Abstract: Discrete Cosine Transforms in a radix-2 block transform method enables true split and merge transform processing of equal sized data halves in the transform domain for enabling recursive merges and splits in transform domain without data degradation. Input data in the time domain or spatial domain during either the split and merge radix-2 forward transform processing is firstly transform processed followed by combinational processing. In the split transform process, whole transformed data is split using combinational processing into first and second transformed data halves in the transform domain as a true split. In the merge transform process, first and second transform halves are merged using combinational processing into a merged transformed data whole in the transform domain. In either case, time or spatial domain input data can be transformed into the transform domain in the form of split halves or merged whole.

Abstract: This specification discloses a process and system for processing a digitally stored image on a digital computer. The system scans and digitizes an image, separate text from non-text components, enhances and deskews the image, compresses the resulting image file, and stores the enhanced, deskewed, and compressed file for later transmission, optical character recognition, or high quality printing or viewing of the image.

Abstract: A special purpose processor and method of computation for performing an N-length real-number discrete cosine transform (DCT). The algorithm is called the split-radix discrete cosine transform because an Nth order DCT consists of an (N/2)th order DCT and two (N/4)th order inverse DCT (IDCT), where N is an integer power of 2 and larger than 4. In the embodiments of the invention, signal flow-graphs for both the forward and inverse scaled DCT have been implemented based on fused multiply and add operations in pipeline architectures. In the two-dimensional cases, over 20% arithmetic operations are saved compared to other known fast two-dimensional DCT algorithms. In addition, the split-radix DCT method offers flexibility of processing different input sizes under software control.

Abstract: A method of enlarging or reducing the size of a halftone one bit per pixel raster image without changing the screen pitch is disclosed. First, there is a determination of whether the current byte is on an image edge using any edge detection means. If the current byte is on an edge, the nearest neighbor algorithm is applied to the original image on a bit by bit basis to prevent edge jaggedness. If the current byte is not on an edge, the nearest neighbor algorithm is applied on a byte by byte basis to prevent Moire patterns in the solid areas of the image.