Hand segmentation plays a vital role in various computer vision tasks such as gesture recognition and human-computer interaction. However, achieving accurate segmentation remains challenging in the presence of complex backgrounds, fine-grained hand structures, and motion blur. To address these issues, we propose a Multi-Scale Feature Weighted Aggregation Network (MSFWAN) for hand segmentation. Specifically, a Hierarchical Feature Aggregation Module (HFAM) is designed to capture fine-grained hand details by aggregating multi-scale contextual information, enabling the model to capture hand boundaries and subtle structures. To enhance robustness under motion blur, we further propose a Multi-Scale Feature Weighted Aggregation Module (MSFWAM) that selectively emphasizes salient features across scales. Extensive experiments on multiple benchmark datasets demonstrate that our method outperforms existing methods.