Analysis of Color Language and Aesthetic Paradigm of Print Art Based on GB-BP Neural Network

Color is the basic element of printmaking art creation and also an important medium for artists to express their emotions. In order to improve the understanding of the print by tourists, the research first carries out image analysis of different colors, summarizes the common color image adjectives in the art of publishing and painting, then optimizes BP neural network algorithm by gradient promotion (GP) algorithm, and constructs a color language analysis technology based on GP-BP neural network review paradigm. Through this technology, the content of color expression of different types of print art works is analyzed. Then, the subjective evaluation of these print art works is carried out by questionnaire adjustment method. Finally, the subjective evaluation and network evaluation results are compared. On the three materials, the expression effect of color image adjectives such as “heavy light fast,” “modern classical,” “lively steady,” and “soft Yang Gang” is consistent (P value is greater than 0.05). The first set of print works mainly reflects the artist's “affinity cold” emotion. The MSE values of the evaluation results are less than 0.01. That is, the color language analysis technology of print art based on GP-BP neural network can reflect the artist's emotion to a certain extent when analyzing the color language of the print.