An optimization of K-mean Algorithm for Feature Extraction based on Computer Programming

Feature extraction is a widely used image algorithm, whose biggest challenge is feature selection. The process of selecting the optimal feature is complex, and different feature selection techniques may produce different feature sets, which may affect the accuracy of classification or regression. However, the current accuracy of feature extraction is not high, and there will be a high error rate, and the labor cost is too high, causing pressure to the society. This paper intends to build a system through k-means algorithm to improve feature extraction, to achieve higher accuracy. In this study, k-means algorithm is used to select the best feature subset. It groups data points into different clusters and selects the cluster with the highest degree of differentiation as the optimal feature subset. In addition, this paper also solves the problem of error rate in feature extraction application, and improves the efficiency and accuracy of feature extraction, with the accuracy rate reaching 97.8%. In the future, this technology is expected to be widely used, which will not only improve work efficiency and reduce costs, but also improve accuracy, which is of great significance to society.