Method and apparatus for progressive encoding for text transmission

1. A method of encoding an image comprising:

analyzing, by a computer having one or more processors, colors and spatial features of pixels of the image to identify a text region of the image that is separate from a picture region of the image;

generating, by the computer, for a portion of the text region, a mask dividing the portion into background pixels and a plurality of text pixels, wherein the background pixels are identified as pixels with a constant color, the plurality of text pixels are identified as pixels contrasting the constant color, and the plurality of text pixels comprise a plurality of colors;

analyzing chrominance values of the plurality of colors to determine a text chrominance;

generating, for each text pixel of the plurality of text pixels, a text pixel value to generate a plurality of text pixel values, wherein each text pixel value is based on a luminance of a text pixel for which it was generated; and

transmitting an encoding of the portion comprising an encoding of each of the constant color, the mask, the text chrominance and the plurality of text pixel values.

2. The method of claim 1 , further comprising, after transmitting the encoding, transmitting a subsequent encoding comprising an encoding of a color of each text pixel of the plurality of text pixels, wherein a combination of the encoding of the mask, the encoding of the constant color and the encoding of the color of each of the plurality of text pixels constitutes a lossless encoding of the portion.

3. The method of claim 1 , wherein the plurality of text pixels comprises a first plurality of luminance levels and the text pixel values comprises a second plurality of luminance levels with fewer luminance levels than the first plurality of luminance levels.

4. The method of claim 1 , further comprising modifying the mask to identify at least one of (i) one of the pixels having the constant color as one of the text pixels of the plurality of text pixels, or (ii) one of the pixels contrasting the constant color as one of the background pixels, wherein modifying the mask reduces a size of the encoding of the mask over an absence of modifying the mask.

5. The method of claim 2 wherein transmitting the subsequent encoding is based on comparing a subsequent image, in a stream of images comprising the image, with the image to determine that a spatially equivalent portion, to the portion, in the subsequent image has the same pixel values as the portion.

6. The method of claim 1 , wherein the constant color is a background color comprising a single set of a red color value, a green color value and a blue color value, and identifying the background pixels comprises determining pixel density information of the constant color in the image.

7. The method of claim 1 , wherein the text chrominance comprises two values and a decoding of the encoding of the portion generates a plurality of decoded text pixels, wherein each decoded text pixel of the plurality of decoded text pixels comprises a red, a green and a blue value determined from a luminance level of a text pixel value of the plurality of text pixel values and the two values of the text chrominance.

8. The method of claim 1 , where generating the text pixel value for each text pixel of the plurality of text pixels comprises selecting a luminance level from a set of luminance levels that is selected based on a transmission bandwidth available for transmitting the encoding of the portion.

9. The method of claim 1 , further comprising, after transmitting the encoding of the portion, transmitting a subsequent encoding comprising an encoding of a luminance level of each text pixel of the plurality of text pixels, wherein a decoding of the subsequent encoding, in combination with the encoding of the portion, generates a more accurate facsimile of the text pixels over only decoding of the encoding of the portion.

10. The method of claim 1 , further comprising generating a subsequent encoding of the image comprising an indication of repeating, from the encoding of the portion, a use of at least one of the constant color, the mask or the text chrominance.

11. The method of claim 1 , further comprising converting the plurality of text pixels from a red, green and blue color space to a chrominance and luminance color space.

12. The method of claim 1 , wherein each text pixel value of the plurality of text pixel values is a quantization of a luminance of an associated text pixel of the plurality of text pixels.