Heterogeneous Graph Structure Learning (HGSL) aims at jointly learning optimized graph structure and representation. Its purpose is to enhance the performance and robustness of Heterogeneous Graph Neural Networks (HGNNs) by solving the problems of redundancy, bias, noise, incompleteness, and unreliability in graph structures. However, the existing research is still insufficient in selecting the multi-order neighborhood information aggregation method, which leads to increased computational complexity and over-convergence of multi-layer semantics. Meanwhile, when dealing with rich node features and complex topology, previous research exhibits the problem of insufficient information utilization. To solve these problems, we propose a novel model named Heterogeneous Graph Structure Learning Based on Feature and Topology Information Extraction (HGSL-FTIE). First, we propose an n-hop meta-path extraction strategy. Based on this strategy, we fuse the node features to construct the feature structural subgraph and fuse the original graph structure to construct the topology structural subgraph. Then, we creatively adopt a structural subgraph fusion method based on the More Confident Fusion (MCF) strategy to guide the fusion of the feature structural subgraph and the topology structural subgraph to obtain the final graph structure and the corresponding node representation. We have conducted comprehensive experiments on three real datasets, and the results demonstrate that HGSL-FTIE outperforms the existing state-of-the-art models.