A method of lossless image data compression including the steps of: (a) receiving an image; (b) forming a residual image by differencing adjacent samples of the received image; (c) forming multiple bitplanes directly from the residual image; (d) predicting polarities of data values of the residual image; (e) determining incorrectly predicted polarities and storing the incorrectly predicted polarities in a table; (f) encoding each of the bitplanes and the incorrectly predicted polarities of the table to form a compressed image, and (g) transmitting both the encoded bitplanes and the encoded incorrectly predicted polarities of the table.