Three-Dimensional Microscopy Image System Based on Depth from Focus

Microscope suffers from the problem of limited depth of focus, which precludes the observation of complete image of a three-dimensional object in a single view. In this paper, extending depth of focus by multi-focus image fusion is investigated. The principle of image fusion based on depth of focus is analyzed, and one spatial domain multi-focus algorithm is implemented. Then a multi-focus micro-image fusion algorithm based on area wavelet transform is presented, which offer improved performance over existing algorithms when processing transparent images. In order to obtain 3D structures of the object, Gaussian interpolation algorithm has been selected to interpolate a small number of focus measure values to obtain accurate depth estimates. Based on the above DEM and Montage images, the 3D surface of micro-object is reconstructed by surface rendering algorithm. And the stereo pair images(left and right) and anaglyph image are also emulated.