基于Depth from Focus的图像三维重建

3D Recontructing of Image Based on Depth from Focus

应用改进Laplacian聚焦算子实现序列图像的融合显示,用高斯插值算法得到图像三维重建中的高度图,并设计一种轮廓线算法对高度索引图进行了三维重建;成功地应用于摄像机拍摄序列图像的三维重建,弥补了摄像机物镜焦深范围有限不能采样清晰大幅图像的不足。理论和实验结果证明,用该方法生成的二维融合图像和三维显示图像,恢复出比较精确的物体表面深度信息,提高了图像的清晰度。

To rebuild the image accurately, the microscope samples a series of images on different depths of focus. But the limitation of the object lens makes it unfit for sampling the large image clearly. The image fusion algorithm for a serial microscopy images is realized by improving the Laplacian algorithm operator. In the algorithm, it uses the Gaussian interpolation algorithm to acquire the depth of 3D recontructing of the image. A special contour line algorithm is designed to rebuild the 3D image accordinig to the depth index. The algorithm is successfully used in 2D image fusion and 3D recontructing of the image of the microscope. It recoveries the information of object surface depth, and making recontructing image more clear and accurate.