| 摘要 |
An image compression and expansion apparatus obtains an average value of a single block (8x8 pixels) in a first matrix (64x64 pixels) of the original image data, to generate reduced image data of a second matrix (8x8 pixels). A sub-matrix, which is displaced from the second matrix by a predetermined amount, is obtained in such a manner that a pixel of the second matrix is contained in the sub-matrix. 8x8 pixels of the sub-matrix are subjected to a two dimensional discrete cosine transformation, to obtain reduced DCT coefficient data. The reduced DCT coefficient data is subjected to an expanded inverse discrete cosine transformation to obtain expanded image data of a matrix (mxn pixels). The displacement direction and the displacement amount of the sub-matrix relative to the second matrix, and the values of "m" and "an" are selected in such a manner that the expanded image data becomes as close to the original image data as possible.
|
