Shape Knowledge Indexing Using Invariant Shape Code

Shape knowledge indexing is crucial in both design reuse and knowledge engineering, in which the pivot issue is to establish the unique representation of the invariant shape properties. Treating the shape of the region of interest as a surface signal, in this paper, a local shape-indexing scheme is developed by applying the affine invariant nature of the Fourier spectrum of the spatial shape distribution. The shape-coding scheme is theoretically proven being strictly invariant under affine transformations. A framework applying the invariant shape code in shape knowledge indexing is presented. Associated examples and the quantity analysis results are provided to justify the robustness, simplicity, and adaptability of the proposed shape knowledge-indexing scheme. Further, the proposed approach could be regarded as an alternative choice to represent local shape knowledge, especially for that of freeform features.