| With the development of the information age,a large amount of image data has been produced all over the world.Nowadays,the amount of data information is increasing rapidly,and the existing mainstream storage devices are obviously insufficient in storage capacity and storage efficiency,which poses a great challenge to the traditional storage media.DNA data storage has the advantages of high storage density,strong stability,energy conservation and environmental protection.It can be used as a substitute for traditional digital storage media.At present,the price of DNA synthesis is high,so improving the storage density of DNA information has important research significance for reducing the synthesis cost and storing large-scale image data in DNA medium.At the same time,in the process of DNA data storage,it is easy to be affected by environment and biological processes,which makes DNA base data prone to errors.Therefore,it is also very important to add error correction codes to DNA coding information to reduce errors.Aiming at the above problems,this thesis studies the methods of DNA coding and error correction code.(1)This thesis proposes an end-to-end DNA coding algorithm based on four segmentation theory,which fully considers the statistical characteristics between multimedia data and bases,directly converts the original data into DNA base sequence,effectively improves the logical storage density of data and improves the efficiency of DNA coding.The DNA coding algorithm can effectively encode image data and text data.(2)In the process of DNA coding,this thesis introduces quaternary low-density parity check code for error correction.The error correction code can effectively reduce the errors in DNA synthesis and sequencing.The introduction of error correction code effectively enhances the stability of DNA coding.(3)In order to further improve the logical density of DNA storage and store some image data with large amount of data.This thesis combines the above four segmentation coding,DNA error correction code and deep learning technology,and proposes a set of efficient DNA coding framework for image deep learning.The framework can restore high-quality images while effectively improving the logic density.(4)In this thesis,the coding method is integrated into the overall scheme of DNA data storage and simulated.Compared with the existing DNA information storage schemes,the proposed scheme has higher logical storage density and can be decoded successfully. |
