Disclosed is a processor-implemented method for processing image data of a document using an apparatus. The processor determines pixel classifications and counts for each classification, classifies the image, selects a billing tier for which to bill the image, determines counts of pixels for marking to output the image (e.g., in device dependent space), and then can adjust from the selected billing tier to a different billing tier based on the image classification (e.g., and/or using determined counts of pixels for each classification and determined counts of pixels for marking). Counts of pixels can be adjusted using an offset, or calculated using a combination of the determined counts (e.g., combining counts from both independent and dependent spaces). By adjusting the tier at which to bill data, more accurate billing can be implemented for documents with contone or grayscale image data, for example, with greater customer satisfaction.