In order to address the problem of decreased classification accuracy due to electrode displacement and variations in subject movements during gesture classification, a gesture classification model based on surface electromyography (sEMG) signals and transfer learning was proposed. Firstly, the four-channel sEMG signals were processed to extract and denoise the active segments. Then, four temporal features and two frequency domain features were extracted from the active segments. The manifold embedded distribution alignment (MEDA) method was utilized to embed the feature matrices of the source and target domains into the Grassmann manifold for manifold feature learning, aiming to reduce the data differences between the two domains and eliminate feature degradation. Additionally, adaptive factors were employed to perform dynamic distribution alignment, dynamically adjusting the relative importance of marginal and conditional distributions under different distribution disparities. Experimental validation was conducted on multiple subjects to demonstrate the rationality of the proposed method. The results indicate that compared to three traditional machine learning methods, namely decision tree (DT), support vector machine (SVM), and k-nearest neighbor (KNN), the proposed method achieves an improvement in recognition accuracy of 13%, 21%, and 9% respectively. Compared to the joint distribution adaptation (JDA) transfer learning method that did not perform manifold learning and dynamic distribution alignment, the recognition accuracy has improved by 52%, reaching 93%. This demonstrates that the proposed method exhibits excellent performance in gesture classification under complex scenarios.