Research on graph reconstruction method based on neural network and genetic algorithm

Aiming at the universality and difference of neural network and genetic algorithm in graphic reconstruction design, this study takes Miao nationality batik graphic design as an example, and makes a comparative analysis of the application of the two. Firstly, neural network and genetic algorithm are used to reconstruct the frame of batik graphics. Combined with the topological configuration, the graphics elements are transformed and filled, and the graphics groups of different elements, the same structure and the same elements are generated, and the comprehensive features of the two algorithms for the example are compared and analyzed. The results show that these two algorithms have their own advantages and disadvantages among the four design elements:area ratio, pattern element, frame configuration and application scene diversity. And in this case, the total weight of BP neural network generation scheme is greater than the total weight of GA generation scheme, which is 0.5625.