Ji Qi, Yingzhong Tian, Wenjun Zhang, Albert Weckenmann, Minglun Fang
3D SURFACE MICRO TOPOGRAPHY RECOVERY FROM MULTI-FOCUS IMAGES USING MODIFIED LAPLACIAN OPERATOR
Optical microscopy enables the observation of highly magnified objects and material structures on micro surfaces, however with the weakness that it can only acquire 2D images. In order to observe the areal features more accurately and intuitively, 3D surface micro topography recovery is applied to form a 3D surface model of an object from its 2D image sequence. Optical microscope has a limited depth of focus in large magnification, which makes the area within the depth of focus cleared and other area blurred. So this paper firstly acquires image sequence which obtains all useful information in one view by vertical scanning of the microscope. Secondly, each image is calculated by an appropriate focus measure operator to find the maximum focus measure value and form a 2D fused image. Then the maximum value of each pixel is accurately transferred into a distance value, forming a discrete depth map. After conducting interpolation, fitting and color mapping, a smooth and authentic 3D color model of the measured surface is obtained. Various focus measure operators such as grey level variance are used to compare their performances in 3D model recovery. The superiority of the modified Laplacian operator proposed in this paper is proved by experimenting on measured objects with different micro topography features such as ditch shaped and slope shaped structure. In addition, surface roughness information of Ra and Rz is extracted from the formed 3D models.