基于改进的无上下文网络解决拼图问题
Solving Jigsaw Puzzle Problem Based on Improved Context-Free Network

本文提出了利用深度学习技术解决拼图问题的新方法。拼图游戏作为一种跨学科的智力挑战，其复杂性随着碎片数量的增加而指数级增长。本文介绍了一种基于改进AlexNet卷积神经网络架构的无监督学习方法，用于解决自然图像的拼图问题。该方法通过构建一个拼图任务预测网络，提高了模型的性能和泛化能力。实验部分在自制的果蔬数据集上进行，验证了模型在不同复杂度拼图还原任务中的有效性。最后，文章总结了研究成果，并对未来的研究方向提出了展望。
In this paper, we propose new methods for solving jigsaw puzzles using deep learning techniques. As an interdisciplinary intellectual challenge, the complexity of jigsaw puzzles grows exponentially with the number of pieces. This paper introduces an unsupervised learning method based on a modified AlexNet convolutional neural network architecture for solving the jigsaw puzzle problem of natural images. This method improves the performance and generalization ability of the model by constructing a puzzle task prediction network. The experimental part is carried out on the self-made fruit and vegetable dataset, and the effectiveness of the model in puzzle restoration tasks of different complexity is verified. Finally, the paper summarizes the research results and puts forward the future research directions.