We present a novel approach to extracting artistic style features from oil paintings, addressing the limitations of existing one-way matrices. Our proposed algorithm focuses on capturing emotional expression in oil painting styles. To ensure relevance to real-world needs and standards, we establish a multi-objective framework with a set of indicators and parameters for fundamental oil painting artistic style extraction. This expansion of the feature extraction calculation range involves the creation of a multi-objective calculation matrix and the development of a pyramid sequence and extraction algorithm for conveying emotional expression. By leveraging local constraints specific to oil paintings, our method effectively extracts artistic style features. The results of our tests, which involved five oil paintings, demonstrate the effectiveness of our approach. By applying the Artistic Style Feature Extraction (ASFE) algorithm based on emotional expression, we successfully control the differences in oil painting texture depth within a range of 0.65. This indicates a high level of authenticity and accuracy in our algorithm, superior calculation performance, and practical utility for various applications.