Image processing device generating dot data using machine learning model and method for training machine learning model
In an image processing device, a controller performs: acquiring target image data representing a target image; inputting first and second datasets into a machine learning model and causing the machine learning model to output first and second partial dot data; and generating dot data specifying a dot formation state for each of a plurality of pixels in a print image corresponding to the target image using the first and second partial dot data. The first dataset includes first partial image data and a first value for an input parameter. The second dataset includes second partial image data and a second value for the input parameter. The machine learning model outputs the second partial dot data different from the first partial dot data according to the second value being different from the first value even when the second partial image data is identical to the first partial image data.
1. An image processing device configured to perform an image process for a printer, the printer being configured to form dots on a printing medium using colorant, the image processing device comprising a controller configured to perform:
(a) acquiring target image data representing a target image, the target image having a plurality of pixels; and
(b) generating dot data specifying a dot formation state for each of the plurality of pixels in a print image corresponding to the target image, the generating in (b) comprising
(b1) inputting a first dataset into a machine learning model and causing the machine learning model to output first partial dot data, the first dataset including first partial image data and a first value for an input parameter, the first partial image data corresponding to the target image data within a first input area in the target image, the first partial dot data specifying a dot formation state for each pixel in a first output area, the first output area including a first corresponding area corresponding to the first input area in the print image,
(b2) inputting a second dataset into the machine learning model and causing the machine learning model to output second partial dot data, the second dataset including second partial image data and a second value for the input parameter, the second partial image data corresponding to the target image data within a second input area different from the first input area in the target image, the second value being different from the first value, the second partial dot data specifying a dot formation state for each pixel in a second output area, the second output area including a second corresponding area corresponding to the second input area in the print image, and
(b3) generating the dot data using the first partial dot data and the second partial dot data,
wherein the machine learning model is configured to output the second partial dot data different from the first partial dot data according to the second value being different from the first value even when the second partial image data is identical to the first partial image data,
wherein the second input area is adjacent to the first input area in the target image, the first corresponding area is adjacent to the second corresponding area in the print image,
wherein the first output area includes the first corresponding area and a first overlapping area overlapping with the second corresponding area,
wherein the second output area includes the second corresponding area and a second overlapping area overlapping with the first corresponding area, and
wherein the first output area and the second output area overlap each other with an overlapping area including the first overlapping area and the second overlapping area.
2. The image processing device according to claim 1 , wherein the overlapping area has a plurality of overlapping pixels including a first pixel and a second pixel, and
wherein the generating in (b3) sets a value of the first pixel in the dot data to a value of the first pixel in the first partial dot data, and sets a value of the second pixel in the dot data to a value of the second pixel in the second partial dot data.
3. The image processing device according to claim 2 , wherein the generating in (b3) randomly selects the first pixel and the second pixel from among the plurality of overlapping pixels in the overlapping area.
4. The image processing device according to claim 1 ,
wherein the controller is configured to further perform
(c) identifying a first region and a second region in the target image, the first region including a photograph, the second region including an image other than the photograph, and
(d) executing a halftone process on the target image data within the second region without using the machine learning model to generate the dot data corresponding to the second region, and
wherein the generating in (b) is performed on the target image data within the first region to generate the dot data corresponding to the first region.