Scanning probe microscopy (SPM) is used widely for the characterization of topographic structure in both material and bio-sciences. However, due to the finite size of the imaging tip, the SPM may not record all of the surface. Dimensional measurements made on such surfaces will not reflect the true nature of the surface. Simulated images have been used to assess the effects of imaging and image reconstruction methods on surface roughness and dimensional measurements. A novel combination of image reconstruction techniques and connectivity criteria is used to set bounds on the certainty of dimensional and topographic measures and to obtain more information about the spatial complexity of the sample surface. A more complete description of the sample is obtained without prior knowledge of the sample or tip geometry.