Construction of image acquisition system based on shear wave transform and transfer learning

Abstract

With the development of the internet and popularization of intelligent devices,the multimedia data type on the internet has also experienced changes accordingly,whichalso contains the label and text information set by users while containing its self-carrying visual information. Under the mixed background of multi-form information, how to provide users with fast and accurate image acquisition results becomes a new challenge of multimedia retrieval. In this paper, an image acquisition method based on the shear wave and transfer learning is put forward. Itstudies the text information of images while studying its visual information. Moreover, it transfers the learned results to the visual information field forcross-form information fusion, so as to produce image features containing cross-form information. And the experiments show that the proposed method can realize better image acquisition results.