Patent US 4,959,868

There is an image processing system comprising: a storage device to store the binarized image; a dividing circuit to divide the storage space in the storage device into rectangular meshes of a predetermined size in which at least one character is included; a discriminating circuit to discriminate the image area included in the mesh; and an encoder to compress and encode the image area on the bases of the first or second encoding method in accordance with the result of the discrimination of the discriminating circuit. This system may include the first or second decoder to decode the image information in correspondence to the first or second encoder. The second encoding method is based on a dot level. When continuous white or blank meshes exist in the image area, namely, no black pixel exists in the image area, the run length, i.e., the number of blank meshes is encoded. With this system, the binarized image signal can be data compressed and stored with a high degree of efficiency and can be decoded as necessary, and a data compressibility is improved.