The aim of this research work is the reconstruction of 3-D model of an object from the range images obtained by scanning the object with a laser beam. Subsequently, it focuses on the feature level fusion of range and intensity images of the object. To achieve the objective, a data acquisition system has been developed with a laser source and a camera interfaced with Silicon Graphics machine. The method of optical triangulation is employed for the calculation of depth information of laser illuminated points. The acquired images are subjected to preprocessing and thinning algorithm. The registration of range images from different views is performed by feature matching technique followed by iterative closest point (ICP) algorithm using unit quaternion. The integration algorithm includes detection of overlapping points, merger of corresponding points and generation of a single connected surface model using Delaunay triangulation. This stage leads to reduction of redundant data points considerably and thus minimizes the computational effort. The next part of the work focuses on the extraction of structural features like edges and corner points from the draft 3-D model as well as intensity image of the object. Extraction of structural features from the 3-D model has been performed by two methods. The first method utilises coordinate thresholding technique to detect edge points in the range images with human intervention to fix the threshold values where as the second method detects the edge points in the interpolated mesh surface using Laplacian of a Gaussian (LoG) zero-crossing edge detector. An approach using shape signatures has been proposed to detect corner points in the edge map obtained using LoG detector. Corner points are detected out of the prominent peak values in the shape signature. Extraction of structural features from the intensity image of the object obtained under lighting conditions has also been performed by two methods. In the first method, gradient based edge detection using Sobel operator is followed by edge linking procedure using the Hough transform technique to obtain the edge map of the object. In the second method, edge map is obtained using Canny edge detector. The corner points in the Canny detected edge map are obtained by implementing the technique using shape signatures. The extracted edge maps from both the modalities should be in same plane for fusion. For