Abstract: In an image compressing apparatus of this invention, a table of extended MH codes is prepared with respect to combinations of consecutive black/white pixels. When a portion where frequent changes between black and white occur is to be coded, consecutive black/white pixels are coded at once without measuring the run lengths of pixels. These extended MH codes are obtained by combining MH codes corresponding to black and white runs whose run lengths are determined in one byte of an original image.

Abstract: In a data processor, image data of a pixel matrix is subjected to orthogonal transformation to coefficient data, the coefficient data is quantized with a first quantization table for character image and a second quantization table for half-tone image, and the image data is discriminated to be character image data or half-tone image data according to the quantization data. The selected data is then coded. In another processor, image data are divided into first and second processing blocks of m*m and M*M pixel matrices wherein M>m and the second processing block includes one of first processing blocks and adjacent pixels. Image data in a first and second processing block are subjected to orthogonal transformation to coefficient data.

Abstract: An apparatus for expanding variable-length codes includes a decoder responsive to compressed data, into which quantized data is compressed on the basis of coding table data, for expanding the compressed data on the basis of decoding table data for variable-length codes to produce symbol data. The apparatus further includes a storage in which the decoding table data for variable-length codes is rewritably stored and a table schemer for generating the decoding table data, from the coding table data, to store the decoding table data thus generated into the storage. The decoder is adapted to consult with the decoding table data stored in the storage to expand the compressed data. The apparatus may be provided with an encoder responsive to the quantized data for compressing the quantized data on the basis of the coding table data to produce compressed data associated with the quantized data.

Abstract: Data compression apparatus is provided for compressing an input stream of n-bit data samples, possible input sample values being notionally divided into groups each containing a number of sample values equal to a respective power of two.

Abstract: An image processing apparatus includes: a first transforming device for transforming encoded data into discrete cosine transform coefficient data segmented into processing blocks, each block being formed of a matrix of m.times.n pixels; and a setting device for setting the values of M and N defining the size of a matrix forming each processing block used in inverse discrete cosine transform. The image processing apparatus also includes a data preparing circuit for adding discrete cosine transform coefficient data to, or deleting discrete cosine transform coefficient data from, a high-frequency domain of the discrete cosine transform coefficient data and segmented into processing blocks, each block being formed of a matrix of m.times.n pixels, and for preparing discrete cosine transform coefficient data segmented into processing blocks, each block being formed of a matrix of M.times.

Abstract: Two dimensional data structures are represented by quadtree codes with embedded Walsh transform coefficients. The quadtree code permits both variable block size inherent in quadtrees, and the calculational simplicity of Walsh transform descriptions of nearly uniform blocks of data. Construction of the quadtree is calculationally simple for implementation in a digital system which does a bottom-up determination of the quadtree because Walsh transform coefficients and a measure of the distortion can be recursively calculated using only Walsh transform coefficients from the previous level in the quadtree. Uniform step size quantization, which is optimal for variable length coding and generalized gaussian distributions, permits fast encoding and decoding of quadtree code.

Abstract: In quantizing and dequantizing circuity for an image data companding device, a selective inputting circuit selects either of image data undergone orthogonal transform and data undergone decoding. A multiplier multiplies the data selected by the inputting circuit by quantization coefficients or reciprocals thereof for dequantization or quantization. The quantization coefficients or the reciprocals thereof are rewritably stored in a RAM (Random Access Memory) in matching relation to input data and transferred from the RAM to the multiplier. The quantization coefficients are read out of a look-up table which stores a plurality of kinds of quantization coefficients. A ROM (Read Only Memory) stores the reciprocals of quantization coefficients beforehand. In the event of quantization, particular reciprocals matching quantization coefficients fed from the outside are read out of the ROM and transferred to the RAM.