3-D reconstruction with cross-sectional images means to reconstruct the 3-D shape of the object with a series of cross-sectional images. The 3-D reconstruction method derived from industrial computed tomography images is focused in this paper. The original data comes from a large industrial computed tomography machine, which uses the 3rd generation fan-beam scan mode. Firstly, images are processed to remove air annulus. Then volume rendering is used to reconstruct the workpiece in Matlab environment. The cross section image resulted from the 2-D convolution projection reconstruction is the original image for 3-D reconstruction, which is image file. The noise caused by converting paper or film image into digital image is reduced in this way. Furthermore the additional step to match images is avoided by applying the fan-beam convolution projection algorithm that determines whether the images are matched.