Bronze-mirror cultural relic identification system and method

Abstract

The invention relates to a bronze-mirror cultural relic identification system which is constructed based on a bronze-mirror cultural relic characteristic matrix and a neural network model. The systemis characterized by acquiring image data of a bronze mirror, and according to a characteristic set, constructing a hierarchical grayscale co-occurrence matrix; and based on a classical convolutional neural network CNN model of a deep learning field, constructing a self-feedback parameter-adjusting neural network model applied to bronze mirror identification. Different hierarchies of effective identification can be performed on the bronze-mirror cultural relic, and through collecting feedback information of a user to an identification result, iteration optimization to the bronze-mirror culturalrelic identification system is reinforced.

Description

Technical field: [0001] The invention relates to the technical field of cultural relics identification, specifically, microscopic comparison and deep learning model identification technology of bronze mirror cultural relics, so as to ensure the accuracy and real-time performance of bronze mirror identification. Background technique [0002] In recent years, with the rapid development of my country's economy, culture and other fields, the collection and investment of cultural relics and artworks have become more active. The vast number of bronze mirror collectors and investors often suffer huge economic losses due to the existence of imitations or fakes of bronze mirrors, and at the same time seriously distort the cultural relics and artistic value of bronze mirrors. Therefore, scientific identification of bronze mirrors has become more and more practical. [0003] However, due to the lack of scientific and effective identification systems and methods, the identification con...

Images