An oil painting image retrieval strategy based on feedback mechanism

Abstract

Image retrieval is mainly a text-based image retrieval technology, which uses text description to describe the characteristics of images, such as the author, age, genre and size of paintings. Later, image retrieval techniques that analyze and retrieve the content semantics of images, such as color, texture, layout, etc. Feedback technology has become an important technology to improve retrieval efficiency in content-based image retrieval. However, in oil painting image retrieval, there is a huge semantic difference between high-level semantics and low-level features. The traditional relevance feedback technology requires multiple feedbacks to obtain satisfactory results, which makes the retrieval task of users time-consuming and cumbersome. In order to further improve the accuracy of retrieval, many systems combine relevant feedback technology to collect the feedback information of users on retrieval results. What they achieve is a gradual refinement of image retrieval process. In the same retrieval process, they need to constantly interact with users. Therefore, this paper proposes a retrieval technology of oil painting image through the feedback mechanism of oil painting image retrieval. On the one hand, the analysis and transformation of user needs form questions that can be retrieved from the index database. On the other hand, collect and process oil painting image resources, extract features, analyze and index, and establish image index database. The last aspect is to calculate the similarity between user questions and records in the index database according to the similarity algorithm, extract records that meet the threshold as the result, and output them in descending order of similarity.