The Research For Image Super-Resolution Reconstruction Based On Wavelet Transformation

The super-resolution reconstruction based on image sequence is a technique for estimating a high-resolution image from a low-resolution image by a way, which without changing the observing systems of image. The low-resolution image use CCD technique to take photographs and can achieve high-resolution image reconstruction on subpixel level. Wavelet method has become the important embranchment of super-resolution restoration. Image super-resolution processing has proved to be useful in many practical applications, such as medical imaging, public security, remote sensing and military detection, etc.Firstly, many super-resolution reconstruction algorithms are classified and discussed, the space field method of multiple images superresolution reconstruction is the mainstream of the current study, and have analyzed its each wavelet transform in this foundation. It not only make up the Fourier analysis , but also retains the advantages of Fourier analysis, and widely used in the field of image analysis.Secondly, the average modulation transfer function of general imaging system is analyzed, and the modulation transfer function of discrete sampling process of CCD sensor is analyzed and the calculating formula of average MTF of CCD sensor in subpixel imaging system is acquired. Compared the average MTF of general imaging and subpixel imaging system, the special resolution of the image acquired by subpixel imaging system is higher than the general one.Finally, bounded domains and irregular samples of first generation wavelet transform are analyzed, and a new kind of algorism which basied on the second wavelet transform is presented. By using hard /soft thresholding, the noise of wavelet coefficients is reduced. The image sequence with displacement is reconstructed by the different wavelets, The results of experiment and computer simulations are analyzed and compared.The paper implemented show that super-resolution reconstruction algorithms of the second wavelet transform remained a wealth of detailed information and improved the resolution of images. Finally, this topic work confirmed the above algorithm feasibility.