Research On Medical Image Security Transmission Based On New Compressed Sensing Technology

With the rapid development of medical information level and the acceleration of population aging,the storage and maintenance of medical data in medical institutions need to cost a lot.In the face of the continuous growth of massive image data,data storage,transfer,information security and full utilization are issues worthy of attention of every medical service institution.The information security of medical images is also a key issue that can not be ignored.Relevant research technology experts and scholars around the world are actively trying to mine more potential information from medical image data.The application of computer technology in the field of medical image makes medical service institutions better serve patients.In order to ensure the information security of medical images in the storage media,prevent the leakage of personal,institutional and national medical and health related information,and protect the information security of medical images,we need to pay close attention to it.Compressed sensing(CS)technology is a signal processing technology different from the traditional way.It only needs to ensure that the signal can be sparse in a certain domain,then it can be compressed by CS technology.The dimension of traditional compressed sensing measurement matrix needs to meet the reconstruction conditions and match with the signal matrix.When the dimension of the signal matrix is large,the dimension of the measurement matrix also needs to increase,which makes the number of elements in the measurement matrix increase exponentially.The time and storage space required for constructing some measurement matrices increase greatly.The compressed sensing technology based on p-tensor product breaks through the bottleneck of traditional compressed sensing technology.The measurement matrix is constructed based on tensor product,which reduces the storage space of measurement matrix and speeds up the construction of measurement matrix.Although compressed sensing technology can realize the image compression and encryption process at the same time,medical image involves the patient’s personal privacy,so the encryption of medical image needs higher security,so it needs to use additional image encryption algorithm to encrypt the image.In order to meet the requirements of spatial efficiency,transmission efficiency and image information security in the storage and transmission scenarios of medical images,this paper proposes the compressed sensing model based on P tensor product for medical image security transmission technology.The main achievements and innovations of this paper are as follows:(1)This paper proposes a p-tensor product compressed sensing model,which is applied to medical image compression and transmission scene.In the p-tensor product compressed sensing model,the sender and the receiver of data do not need to save a measurement matrix with large dimension.They only need to save the parameters that generate the chaotic matrix,which can save storage space and speed up the measurement.The construction of matrix can speed up the process of data compression.Compressed sensing can bring a certain encryption effect for the image.The new compressed sensing technology reduces the requirements of storage space and the amount of data to be transmitted,and speeds up the process of data transmission to the receiver.(2)This paper proposes a reconstruction algorithm that can adapt to the p-tensor product compressed sensing model,and determines the iterative weights and corresponding parameters.When the iterative weight in lρ-norm and ρ=0.8,the reconstruction effect of the iterative reweighted least square(IRLS)algorithm in low sampling rate image is the best.IRLS algorithm can adapt to the performance degradation caused by the reduction of the dimension of the initial measurement matrix,ensure that the reconstructed image has good quality,and the algorithm has stability.Compared with the orthogonal matching pursuit(OMP)algorithm,the IRLS algorithm needs less sampling data on the premise of the same reconstructed image quality,which can reduce the number of data transmitted and speed up the process of data transmission.(3)An image encryption technology based on double chaotic system is proposed.The proposed image encryption technology has sufficient key space and high key sensitivity.It can achieve very superior results in information entropy,adjacent pixel correlation,image gray distribution histogram and so on,which brings sufficient information security for the storage and transmission of medical images.