Similarity Evaluation Method of Robotic Cow-Nosed Ray Based on Shape Context Algorithm

In order to solve the problem that it is difficult to accurately judge whether the Robotic Cow-Nosed Ray is similar to the real cow-nosed ray, the similarity evaluation method of the Robotic Cow-Nosed Ray is presented in this paper. First, we collect images from different perspectives of the Robotic Cow-Nosed Ray and the cow-nosed ray, extracted their profiles and used Jitendra’s sampling to ensure that the points were as uniform as possible to fully reflect the profile features of the cow-nosed ray. Second, we use the shape context operator to calculate the similarity evaluation matrix, and measure the matching degree of point pairs by normalized shortest distance. Last, calculate the similarity between each view through the shape context distance. The similarity of each view is weighted and summed to obtain the overall shape similarity of the Robotic Cow-Nosed Ray. The value of similarity is calculated based on the shape context algorithm, which points out the direction for the shape and structure optimization of the Robotic Cow-Nosed Ray.