Cell-based compression with edge detection

A non-transitory computer-readable medium, having stored thereon program instructions that, upon execution by a processor, cause the processor to perform operations comprising:

• obtaining, from a memory, an a×b pixel macro-cell from a digital image that contains 4 non-overlapping m×n pixel cells, wherein pixels in the a×b pixel macro-cell have respective pixel values and are associated with respective tags;
• determining that less than e of the respective tags indicate that their associated pixels represent edges in the digital image, wherein e is at least 1;
• based on (i) less than e of the respective tags indicating that their associated pixels represent edges, and (ii) ranges of pixel values in each of the 4 non-overlapping m×n pixel cells, selecting a compression technique, wherein selecting the compression technique comprises determining that a range of pixel values in at least one of the 4 non-overlapping m×n pixel cells exceeds a non-zero threshold value;
• applying the selected compression technique to the a×b pixel macro-cell to form a compressed representation of the a×b pixel macro-cell, wherein applying the selected compression technique comprises encoding a particular pixel in a downsampled m×n pixel cell with an average value of all pixels in a particular 2×2 block of pixels in the a×b pixel macro-cell, wherein the average value of all pixels in the particular 2×2 block of pixels does not consider whether pixels represent edges;
• and
• storing, in the memory, the compressed representation of the a×b pixel macro-cell.

A printing device comprising:

• at least one processor;
• memory;
• and
• program instructions, stored in the memory, that upon execution by the at least one processor cause the printing device to perform operations comprising:
• obtaining, from the memory, an a×b pixel macro-cell from a digital image that contains 4 non-overlapping m×n pixel cells, wherein pixels in the a×b pixel macro-cell have respective pixel values and are associated with respective tags;
• determining that less than e of the respective tags indicate that their associated pixels represent edges in the digital image, wherein e is at least 1;
• based on (i) less than e of the respective tags indicating that their associated pixels represent edges, and (ii) ranges of pixel values in each of the 4 non-overlapping m×n pixel cells, selecting a compression technique, wherein selecting the compression technique comprises (i) determining that ranges of pixel values in each of the 4 non-overlapping m×n pixel cells are less than or equal to a non-zero threshold value, (ii) determining respective average pixel values for each of the 4 non-overlapping m×n pixel cells, and (iii) determining that either (a) at least one of the ranges of pixel values is greater than a cell-level threshold value, or (b) a range of the respective average pixel values is greater than or equal to a macro-cell-level threshold value;
• applying the selected compression technique to the a×b pixel macro-cell to form a compressed representation of the a×b pixel macro-cell, wherein applying the selected compression technique comprises (i) downsampling blocks of pixels in the a×b pixel macro-cell to a downsampled m×n pixel cell, wherein the downsampling does not consider whether pixels in the a×b pixel macro-cells represent edges in the digital image, and (ii) encoding a transform of the downsampled m×n pixel cell;
• and
• storing, in the memory, the compressed representation of the a×b pixel macro-cell.