Analysis of the spinning behavior of silkworms by 3D computer graphics

点击翻译按钮获取中文摘要

We devised a simple method to determine the position of a silkworm by using two video cameras and a personal computer. The coordinates of several points marked on the body of a silkworm were read on the CRTbased on the two pictures taken by two video cameras from two different angles. When the mark on a silkworm was not observed on the CRT, because it was hidden by the body of the silkworm, the coordinates of two vertexes defining an appropriate rectangle, which includes the marked point, were read. The iterative reweighted least squares method was used to estimate the position in three dimensions. Then we considered the accuracy of the reconstructed position of a silkworm obtained by the method proposed in this research. The variances in the estimates of two transformation matrices were given. The variance in the reconstructed position was also obtained, and the relationship between the accuracy and the measurement error was considered. We presented a new method for estimating the spinning speed using the reconstructed three-dimensional positions of the spinneret of a silkworm. The approximate variance was given to evaluate the accuracy of the estimated speed using a stochastic model. Furthermore, generalized Procrustes analysis was applied to the shape data of the silkworm so as to investigate the change of shape in spinning behavior. We determined that it was very efficient to use 3D computer graphics in the research of silkworm behavior. In this research we developed 3DASBS and applied it in all aspects of the analysis.