Digital Image Compression using Sparse Matrix

The multimedia applications are widely adopting in all the fields and in the day-to-day activities. All multimedia file storages are attempted to store in tiny devices with minimal memory area. The storage process and its functionalities differs with one another, but the logical processes are same. As per the file storage techniques, the file format differ one with another. The resource availability, utilizations become a challenging task to the multimedia user. Various research works are carried out in the field of file size minimization especially of images and video. Most of the file format presentation and minimization work carried out in the wavelet algorithms. The same work has been done by the researchers by using mathematical approach in which the digital image is converted into sparse matrix. However, the video compression is not at the appreciated level of the researchers. There are many possible avenues for the researches to reduce the frame image into minimized level in terms of size, storage approach and retrieval process of the application presentations. This research work attempted to reduces the image storage size via the functional process of the reduction mathematical tool approach, such as sparse matrix. In the sparse matrix reduction approach, the video image is converted into frames according to the scaling standards. The frames converted from the original accessible file format into three-dimensional layer based mathematical set. Each frames sequences is named and generated and then compared with transactional mapping set. The transactional binary set, one represents the difference of the pixel value and the zero represents the equivalent pixels. From the transactional matrix, the sparse matrix is generated and compared with the converted three-dimensional matrix. If the sparse is combined with the first conventional sparse that could generate the sequence of next frame. The overall numerical representation of the image and is size after decompression is compared to the original image model and its size.