Recognizing text in a multicolor image

A computer-implemented method for recognizing and reproducing a background in a multicolor image stored in a computer, the method comprising:

• dividing an image into multiple blocks;
• analyzing color distributions for each of the blocks;
• identifying blocks having two main colors;
• grouping two-color blocks having similar colors into two-color zones;
• determining a background color for each two-color zone, including comparing the determined background component for a tile to the background component determined for one or more neighboring tiles not including the tile,
• if the determined background component for the tile does not match the background component determined for one or more of the one or more neighboring tiles, adjusting the background component for the tile to match the background component determined for one or the one or more neighboring tiles;
• and
• building a low resolution pixelmap representing the background of the multicolor image using the determined background component for each two-color zone.

A computer-implemented method for processing and reproducing a multicolor image represented as a pixelmap, the method comprising:

• dividing a pixelmap into a grid of tiles;
• classifying each tile as one of either monochrome, picture, or text tiles;
• assigning a background component to each monochrome tile that is equivalent to a monochrome color associated with a given monochrome tile;
• determining for each text tile a background component;
• building a low resolution pixelmap representing a background of the pixelmap using the determined background component for each text tile and the assigned background component for each monochrome tile;
• processing the text tiles with an optical character recognition process to produce recognized text;
• and
• reproducing the multicolor image including
• rendering directly each picture tile;
• rendering the low resolution pixelmap, and
• rendering the recognized text.

A computer program product, tangibly stored on a computer-readable medium, for identifying and reproducing a background of a pixelmap, the product comprising instructions operable to cause a programmable processor to:

• divide a pixelmap into a grid of tiles;
• determine for each tile a background component, including compare the determined background component for a tile to the background component determined for one or more neighboring tiles not including the tile,
• if the determined background component for the tile does not match the background component determined for one or more of the one or more neighboring tiles, adjust the background component for the tile to match the background component determined for one of the one or more neighboring tiles;
• build a low resolution pixelmap representing a background of the pixelmap using the determined background component for each tile;
• and
• store the low resolution pixelmap.