Correlation \(\vartheta\) Measure Neighborhood Constraint Features Generation Small Data Targets Space Reconstruction

The process is detailed as follows. Firstly, the small data samples and their space points are obtained, and the quadratic elements are established to generate the projection transformation matrix with surface constraints. Secondly, a variety of construction factor transformations based on small data are obtained. Thirdly, the planar points are inversely projected into a sparse stereo point set, based on the F-norm distance domain, the correlation \(\vartheta\) measure factor is set and the surface is generated by infinite approximation. Finally, the seed points expend to generate dense sets and are iterated into exact models. The experimental results show that this method is effective, especially for small data samples, with low computation and high visibility of reconstructed objects.