Load assignment in image processing by parallel processing

• US 20050206920A1
• Filed: 02/28/2005
• Published: 09/22/2005
• Est. Priority Date: 03/01/2004
• Status: Active Grant

• Current Assignee: Seiko Epson Corporation (Seiko Group)
• Original Assignee: Seiko Epson Corporation (Seiko Group)
• Inventors: Yamazaki, Satoshi, Kakutani, Toshiaki, Utsunomiya, Kohei, Takata, Teruyuki
• Granted Patent: US 7,436,559 B2
• Time in Patent Office: Days
• US Class Current: 358/1.800
• CPC Class Codes: G06K 15/02 using printers
• CPC Class Codes: G06K 15/1857 involving parallel processing in the same printing apparatus
• CPC Class Codes: H04N 1/52 Circuits or arrangements for halftone screening

Abstract

In image processing carried out by means of repeated execution of process set which includes N unit processes (where N is an integer equal to 3 or greater), prior to execution of the process groups, the N unit processes are assigned to a number M (where M is an integer equal to 2 or greater, but less than N) of processing modules for executing the unit processes. First, on the basis of execution results of unit processes included in the process group executed immediately prior, there is estimated an estimated load of unit processes of each class in the next process group. Then, on the basis of the estimated load, the N unit processes are assigned to the processing modules. During this process, a selected unit process which is one of the unit processes not yet assigned to a processing module, is assigned to one of the processing modules.

Claims

1. A method of executing image processing carried out by means of repeated execution of a process set which includes N unit processes, where N is an integer equal to 3 or greater, comprising:

assigning the N unit processes of a next process set to be executed to M processing modules prior to execution of the next process set, where M is an integer equal to 2 or greater, but less than N,

the step of assigning the N unit processes including the steps of;

(a) estimating loads of respective unit process in the next process set based on execution results of unit processes included in a process set executed immediately prior to execution of the next process set; and

(b) assigning the N unit processes to the processing modules based on the estimated loads;

wherein the step (b) includes the step of

(b-1) selecting a unit process from unassigned unit processes which are not yet assigned to the processing module and assigning the selected unit process to one of the processing modules.

2. Method according to claim 1 wherein

the selected unit process is a unit process selected from among the unassigned unit processes, and other than a unit process with the smallest estimated load.

3. Method according to claim 1 wherein

the selected unit process has the largest estimated load, among the unassigned unit processes.

4. Method according to claim 1 wherein

each unit process is a halftone process for a predetermined ink color relating to an area of predetermined size in an image, ‘

the process set is a set of unit processes of mutually different ink colors, targeted to a same area of the image, and

a plurality of the process sets for adjacent areas in the image are executed in order of arrangement of the adjacent areas in the image.

5. Method according to claim 1 wherein

the M processing modules have mutually different basic ranks;

and

the step (b) further includes the step of;

(b2) prior to assigning the selected unit process, selecting one processing module to which the selected unit process is to be assigned, in accordance with an alternate assignment order which includes alternate use of a first order going from a processing module with the highest basic rank to a processing module with the lowest basic rank, and a second order going from the processing module with the lowest basic rank to the processing module with the highest.

6. Method according to claims 1 wherein

the one processing module to which the selected unit process is assigned is a processing module with the smallest total estimated load of previously assigned unit processes.

7. Method according to claim 1, wherein

the step (b) further includes the step of

(b2) based on the estimated loads, grouping the N unit processes into a first process group and a second process group, wherein the estimated load of each unit process in the second process group is smaller than that of unit processes included in the first process group;

(b3) assigning unit processes included in the first process group to the processing modules; and

(b4) after completing assignment of all unit processes included in the first process group to the processing modules, assigning unit processes included in the second process group to the processing modules;

wherein the step (b4) includes the steps of

selecting from processing modules a minimum load processing module which has the smallest total of estimated load of unit processes previously assigned; and

assigning one unit process included in the second process group that has not yet been assigned to the processing module to the minimum load processing module.

8. Method according to claim 7 wherein

each unit process is a halftone process for a predetermined ink color relating to an area of predetermined size in an image;

the process set is a set of unit processes of mutually different ink colors, targeted to a same area of the image, and

a plurality of the process sets for adjacent areas in the image are executed in order of arrangement of the adjacent areas in the image.

9. Method according to claim 8 wherein

the one unit process included in the second process group assigned to the minimum load processing module is an unit process that has the largest estimated load among unit processes included in the second process group unit processes and not yet assigned to the processing modules.

10. Method according to claim 8 wherein

the unit processes include:

a first type of unit process wherein the halftone process is performed using a first halftone method; and

a second type of unit process wherein the halftone process is performed using a second halftone method different from the first halftone method;

unit processes assigned to the processing modules in the step (b3) are executed in an assigned processing module in order in which the unit processes were assigned in the step (b3); and

the step (b3) includes the step of assigning the unit processes to the processing modules in such a way that while the first type of unit process is being executed by a processing module, the second type of unit process is executed by another processing module.

11. Method according to claim 10 wherein

the step (b3) further includes the step of assigning to the minimum load processing module a unit process having the largest estimated load among unit processes that have not yet been assigned to processing units and whose halftone method differs from that of an unit process assigned just prior.

12. Method according to claim 11 wherein

the first halftone method is an error diffusion method;

and

the second halftone method is a dither method.

13. A method of executing image processing carried out by means of repeated execution of a process set which includes N unit processes, where N is an integer equal to 3 or greater, comprising:

assigning one of the unit processes of a next process set to be executed to one of M processing modules for executing the unit processes while another of the unit processes of the process set is being executed by another of the M processing modules, where M is an integer equal to 2 or greater, but less than N, wherein

the step of assigning one of the unit processes including the steps of;

(a) estimating loads of respective unit process in the next process set based on execution results of unit processes included in a process set executed immediately prior to the execution of the next process set;

(b) selecting a unit process which is one of unassigned unit processes, the unassigned unit processes being unit processes which are not assigned to the processing modules; and

(c) assigning the selected unit process to a processing module that has completed unit processes previously assigned to.

14. Method according to claim 13 wherein

the selected unit process is a unit process selected from among the unassigned unit processes, and other than a unit process with the smallest estimated load.

15. Method according to claim 13 wherein

the selected unit process has the largest estimated load, among the unassigned unit processes.

16. Method according to claim 13 wherein

each unit process is a halftone process for a predetermined ink color relating to an area of predetermined size in an image,

the process set is a set of unit processes of mutually different ink colors, targeted to a same area of the image, and

a plurality of the process sets for adjacent areas in the image are executed in order of arrangement of the adjacent areas in the image.

17. Method according to claim 13, further comprising the step of:

(d) based on the estimated loads, grouping the N unit processes into a first process group and a second process group, wherein the estimated load of each unit process in the second process group is smaller than that of unit processes included in the first process group, wherein

the step (c) is one of the steps of.

(c1) assigning as the selected unit process an unit process in the first process group to the processing module that has completed unit processes previously assigned to; and

(c2) after completing assignment of all of unit processes in the first process group to the processing modules, assigning as the selected unit process an unit process in the second process group to the processing module that has completed unit processes previously assigned to.

18. Method according to claim 17 wherein

each unit process is a halftone process for a predetermined ink color relating to an area of predetermined size in an image;

the process set is a set of unit processes of mutually different ink colors, targeted to a same area of the image, and

a plurality of the process sets for adjacent areas in the image are executed in order of arrangement of the adjacent areas in the image.

19. Method according to claim 18 wherein

the step (b) includes the step of selecting as the selected unit process an unit process that, of the second process group unit processes not yet assigned to the processing modules, has the largest estimated load.

20. Method according to claim 18 wherein

the unit processes include:

a first type of unit process wherein the halftone process is performed using a first halftone method; and

a second type of unit process wherein the halftone process is performed using a second halftone method different from the first halftone method;

unit processes assigned to the processing modules in the step (c1) are executed in an assigned processing module in order in which the unit processes were assigned in the step (c1); and

the step (c1) includes the step of assigning the selected unit process to the processing module in such a way that while the first type of unit process is being executed by a processing module, the second type of unit process is executed by another processing module.

21. Method according to claim 20 wherein

the step (c1) includes the step of assigning a unit process having the largest estimated load, among unit processes that have not yet been assigned to processing modules and whose halftone method differs from that of an unit process assigned just prior.

22. Method according to claim 21 wherein

the first halftone method is an error diffusion method;

and

the second halftone method is a dither method.

23. A computer program product for image processing carried out by means of repeated execution of a process set which includes N unit processes, where N is an integer equal to 3 or greater, wherein

the computer program product comprising:

a computer readable medium; and

a computer program stored on the computer readable medium, the computer program comprising;

a portion for assigning the N unit processes of a next process set to be executed to M processing modules prior to execution of the next process set, where M is an integer equal to 2 or greater, but less than N, wherein

the portion for assigning the N unit processes includes;

a first portion for estimating loads of respective unit process in the next process set based on execution results of unit processes included in a process set executed immediately prior to execution of the next process set; and

a second portion for assigning the N unit processes to the processing modules based on the estimated loads,;

wherein the second portion includes

the third portion for selecting a unit process from unassigned unit processes which are not yet assigned to the processing module and assigning the selected unit process to one of the processing modules.

24. A computer program product for image processing carried out by means of repeated execution of a process set which includes N unit processes, where N is an integer equal to 3 or greater, wherein

the computer program product comprising:

a computer readable medium; and

a computer program stored on the computer readable medium, the computer program comprising;

a portion for assigning one of the unit processes of a next process set to be executed to one of M processing modules for executing the unit processes while another of the unit processes of the process set is being executed by another of the M processing modules, where M is an integer equal to 2 or greater, but less than N, wherein

the portion for assigning one of the unit processes includes;

a first portion for estimating loads of respective unit process in the next process set based on execution results of unit processes included in a process set executed immediately prior to the execution of the next process set;

a second portion for selecting a unit process which is one of unassigned unit processes, the unassigned unit processes being unit processes which are not assigned to the processing modules; and

a third portion for assigning the selected unit process to a processing module that has completed unit processes previously assigned to.

25. An image processing device for repeatedly executing a process set which includes N unit processes, where N is an integer equal to 3 or greater wherein the device comprises:

an image processing portion having M processing modules for executing the process set to perform image processing, where M is an integer equal to 2 or greater, but less than N; and

an assigning portion which is configured to assign the N unit processes of a next process set to be executed to the M processing modules prior to execution of the next process set;

wherein the assigning portion estimates loads of respective unit process in the next process set based on execution results of unit processes included in a process set executed immediately prior to execution of the next process set; and

assigns the N unit processes to the processing modules based on the estimated loads, and

wherein the assigning portion selects a unit process from unassigned unit processes which are not yet assigned to the processing module, and assigns the selected unit process to one of the processing modules, when assigning the N unit processes to the processing modules.

26. Image processing device according to claim 25 wherein the image processing module is configured to execute the process set by means of hyper-threading, and

the processing modules are threads.

27. An image processing device for repeatedly executing a process set which includes N unit processes to perform image processing, where N is an integer equal to 3 or greater wherein the device comprises:

an image processing portion including M processing modules, for executing the process set to perform image processing, where M is an integer equal to 2 or greater, but less than N; and

an assigning portion which is configured to assign one of the N unit processes of a next process set to be executed to one of the M processing modules during another of the unit processes of the process set is being executed by another of the M processing modules,

wherein the assigning portion

estimates loads of respective unit process in the next process set based on execution results of unit processes included in a process set executed immediately prior to the execution of the next process set;

selects a unit process which is one of unassigned unit processes,the unassigned unit processes being unit processes which are not assigned to the processing modules; and

assigns the selected unit process to a processing module that has completed unit processes previously assigned to.

28. Image processing device according to claim 27 wherein the image processing module is configured to execute the process set by means of hyper-threading, and

the processing modules are threads.