JP2020098493A - Image processing device - Google Patents

Abstract

To provide an image processing device capable of controlling heating value of a device equal to or lower than an upper limit while preventing quality from degrading even when a plurality of related processing are executed simultaneously.SOLUTION: A device comprises: a plurality of processing units for executing a series of processing relating to predetermined image processing; and a control unit for controlling at least one processing performance of the plurality of processing units so that the total consumption power of the plurality of processing units becomes equal to or less than a predetermined threshold. The control unit selects at least one piece of means from a plurality of means causing processing throughput in the processing units to reduce, reduces the processing performance by the selected means, and controls so that the total consumption power becomes equal to or less than the predetermined threshold.SELECTED DRAWING: Figure 5

Description

The present invention relates to an image processing device.

In Patent Document 1, the internal hardware module is stopped so as not to exceed the SOC (System on a Chip) package withstand voltage and heat capacity, the internal frequency of the CPU is lowered, and the L2 cache is dynamically turned on/off. There is disclosed an SOC that suppresses power consumption and performs job control so that performance can be maintained as it is while the package withstand voltage is not exceeded.

JP, 2007-157068, A

It is an object of the present invention to provide an image processing apparatus capable of controlling the heat generation amount of the apparatus to be equal to or lower than the upper limit while suppressing deterioration of quality even if a plurality of related processes are simultaneously executed.

The image processing apparatus according to the first aspect is configured such that a plurality of processing units that execute a series of processes related to a predetermined image processing and a total power consumption of the plurality of processing units are equal to or less than a predetermined threshold value. And a control unit that controls at least one processing capacity of the plurality of processing units, the control unit selecting and selecting at least one unit from a plurality of units that reduce the throughput of the processing in the processing unit. In addition, the processing capacity is reduced by the means, and the total power consumption is controlled to be equal to or less than a predetermined threshold value.

An image processing apparatus according to a second aspect is the image processing apparatus according to the first aspect, wherein the plurality of means include means for decreasing the throughput by increasing the number of clocks required for processing in the processing section. It is a thing.

An image processing apparatus according to a third aspect is the image processing apparatus according to the second aspect, wherein the control unit reduces the throughput by increasing the number of clocks required for processing one pixel.

An image processing apparatus according to a fourth aspect is the image processing apparatus according to the first aspect, wherein the plurality of means reads the image information from a storage unit in which image information to be subjected to the image processing is stored. A unit for reducing the throughput by reducing the amount of the image information read per unit time is included.

An image processing apparatus according to a fifth aspect is the image processing apparatus according to the fourth aspect, wherein the control unit lengthens a read interval when reading from the storage unit for each predetermined unit information amount The amount of the image information to be read out is reduced.

An image processing apparatus according to a sixth aspect is the image processing apparatus according to the fourth aspect, wherein the control unit reduces the unit information amount when reading out from the storage unit for each predetermined unit information amount, and The amount of the image information read out per unit time is reduced.

An image processing apparatus according to a seventh aspect is the image processing apparatus according to the first aspect, wherein the plurality of means include at least one parallel processing of a plurality of parallel processing units that performs one processing in parallel and another parallel processing. Means for reducing the throughput by executing the processing after completion of the above.

An image processing apparatus according to an eighth aspect is the image processing apparatus according to any one of the first to seventh aspects, calculates the total power consumption according to the content of the received image processing, and calculates the calculated total consumption. The control unit further includes a determination unit that determines whether the power exceeds the threshold, and the control unit determines the total power consumption in advance when the determination unit determines that the total power consumption exceeds the threshold. The control is performed so as to be the threshold value or less.

An image processing apparatus according to a ninth aspect is the image processing apparatus according to the eighth aspect, wherein the control unit determines by the determination unit when the determination unit determines that the total power consumption exceeds the threshold. According to the result, a priority order is given to the plurality of means, and at least one means is selected from the plurality of means according to the priority order.

According to the first aspect, it is possible to provide the image processing apparatus capable of controlling the heat generation amount of the apparatus to be equal to or lower than the upper limit while suppressing the deterioration of quality even if a plurality of related processes are simultaneously executed. Produce an effect.

According to the second aspect and the third aspect, there is an effect that the processing throughput can be reduced regardless of access to the storage unit.

According to the fourth aspect, the fifth aspect, and the sixth aspect, there is an effect that fine adjustment can be performed to the extent of reduction in throughput.

According to the seventh aspect, there is an effect that the throughput can be reduced in units of a large reduction degree.

According to the eighth aspect, it is possible to more reliably control the total power consumption to be equal to or less than the threshold, as compared with the case where the total power consumption is controlled to be equal to or less than the predetermined threshold during the image processing. Has the effect.

According to the ninth aspect, there is an effect that at least one means can be selected from the plurality of means in a shorter time, as compared with the case where priority is not given to the plurality of means.

It is a block diagram showing an example of composition of an image processing device concerning an embodiment. In the first throughput reduction method according to the embodiment, (a) is a configuration diagram, (b) is a diagram illustrating a reduction in a power peak value, and (c) is a diagram showing original conditions of each process. In the second throughput reduction method according to the embodiment, (a) is a configuration diagram, (b) is a diagram for explaining a reduction in a power peak value, and (c) is a diagram showing an original condition of each process. In the third method of reducing throughput according to the embodiment, (a) is a configuration diagram, (b) is a diagram illustrating a reduction in a power peak value, and (c) is a diagram showing original conditions of each process. It is a flow chart which shows a flow of processing of a peak power reduction processing program concerning an embodiment. 6A and 6B are diagrams illustrating a temporal change in electric power in image processing of the image processing apparatus according to the embodiment, FIG. 7A is a diagram showing movement of a sheet in each processing, and FIG. 9B is each processing corresponding to FIG. FIG. 4C is a diagram showing the electric power in FIG. 4C, and FIG. 7C is a diagram showing a comparison between the electric power shown in FIG.

Hereinafter, embodiments for carrying out the present invention will be described in detail with reference to FIGS. 1 to 6.
In the present embodiment, in the image processing apparatus according to the present invention, the image information (image data) input from the image reading apparatus is subjected to image processing corresponding to the operation mode and output to the image forming apparatus. The case will be described as an example. It should be noted that the “operation mode” in the present embodiment refers to the content of the processing specifically executed in the image processing apparatus 10.

As shown in FIG. 1, the image processing apparatus 10 according to the present embodiment includes an image input unit 12, an image processing unit 14, an image output unit 16, and storage units 18, 20, 22 (in FIG. 1, DDR (Double Data Rate). Further, FIG. 1 also shows a scanner 50 as an image reading apparatus as a device related to the image processing apparatus 10 and a printer 52 as an image forming apparatus.

The image input unit 12 is an interface for inputting image information read by the scanner 50, and may be called an IIT (Image Input Terminal). Below, the process in the image input part 12 may be called "IIT process."

The image output unit 16 is an interface that outputs the image information processed by the image processing unit 14 to the printer 52, and may be referred to as IOT (Image Output Terminal). Hereinafter, the process in the image output unit 16 may be referred to as “IOT process”.

The image processing unit 14 is a unit that executes a predetermined process such as noise removal on the image information input via the image input unit 12. In the following, when simply referring to “image processing”, this predetermined processing in the image processing unit 14 is referred to. The image processing apparatus 10 usually executes a plurality of types of image processing. Each image processing is mainly a memory-to-memory processing of reading image information from the storage unit, executing the processing, and storing the processed result in the storage unit.
In that sense, hereinafter, the processing in the image processing unit 14 may be referred to as “copy processing”. In the present embodiment, a plurality of processes are assumed as the image process, but needless to say, a single process may be performed.

The storage unit 18 is a storage unit used when the image information is transferred between the image input unit 12 and the image processing unit 14. The storage unit 22 is a storage unit that is used when the image information is transferred between the image processing unit 14 and the image output unit 16. The storage unit 20 is a storage unit that stores the execution result of each process performed by the image processing unit 14. The form of the storage units 18, 20, 22 is not particularly limited, but a DDR type memory is used as an example in the present embodiment.

In addition to the above respective configurations, the image processing apparatus 10 includes a CPU that controls each of the image input unit 12, the image processing unit 14, the image output unit 16, and the storage units 18, 20, and 22, and a peak power reduction process described below. A ROM that stores programs and various control programs, a RAM that expands the programs when the peak power reduction processing program is executed, and the like are provided.

Here, the process of reducing the peak value (hereinafter, “peak power”) of the power consumption (hereinafter, “power”) consumed by the process in the image processing device 10, which is executed by the image processing device 10, will be described.

The electric power consumed not only by the image processing apparatus but also by electric equipment has a limit. The inside of the electric device is usually divided into a plurality of processing units that each perform a predetermined process. In general, the electric power allowed for each processing unit (hereinafter, “electric power allowable amount”) is different. When the electric power allowable amount is exceeded in a specific processing unit, malfunctions, damages, etc. occur in the processing unit.

In order to solve the above problems, the following measures have been conventionally taken. That is, the first measure is a measure for releasing heat generated by excessive electric power. More specifically, heat is released by using, for example, a heat sink, a fan, a package having high heat dissipation performance, or the like. The second measure is a measure that does not generate heat. More specifically, a part of the processing unit is stopped by using power gating (control of power supply and cutoff), clock gating (control of supply of clock signal and cutoff), and heat generation is suppressed.

However, in the first measure, the problem that the cost increases due to the additional member is derived. In addition, the second measure is not suitable for fine control, which causes a problem that the software becomes complicated. Therefore, as a third measure, there is a case where even if all the processing units operate at the same time, a measure to reduce the copy processing is taken so as not to exceed the heat allowable amount. However, in this case, since originally necessary processing is omitted, the image quality and the function may be deteriorated.

Therefore, in the present embodiment, the operating processing unit (hereinafter sometimes referred to as “module”) is specified, and the operating module is controlled so that the electric power (that is, the heat generation amount) does not exceed the allowable amount. More specifically, at least one means is selected from a plurality of means for reducing the throughput of processing in the processing unit, the processing capacity is reduced by the selected means, and the peak power of the total power of the plurality of processing units is predetermined. It was decided to control so as to be equal to or less than the threshold value (allowable power amount). Accordingly, it is possible to provide an image processing apparatus capable of controlling the heat generation amount of the apparatus to be equal to or lower than the upper limit while suppressing deterioration of quality even if a plurality of related processes are simultaneously executed. The “throughput” generally refers to the processing capacity per unit time.

The peak power control according to the present embodiment will be specifically described with reference to FIG. FIG. 6 shows "IIT processing", "copy processing", and "IOT processing" as processing in the image processing apparatus 10, and shows temporal changes in the flow of processing and temporal changes in power. The IIT process and the IOT process are “real time processes” as shown in FIG. 1 in the sense that the processes flow (execute) in a pipeline once the processes are started, and the “copy process” is a control not shown. This is “non-real time processing” in the sense that it is executed according to the scheduling set by the unit (CPU).

In FIG. 6A, three recording media (hereinafter referred to as “paper”) on which images are formed, indicated by “p1”, “p2”, and “p3”, are indicated as “intervals between pages”. With the opening of the scanner 50, the scanner 50 sequentially reads (IIT processing), the read image information of p1, p2, and p3 is subjected to copy processing, then IOT processing is performed, and the printer 52 sequentially prints. Is shown.

FIG. 6B is a diagram showing power in each of the IIT process, the copy process, and the IOT process. FIG. 6C is a diagram showing the electric power of FIG. 6B accumulated in each time. In the present embodiment, as shown in FIG. 6C, the time series data of the accumulated power and the power allowance are compared to determine whether the peak power is less than or equal to the power allowance.

Since the throughput of the processing unit may be reduced as a result, the method for reducing the throughput is not particularly limited, but the following three methods are used as an example in the present embodiment. .. In each of the following methods, the copy processing is assumed as the processing in the image processing unit 14, and the throughput of the module that executes the copy processing is reduced so that the total power does not exceed the power allowable amount. That is, the operating power of each module is known in advance, and if the peak value of the total power of the operating modules (peak power) is predicted to exceed the power allowable amount, the throughput of one of the modules is reduced. By doing so, the peak power is suppressed.

<First method>
A first method for reducing the throughput will be described with reference to FIG. In this method, three processing units 1, 2, and 3 are assumed as the processing units to be controlled, as shown in FIG. The processing unit 1, the processing unit 2, and the processing unit 3 are a part of a plurality of processing units (for example, 20 to 30 processing units) of the image processing apparatus 10. Each processing of the processing unit 1, the processing unit 2, and the processing unit 3 reads image information from the memory 24 (denoted as “R” in FIG. 2A), executes a predetermined process, and The execution result is written in the memory 24 (denoted as “W” in FIG. 2A). The content of the processing in the processing unit is defined by using, for example, “data amount”, “throughput”, and “processing time”. The original data amount (before the reduction in throughput), throughput, and processing time of each of the processing unit 1, the processing unit 2, and the processing unit 3 are as shown in FIG. The units of the data amount, the throughput, and the processing time are, for example, [byte], [byte/second], and [second].

In this method, the processing time is lengthened and the throughput is reduced. More specifically, for example, the number of clocks required for processing per pixel included in the image information is increased to reduce the throughput. As shown in FIG. 2C, in this example, the processing times of the processing unit 1, the processing unit 2, and the processing unit 3 are Time1, Time2, and Time3, respectively, and as shown in FIG. Time2<Time3. In this example, as shown in <2> of FIG. 2B, the peak power exceeds the power allowable amount.

Therefore, in this method, as shown in <3> of FIG. 2B, the length of the processing time Time1 of the processing unit 1 and the processing time Time2 of the processing unit 2 is increased by the above method, and the time of the processing unit 3 is increased. To match. For example, if originally, Time1=(1/2).Time3, Time2=(2/3).Time3, and Time3=1, Time1 is doubled to 1 and Time2 is multiplied by 3/2 to 1 And In this case, the throughput of the processing unit 1 is 1/2 times the original capacity, and the throughput of the processing unit 2 is 2/3 times the original capacity.

As a result of performing the above, the total power of the processing unit 1, the processing unit 2, and the processing unit 3 is as shown in <4> of FIG. 2B, and the peak power of the total power is less than the power allowable amount. As a result, the power inside the module and the power of the data transfer unit with the memory 24 in the module are suppressed. This method is characterized in that the throughput can be reduced regardless of the access to the memory 24.

<Second method>
A second method for reducing the throughput will be described with reference to FIG. In this method, the number of modules in the copy process is reduced to reduce the throughput. That is, in some copy processing, processing performance is improved by simultaneously operating a plurality of pieces of hardware having the same function. When the peak power of the image processing apparatus 10 exceeds the power allowable amount, in a process in which a plurality of modules are used at the same time, a part of the operation of the modules is extended first to be a serial process. As a result, the power of the target module is suppressed and the peak power of the entire image processing apparatus 10 is controlled to be the power allowable amount or less.

In this example, as shown in FIG. 3A, the processing section 2 includes a processing section 2-1 and a processing section 2-2. The processing unit 2-1 and the processing unit 2-2 have the same hardware, and the processing unit 2-1 and the processing unit 2-2 read image information from the memory 24 in parallel, execute parallel processing, and then store the image information in the memory 24. Write. The original data amount, throughput, and processing time of each processing unit are as shown in FIG.

As shown in FIG. 3B<1>, the processing unit 2 is processed in parallel in this example, and as shown in FIG. 3B<2>, the peak power of the image processing apparatus 10 is equal to the allowable power amount. It is higher. Therefore, as shown in FIG. 3C, one processing of the processing unit 2 is first extended so as to be serial processing. In this case, the processing time of the processing unit 2 is changed from Time2 to 2×Time2 (the throughput is (1/2)·TP2). As a result, the peak power of the image processing apparatus 10 is suppressed to the power allowable amount or less as shown in <4> of FIG. In the example of <3> in FIG. 3B, 2×Time2>Time3, which means that the total processing time is longer than the original processing time, but of course the number of modules to be processed in series is adjusted. The method may be controlled so as to be within the initial processing time.

In the present embodiment, the method of changing a part of the parallel processing to the serial processing to reduce the throughput has been described as an example, but the present invention is not limited to this, and a part of the parallel processing is stopped and the remaining parallel processing is performed. Throughput may be reduced by executing the entire process. Since this method reduces the throughput in units of parallel processing, it is an effective method when it is desired to greatly reduce the throughput at one time.

<Third method>
A third method for reducing the throughput will be described with reference to FIG. In this method, the throughput is reduced by reducing the transfer amount of image information in copy processing in some modules. As shown in FIG. 4A, the configuration of the processing unit is the same as that of the processing unit shown in FIG. 2A described above, and the original data amount, throughput, and processing time in each process are also the same.

When reading data from the memory 24 in a module, a request signal RD_REQ is issued to the memory 24 via a bus (not shown) as shown in <1> of FIG. 4B. A data amount TD (number of bytes) to be transferred is attached to the request signal RD_REQ.
In response to the request of the request signal RD_REQ, the data of the specified data amount TD is sent from the memory 24 to the module as the data signal RD_DATA. In this example, as shown in <3> of FIG. 4B, the data amount TD is made constant and the interval of the request signal RD_REQ (indicated as “REQ interval” in FIG. 4B) is lengthened. As a result, the amount of data transferred per unit time decreases, and the throughput of the module decreases. The example shown in FIG. 4B <4> shows an example in which the present method is applied to the processing unit 1 and the processing unit 2. As a result, Time1 and Time2 increase and throughputs TP1 and TP2 decrease. doing.

Although the present embodiment has been described by exemplifying a mode in which the data amount TD to be transferred is constant and the REQ interval is lengthened, the present invention is not limited to this, and the data amount TD to be transferred is TD without changing the REQ interval. '(<TD). This also reduces the amount of data transferred per unit time, which reduces the throughput of the module. The present method and the above-mentioned method have a common point in that throughput is reduced by preventing data from flowing abundantly. Since this method changes the REQ interval or the data amount TD to be transferred, it is possible to finely adjust the decrease in throughput.

Next, the peak power reduction processing executed in the image processing apparatus 10 will be described with reference to FIG. FIG. 5 is a flowchart showing a processing flow of the peak power reduction processing program. This program is stored in, for example, a storage unit such as a ROM (not shown) of the image processing apparatus 10, and the CPU (not shown) reads the program from the storage unit, expands it in the RAM, and executes it.

As shown in FIG. 5, in step S100, the operation module in the current operation mode (denoted as “mode” in FIG. 5) is confirmed. The operation mode refers to the content of the process specifically executed in the image processing apparatus 10 as described above, and is set in accordance with, for example, the user's instruction. The operation module refers to, for example, the processing unit 1, the processing unit 2, and the processing unit 3 described above.

In step S102, the peak power is calculated. More specifically, in the image processing apparatus 10, the power consumption for each mode of each module is stored in advance in a storage unit such as a ROM. Then, the CPU obtains the power of each module corresponding to the set mode from the ROM and totals it to calculate the peak power.

In step S104, it is determined whether the peak power calculated in step S102 exceeds the allowable power amount. If the determination is negative, the peak power reduction processing program is terminated, while if the determination is positive, the process proceeds to step S106.

In step S106, the order of execution of the plurality of throughput reduction methods (for example, the above-mentioned first method, second method, and third method) is determined. The execution order may be created in a table format as an example. The order of execution of the throughput reduction method may be set, for example, in such a manner that the image processing quality (for example, image quality) and the function are least influential, and may be set in order from the one having the least effect on the overall processing time. It may be set to execute.

In step S108, the method of decreasing the throughput in the high order is executed.

In step S110, it is determined whether or not the peak power as a result of the execution of step S108 is less than or equal to the allowable power amount. If the determination is affirmative, the peak power reduction processing program is terminated, while if negative, the process proceeds to step S108, and the next throughput reduction method is executed. It should be noted that steps S106 to S110 are processes for determining a method of reducing the throughput for a specific processing unit so that the peak power falls within the power allowable amount, and the image processing apparatus 10 performs image processing according to a user instruction. This is a pre-process to be executed before performing.

By the above processing, when various processing is executed in the image processing apparatus 10, the peak power is suppressed to be equal to or lower than the allowable power amount.

It should be noted that in the above-described embodiment, the method of reducing the throughput in consideration of the power consumption of each processing unit corresponding to the operation mode before the actual processing in the image processing apparatus 10 has been described as an example, but the present invention is not limited to this. Absent. Immediately after executing the image processing, the peak power did not exceed the allowable power amount, but if the peak power exceeds the allowable power amount due to the influence of heat, etc., as the image processing progresses, in the middle of the image processing The peak power may be reduced by lowering the throughput of either processing unit.

1, 2 and 3 processing unit 10 image processing device 12 image input unit 14 image processing unit 16 image output unit 18, 20, 22 storage unit 24 memory 50 scanner 52 printer TD data amount

Claims (9)

A plurality of processing units that execute a series of processing related to a predetermined image processing, and at least one of the plurality of processing units so that the total power consumption of the plurality of processing units is equal to or less than a predetermined threshold value. A control unit for controlling the processing capacity,
The control unit selects at least one means from a plurality of means for reducing the throughput of processing in the processing unit, reduces the processing capacity by the selected means, and the total power consumption is equal to or less than a predetermined threshold value. Image processing device that controls so that. The image processing apparatus according to claim 1, wherein the plurality of units include a unit that reduces the throughput by increasing the number of clocks required for processing in the processing unit. The image processing apparatus according to claim 2, wherein the control unit reduces the throughput by increasing the number of clocks required for processing one pixel. When the image information is read from the storage unit in which the image information to be subjected to the image processing is stored, the plurality of units reduce the amount of the image information read per unit time to reduce the throughput. The image processing apparatus according to claim 1, further comprising means. The image processing device according to claim 4, wherein the control unit reduces the amount of the image information read per unit time by lengthening a read interval when reading from the storage unit for each predetermined unit information amount. The image processing according to claim 4, wherein the control unit reduces the unit information amount when reading from the storage unit for each predetermined unit information amount and reduces the amount of the image information read per unit time. apparatus. The plurality of units includes a unit that reduces the throughput by causing at least one parallel processing of a plurality of parallel processing units that perform one processing in parallel to be executed after another parallel processing is completed. The image processing device according to. The total power consumption is calculated according to the content of the received image processing, and further includes a determination unit that determines whether the calculated total power consumption exceeds the threshold value.
The control unit controls the total power consumption to be equal to or less than a predetermined threshold value when the determination unit determines that the total power consumption exceeds the threshold value. The image processing device according to item 1. When the determination unit determines that the total power consumption exceeds the threshold value, the control unit assigns a priority order to the plurality of units according to the determination result of the determination unit, and the control unit assigns the priority order according to the priority order. The image processing apparatus according to claim 8, wherein at least one means is selected from a plurality of means.