The conclusion of the paper discusses a simple algorithm for matching three-dimensional objects using observed silhouettes. Experiments show that as more silhouettes are added to modify the three-dimensional structure, consistent matching results from moments and Fourier transform coefficients. The speed of convergence depends on the amount of information provided by the sequence of silhouettes, with small modifications requiring fewer new information. For man-made objects with clearly defined faces and edges, fast convergence is achieved when principal or orthogonal silhouettes are observed. However, the technique fails when parts of the object are missing or the object is not isolated, as the principal direction calculation becomes inaccurate. The paper also introduces a method for multiple resolution texture analysis and classification based on changes in texture properties with varying resolution. The area of the gray level surface is measured at different scales, and the change in this area is used as the "fractal signature" for texture classification. The method is compared with earlier work using fractal techniques and is shown to provide effective classification results. The paper discusses the measurement of curve length and surface area, and how these measurements can be used to compute the fractal signature. The fractal signature is used to compare textures, and the method is extended to consider the directional properties of textures. The paper concludes by highlighting the potential applications of this technique in industrial materials and surface-texture probing.The conclusion of the paper discusses a simple algorithm for matching three-dimensional objects using observed silhouettes. Experiments show that as more silhouettes are added to modify the three-dimensional structure, consistent matching results from moments and Fourier transform coefficients. The speed of convergence depends on the amount of information provided by the sequence of silhouettes, with small modifications requiring fewer new information. For man-made objects with clearly defined faces and edges, fast convergence is achieved when principal or orthogonal silhouettes are observed. However, the technique fails when parts of the object are missing or the object is not isolated, as the principal direction calculation becomes inaccurate. The paper also introduces a method for multiple resolution texture analysis and classification based on changes in texture properties with varying resolution. The area of the gray level surface is measured at different scales, and the change in this area is used as the "fractal signature" for texture classification. The method is compared with earlier work using fractal techniques and is shown to provide effective classification results. The paper discusses the measurement of curve length and surface area, and how these measurements can be used to compute the fractal signature. The fractal signature is used to compare textures, and the method is extended to consider the directional properties of textures. The paper concludes by highlighting the potential applications of this technique in industrial materials and surface-texture probing.