Whale optimized damaged terracotta figure fragment registration method based on chaos reverse learning

Abstract

The invention belongs to the technical field of image processing, and discloses a whale optimized damaged terracotta figure fragment registration method based on chaos reverse learning. The method comprises the steps: carrying out the data preprocessing, and generating source point cloud data P and to-be-registered data Q; establishing a mapping relation between population individuals and a model according to a target function, and obtaining an optimal transformation matrix by using a rotation R parameter and a translation T parameter; initializing parameters; saving the elite data in an elite database by using an elite retention mechanism; designing a weight factor and a convergence factor, updating the position of the whale individual, and calculating the fitness of the whale individual; updating the partship degree, the non-membership degree and the hesitation degree; updating the distance between the whale individuals; calculating a shared function value of the niche technology, and updating a fitness value; judging whether an iteration condition is met or not; and applying the searched optimal solution to the R and T parameters to obtain a final registration model. The whale optimization algorithm is optimized, and the improved whale optimization algorithm is more suitable for fragment registration.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a method for registering fragments of damaged figurines optimized by whales based on chaotic reverse learning. Background technique [0002] At present: With the rapid improvement of computer performance, computer-based image processing technology develops very rapidly. Since mobile devices can collect two-dimensional images, a large amount of available data is provided for two-dimensional images. Therefore, the two-dimensional image Visual processing technology is developing rapidly. With the continuous development of image processing technology, two-dimensional images can no longer meet people's requirements for three-dimensional description of the real world, and researchers have gradually shifted their research focus to three-dimensional images. [0003] At present, 3D point cloud image processing technology plays an important role in the fields of 3D re...

AI Technical Summary

Problems solved by technology

Point cloud registration generally has the following problems: (1) the data itself has noise, which affects the accuracy of registration; (2) in the process of data acquisition, due to the problems of self-occlusion light and viewing angle of the 3D scanner, there are data missing or Partial overlap and other problems make it difficult to find the corresponding relationship of registration; (3) The initial position of point cloud data has a great influence on the performance of registration

[0005] During the registration process of damaged figurine fragments, there may be the following problems: (1) The environment (long-term age, fading of fragments) problems in the data collection process, that is, there is noise in itself, which affects the accuracy of registration; (2) During the data acquisition process, due to the problems of self-shielding light and viewing angle of the 3D scanner, there are problems such as missing data or partial overlap, which makes it difficult to find the corresponding relationship of registration; Great performance impact

These problems may lead to problems such as large-scale sample data, redundant or missing information, high noise, fuzzy uncertainty, etc.

[0006] Through the above analysis, the existing problems and defects of the existing technology are: the registration efficiency of the existing point cloud data registration technology is low due to factors such as high-dimensional, fuzzy, abstract, redundant, and uncertain data characteristics, and the simple ICP algorithm The registration speed is slow, there are high requirements for the initial registration data, and the accuracy is difficult to improve

[0007] The difficulties in solving the above problems and defects are: slow registration speed and low registration accuracy

Images