Research on the analysis of image characteristics in fine art painting under the application of machine learning technology

Fine art painting has shaped civilization throughout history. Fine art painting best expresses the human experience. Smartphones have helped fine art internet galleries grow rapidly. Thus, many internet services and institutions offer substantial digital collections. With so many digital artworks to study, automated processing and analysis of fine arts is essential. Classifying fine art paintings automatically is a crucial problem for facilitating the analysis of such works. This article suggests a new machine learning (ML) classifier for fine-art painting images called stochastic swarm intelligent-based logistic regression (SSI-LR). In this instance, through generating optimal image features, the classification performance of the LR approach is optimized using the stochastic particle swarm optimization (SPSO) algorithm. The fine art painting source data samples are taken for this examination, and they can be de-noised using a median filter (MF) to remove imperfections. The contrast aspect of the painting images is then improved using the contrast improvement (CI) method. The essential characteristics of the painting images are extracted from the augmented images using linear discriminant analysis (LDA). Additionally, the retrieved data is employed to effectively characterize the images of fine art paintings using the augmented feature representations from the suggested SSI-LR approach. The experimental findings demonstrate that, compared to other approaches already in use, the proposed method achieves excellent categorization of images of fine art paintings.