Method, apparatus, and non-transitory computer-readable storage medium for altering a digital image for a printing job

A non-transitory computer-readable storage medium storing computer-readable instructions that, when executed by a computer, cause the computer to perform a method for altering a digital image for a printing job, comprising:

• receiving a requested printing job including the digital image;
• performing a segmentation on the digital image;
• extracting values of properties for a segment of the segmented digital image;
• determining, based on the extracted values of the properties for the segment, whether the segment of the digital image should be altered;
• and
• altering the segment of the digital image when it is determined the segment of the digital image should be altered, a resulting altered digital image being transmitted to a printer for printing, wherein
• the determining whether the segment of the digital image should be altered includes applying a machine learning model to the extracted values of the segment of the segmented digital image, the machine learning model being trained to generate a probability that the segment should be altered.

An apparatus for altering a digital image for a printing job, comprising:

• processing circuitry configured to
• receive a requested printing job including the digital image,
• perform a segmentation on the digital image,
• extract values of properties for a segment of the segmented digital image,
• determine, based on the extracted values of the properties for the segment, whether the segment of the digital image should be altered, and
• alter the segment of the digital image when it is determined the segment of the digital image should be altered, a resulting altered digital image being transmitted to a printer for printing, wherein
• the processing circuitry is configured to determine whether the segment of the digital image should be altered by applying a machine learning model to the extracted values of the segment of the segmented digital image, the machine learning model being trained to generate a probability that the segment should be altered.