Fundamental Study of Very High Resolution Image Acquisition Method using Multiple Image Processing for Very High Resolution Image Communication

针对VHD图像采集系统的开发，我们提出了一种新的基于信号处理的方法来提高现有图像采集设备的空间分辨率。基于信号处理的图像采集方法通过对多个不同相机同时拍摄的多幅图像进行处理和整合，生成具有足够高信噪比的改进的分辨率图像。经过理论分析和实验模拟，得出以下结论：1.在两个摄像机的光轴相交的情况下，或者在对三维物体成像的情况下，一个摄像机拍摄的图像的一个像素几乎与另一个摄像机拍摄的另一个图像的某个像素重合。另一方面，利用像素光圈不同的多个相机可以消除这种不均匀改善的问题，因为在不同光圈的情况下，不同于相同光圈的情况，一个相机拍摄的一个相机的像素与另一个相机拍摄的另一个相机的任何像素永远不会重合，无论多个相机的排列如何。在本研究中，我们展示了利用多个像素光圈不同的相机的优势，但需要研究即使在严重的混叠效应下也适用于不同光圈的稳健配准算法。

Aiming at the development of an imaging system for acquiring VHD images, we presents a new signal-processing based approach to improve the spatial resolution of existing image acquisition devices. The signal-processing based image acquisition method produces an improved resolution image with sufficiently high SNR by processing and integrating multiple images taken simultaneously with multiple different cameras. After the theoretical analysis and experimental simulation, we have had the next results.1.In the case of the optical axes of the two cameras intersect or the case of imaging 3-dimensional object, a pixel of one image taken with one camera almost coincides with some pixel of the other image taken with the other camera. And hence in such portion spatial resolution is not improved satisfactorily.2.On the other hand, the utilization of multiple cameras various in their pixel apertures enables us to remove this non-uniform improvement problem, because in the different-aperture case, unlike the same-aperture case, a pixel of one camera taken with one camera never coincides with any pixel of another camera taken with another camera irrespective of the arrangement of multiple imagers.In this research, we show the advantage of utilization of multiple cameras various in their pixel apertures, but it is requires that investigation of robust registration algorithm even under severe aliasing effects and available in the different aperture case.