Tree Structured Cooperative Coevolutionary Genetic Algorithm for Fragment Reconstruction

The fragment reconstruction problem aims to assemble the original object from a collection of fragmented pieces. Traditional manual reconstruction techniques heavily rely on expert knowledge and can potentially damage fragile fragments, necessitating the development of automated reconstruction methods. Current reconstruction algorithms often suffer from the curse of dimensionality, compromising both accuracy and efficiency as the number of fragments increases. These algorithms primarily rely on fragment content, limiting their adaptability and scalability. To address these challenges, this paper introduces a novel reconstruction method grounded in a cooperative coevolutionary (CC) optimization framework. This approach encompasses both the formalization of the fragment reconstruction problem and the development of a tailored algorithm to solve it. Notably, our modeling approach is content-independent, relying solely on the edge shapes of the fragments. With this modeling approach, the solution itself represents the reconstruction process of the fragments. To encode candidate solutions efficiently, we employ a tree structure. This encoding scheme renders traditional CC processes and genetic algorithm operators, such as crossover and mutation, inapplicable. Therefore, this paper proposes a tree-structured CC genetic algorithm (T-CCGA) specifically tailored to our reconstruction task. We aim to overcome the limitations of current reconstruction algorithms and pave the way for more accurate and efficient fragment reconstruction methods. To evaluate the effectiveness of the proposed method, we conducted a series of comprehensive experiments. The results demonstrate that T-CCGA achieves promising outcomes in terms of solution quality, convergence speed, and robustness.