| The development of semiconductor technology has followed Moore's Law for half a century. However, this growth has encountered a bottleneck. Thus, human need a better storage device to replace the transistor. On the other hand, the development of image processing has penetrated deeply into every aspect of human activity. The memristor, as a kind of new nano-memory device has a smaller size and excellent characteristics than conventional memory devices. In this thesis, the method of image storage and compression is studied combining memristor with image processing.Firstly, the basic principle and characteristics of memristor are introduced in this thesis, including conductive mechanism of memristor and the basic model of memristor. The characteristic curve and important features of memristor are obtained by simulating in MATLAB. According to the basic model of memristor, two kinds of structure based on memristor array are proposed, one can store binary value, another can store multi-value. The read and write operations of the memristor array are realized in MATLAB. On this basis, two methods of image storage based on memristor array are proposed. The relationship between the gray value and the current is obtained which can be used to store image with memristor array. Two methods are proved to be feasible by simulating in MATLAB. Then, the method of image compression with memristor is studied and simulated in MATLAB. In addition, the image compression of memristor array is compared with the conventional memory devices. Finally, the image compression algorithm of memristor is improved which can enhance the utility of image compression with memristor.Compared with the conventional memory devices, memristor has a better storage and a higher compression ratio in image storage and compression. The theoretical basis and experimental evidence will be provided for image processing with memristor by this research. It will be used to study memristor and image processing in the future. |
PDF Full Text Request