The use of computer technology for three-dimensional (3 D) reconstruction is one of the important development directions of social production. The purpose is to find a new method that can be used in traditional handicraft design, and to explore the application of 3 D reconstruction technology in it. Based on the description and analysis of 3 D reconstruction technology, the 3 D reconstruction algorithm based on Poisson equation is analyzed, and the key steps and problems of the method are clarified. Then, by introducing the shielding design constraint, a 3 D reconstruction algorithm based on shielded Poisson equation is proposed. Finally, the performance of two algorithms is compared by reconstructing the 3 D image of rabbit. The results show that: when the depth value of the algorithm is 11, the surface of the rabbit image obtained by the proposed optimization algorithm is smoother, and the details are more delicate and fluent; under different depth values, with the increase of the depth value, the number of vertices and faces of the two algorithms increase, and the optimal depth values of 3 D reconstruction are more than 8. However, the proposed optimization algorithm has more vertices, and performs better in the reconstruction process; the larger the depth value is, the more time and memory are consumed in 3 D reconstruction, so it is necessary to select the appropriate depth value; the shielding parameters of the algorithm have a great impact on the fineness of the reconstruction model. The larger the parameter is, the higher the fineness is. In a word, the proposed 3 D reconstruction algorithm based on shielded Poisson equation has better practicability and superiority.

使用计算机技术进行三维（3D）重建是社会生产的重要发展方向之一。目的是找到一种可用于传统手工艺品设计的新方法，并探索3D重建技术在其中的应用。在对3D重建技术进行描述和分析的基础上，分析了基于泊松方程的3D重建算法，阐明了该方法的关键步骤和存在的问题。然后，通过引入屏蔽设计约束条件，提出了一种基于屏蔽泊松方程的3D重构算法。最后，通过重建兔子的3D图像比较两种算法的性能。结果表明：当算法的深度值为11时，提出的优化算法获得的兔子图像表面更光滑，细节更细腻流畅。在不同的深度值下，随着深度值的增加，两种算法的顶点数和面数均增加，并且3D重构的最佳深度值大于8。但是，所提出的优化算法具有更多的顶点，并且在重建过程中表现更好；深度值越大，在3D重建中消耗的时间和内存就越多，因此有必要选择合适的深度值；该算法的屏蔽参数对重建模型的精细度有很大影响。参数越大，精细度越高。一言以蔽之