Defect detection for multi-function devices using machine learning
A method is disclosed. For example, the method executed by a processor of a multi-function device (MFD) includes executing a defect learning routine to identify defects, cataloging the defects based on a job function, a type of paper, and a machine state, receiving a job request, determining a known defect that has been catalogued based on the job function, the type of paper, and the machine state, and presenting a visualization of the known defect on a display of a user interface before executing the job request.
1. A method, comprising:
executing, by a processor of a multi-function device, a defect learning routine to identify defects;
cataloguing, by the processor, the defects based on a job function, a type of paper, and a machine state of the multi-function device;
receiving, by the processor, a job request;
determining, by the processor, a known defect that has been catalogued based on the job function, the type of paper, and the machine state; and
presenting, by the processor, a visualization of the known defect on a display of a user interface before executing the job request.
2. The method of claim 1 , wherein the defect learning routine comprises a diagnostic routine or a maintenance routine.
3. The method of claim 2 , wherein the diagnostic routine comprises:
generating, by the processor, a first print out of an original image to be printed on a first type of paper;
generating, by the processor, a first scanned image of the first print out;
generating, by the processor, a copy of an inverse of the first print out on a second type of paper;
generating, by the processor, a second scanned image of the copy; and
analyzing, by the processor, defects on the first print out, the first scanned image, the copy, and the second scanned image to identify defects caused by the job function and the type of paper.
4. The method of claim 3 , wherein the analyzing comprises:
comparing, by the processor, the original image to the first scanned image to identify defects generated by a print function, a scan function and the first type of paper;
comparing, by the processor, the first scanned image to the second scanned image to identify the defects generated by the first type of paper, the second type of paper, the print function, and the scan function;
identifying, by the processor, a first set of defects from the defects that are flipped as a first type of defect associated with the scan function;
identifying, by the processor, a second set of defects from the defects that are mirrored as a second type of defect associated with the print function;
identifying, by the processor, a third set of defects from the defects that stayed as a third type of defect associated with the first type of paper; and
identifying, by the processor, a fourth set of defects from the defects that are new as a fourth type of defect associated with the second type of paper.
5. The method of claim 2 , wherein the maintenance routine comprises:
identifying, by the processor, a defect due to the job function;
determining, by the processor, a change in the machine state;
identifying, by the processor, that the defect is absent on a subsequent execution of the job function; and
tagging, by the processor, the defect to be associated with the machine state before the change.
6. The method of claim 5 , wherein the job function comprises a print job and the change in the machine state comprises at least one of a print component replacement, a photoreceptor replacement, or a print component cleaning operation.
7. The method of claim 5 , wherein the job function comprises a scan job and the change in the machine state comprises a change in a full width array of an optical scanner.
8. The method of claim 2 , wherein the maintenance routine comprises:
determining, by the processor, a change in the machine state;
identifying, by the processor, a defect on a subsequent execution of the job function; and
tagging, by the processor, the defect to be associated with the machine state after the change.
9. The method of claim 8 , wherein the job function comprises a print job and the change in the machine state comprises a change in the type of paper loaded in a paper tray.
10. The method of claim 1 , wherein the visualization illustrates the known defect and presents an option for a user to continue with the job request.
11. The method of claim 1 , wherein the visualization comprises an option for a user to fix the known defect with a maintenance operation executed by the multi-function device.
12. A non-transitory computer-readable medium storing a plurality of instructions, which when executed by a processor of a multi-function device, causes the processor to perform operations comprising:
executing a defect learning routine to identify defects;
cataloguing the defects based on a job function, a type of paper, and a machine state of the multi-function device;
receiving a job request;
determining a known defect that has been catalogued based on the job function, the type of paper, and the machine state; and
presenting a visualization of the known defect on a display of a user interface before executing the job request.
13. The non-transitory computer-readable medium of claim 12 , wherein the defect learning routine comprises a diagnostic routine or a maintenance routine.
14. The non-transitory computer-readable medium of claim 13 , wherein the diagnostic routine comprises:
generating a first print out of an original image to be printed on a first type of paper;
generating a first scanned image of the first print out;
generating a copy of an inverse of the first print out on a second type of paper;
generating a second scanned image of the copy; and
analyzing defects on the first print out, the first scanned image, the copy, and the second scanned image to identify defects caused by the job function and the type of paper.
15. The non-transitory computer-readable medium of claim 14 , wherein the analyzing comprises:
comparing the original image to the first scanned image to identify defects generated by a print function, a scan function and the first type of paper;
comparing the first scanned image to the second scanned image to identify the defects generated by the first type of paper, the second type of paper, the print function, and the scan function;
identifying a first set of defects from the defects that are flipped as a first type of defect associated with the scan function;
identifying a second set of defects from the defects that are mirrored as a second type of defect associated with the print function;
identifying a third set of defects from the defects that stayed as a third type of defect associated with the first type of paper; and
identifying a fourth set of defects from the defects that are new as a fourth type of defect associated with the second type of paper.
16. The non-transitory computer-readable medium of claim 13 , wherein the maintenance routine comprises:
identifying a defect due to the job function;
determining a change in the machine state;
identifying that the defect is absent on a subsequent execution of the job function; and
tagging the defect to be associated with the machine state before the change.
17. The non-transitory computer-readable medium of claim 16 , wherein the job function comprises a print job and the change in the machine state comprises at least one of a printhead replacement, a photoreceptor replacement, or a printhead cleaning operation.
18. The non-transitory computer-readable medium of claim 16 , wherein the job function comprises a scan job and the change in the machine state comprises a change in a full width array of an optical scanner.
19. The non-transitory computer-readable medium of claim 13 , wherein the maintenance routine comprises:
determining a change in the machine state;
identifying a defect on a subsequent execution of the job function; and
tagging the defect to be associated with the machine state after the change.
20. A method, comprising:
executing, by a processor of a multi-function device, a diagnostic routine and a maintenance routine to identify defects;
cataloguing, by the processor, the defects based on a job function, a type of paper, and a machine state of the multi-function device;
receiving, by the processor, a job request;
determining, by the processor, a known defect that has been catalogued based on the job function, the type of paper, and the machine state; and
presenting, by the processor, a visualization of the known defect on a display of a user interface before executing the job request, wherein the visualization illustrates the known defect and presents an option for a user to ignore the known defect and continue with execution of the job request.