can directly affect the accuracy of the measurement. So study the scaling method and improve the precision of calibration become one of the important topics in the measurement of the millimeter wave. This paper gives for the nonlinear scaling function and improves the accuracy of the calibration using Newton interpolation algorithm which is relatively simple and practical advantages.