The k-means is a classic clustering algorithm widely used in data mining. However, this algorithm is sensitive to outliers and requires the number of clusters to be determined in advance, along with the random selection of initial centers. This leads to lower accuracy in the clustering results. To address these issues, this paper proposes an enhanced k-means clustering algorithm integrating outlier detection and density peaks, named DPKM. This scheme aims to improve clustering accuracy. First, the algorithm preprocesses the data using outlier detection methods to identify and remove outliers that may affect subsequent analysis. Then, it employs a density peak clustering method to determine the optimal number of clusters and initial centers. Finally, the algorithm classifies and clusters the data based on these centers to improve the clustering results. The experimental results show that the proposed algorithm achieves significant improvements in Mean Squared Error (MSE) metrics and Adjusted Rand Index (ARI) compared to existing algorithms, verifying its effectiveness and superiority in clustering tasks.