Skip to Main Content
This paper describes a mechanical part similarity assessment algorithm based on shape distribution, which represents a 3D shape as a D2 shape histograms. First, generating a large number of random sample points from a model. Next, we adopt the D2 shape function to compute the distance of any two random points and construct a shape distribution curve by having a statistic of the distance values. Last, compare each pair of shape distribution histograms to implement the similarity assessment of the parts. Experimental results show that this method can efficiently achieve the similarity assessment of the mechanical parts and as well reflect human perceptual similarity.