Image Data Compression and Decompression

Sheffield Hallam University (2016). Image Data Compression and Decompression. UK patent application GB 1503433.3, 28 Feb 2016.. [Patent]

Documents
11796:36107
[thumbnail of Confirmation of patent application and invitation to apply to international PCT]
PDF (Confirmation of patent application and invitation to apply to international PCT)
Letter to Sheffield Hallam Uni_160217.pdf - Other
Restricted to Repository staff only

Download (48kB)
Abstract
A compression-encryption method is described where a unique symmetric key is automatically generated as part of the compression step and it is dependent on the characteristics of the data. The file can only be decompressed with the corresponding generated key. It is the only method that can offer compression and encryption as part of the same process. Compression rates are higher than the vast majority of existing methods making it suitable for transmission over the network with reduced bandwidth requirements, massive storage reduction especially in a Cloud environment, reduced energy costs, and faster access times. The method involves a discrete cosine (DCT) transform applied to non-overlapping variable size pixel blocks to generate a set of coefficients (DC-coefficients and AC-coefficients) for each block. Each set of coefficients is quantised resulting in a DC array and an AC matrix. The AC matrix is compressed by eliminating blocks of data having only zero values and forming a reduced AC array from blocks of non-zero values. The reduced AC array is compressed by generating a unique key that is applied to each element in the array and summed over in a particular way to form a reduced, coded AC array. Both coded AC and DC arrays are subject to arithmetic coding whose outputs are included in the compressed file together with information about the unique key for that file. If the unique key is lost, the file cannot be decompressed. The decompression method is essentially the reverse process: reverse arithmetic coding, use the key to undo the sum, recover individual elements in the AC array and reconstruct the original AC matrix. These are then assembled together with the DC coefficients and the inverse DCT is applied recovering the original data.
More Information
Share
Add to AnyAdd to TwitterAdd to FacebookAdd to LinkedinAdd to PinterestAdd to Email

Actions (login required)

View Item View Item