KR101175213B1 - Image compression method for multiprocessor by considering energy consumption and recording medium thereof - Google Patents

Abstract

PURPOSE: A video compression method for a multi-core processor considering of energy consumption and recording medium thereof are provided to effectively compress videos using a compression parameter. CONSTITUTION: A multi-core processor calculates compression energy consumed in video compression(110). The multi-core processor calculates transmission energy consumed in the transmission of the compressed video(130). The multi-core processor calculates non-compression transmission energy consumed in the transmission without compressing the video(150). When the compression transmission energy is smaller than the non-compression transmission energy, the multi-core processor creates the compressed video(170,190). When the compression transmission energy is bigger than the non-compression transmission energy, the multi-core processor creates a non-compression video.

Description

Image compression method for multiprocessor by considering energy consumption and recording medium

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image compression method for a multicore processor. The present invention relates to an image compression method for a multicore processor and a recording medium capable of determining whether to compress an image in consideration of energy consumption and to set an efficient compression parameter for energy consumption. .

Regarding a method of determining the quality of a compressed image, Prior Art 1 (Application No. KR 10-2007-0090466) relates to a method of determining an initial quantization parameter value, and is based on a frame rate, a screen size, and a bit rate. And a method of calculating a bit amount using the QP value and determining a quantization parameter value using a feature of the detected input image based on the bit amount.

Regarding a method of adjusting the compression rate, Prior Art 2 (Application No. KR 10-2007-0139804) relates to a compression rate adjusting system and a method for measuring the quality of an image. We propose a method to compress and encode and adjust the compression rate according to the image quality.

Prior Art 1 proposes a method using a frame rate, screen size, and bit rate to determine an efficient quantization parameter. Prior Art 2 determines an image compression quality in consideration of image compression quality, and determines an image according to the compression quality. It is merely proposing a way to control the compression rate. However, energy efficiency is an important issue when transmitting data over a wireless sensor network (WSN). In particular, as the video sensor network (VSN) for transmitting video is widely used, energy-efficient video transmission is an important problem. In general, in the case of a VSN transmitting an image, a compression algorithm is used in consideration of a bandwidth of a network. However, depending on the network configuration (bandwidth and power consumption), the use of the compression algorithm may be more energy efficient than the use of the compression algorithm.

In order to decide whether to compress or transmit such images or uncompressed images, a study result of analyzing energy consumption of image compression and transmission according to the network configuration environment was recently published. However, these findings have the disadvantage that the consideration of image compression parameter setting is insufficient. In general, since compression rate, image quality, and execution time vary according to image compression parameter settings, it is necessary to determine image compression transmission / no compression transmission by reflecting all these characteristics.

The first problem to be solved by the present invention is to provide an image compression method for a multi-core processor to compress the image in consideration of the energy consumption.

The second problem to be solved by the present invention is to provide an image compression method for a multi-core processor using a compression parameter that is efficient for energy consumption.

Further, the present invention provides a computer-readable recording medium having recorded thereon a program for executing the above method on a computer.

According to an aspect of the present invention, there is provided a method of calculating a compression energy consumed for compressing an image, calculating a transmission energy consumed for transmitting the compressed image, and transmitting the image without compressing the image. Calculating uncompressed transmission energy, comparing the compressed transmission energy, which is the sum of the compressed energy and the transmission energy, and the uncompressed transmission energy; and, as a result of the comparison, the compressed transmission energy is smaller than the uncompressed transmission energy. And generating an uncompressed image when the compressed transmission energy is greater than or equal to the uncompressed transmission energy.

According to an embodiment of the present invention, the compression energy is calculated in consideration of the compression performance time and compression consumption time according to the size and compression parameters of the image, the transmission energy is the size and compression of the image It is calculated by considering the data size after compression, network bandwidth, and transmission power consumption for transmitting the compressed image. The uncompressed transmission energy is the size of the image, the network bandwidth, and the transmission necessary for transmitting the image. The image compression method for a multicore processor may be calculated in consideration of power consumption.

In order to solve the second problem, the present invention provides a method of calculating compression energy consumed to compress an image, calculating transmission energy consumed to transmit the compressed image, and a sum of the compression energy and the transmission energy. Comprising a step of calculating a compression parameter is the minimum transmission energy, and compressing the image using the calculated compression parameters provides a video compression method for a multi-core processor.

In order to solve the above other technical problem, the present invention provides a computer readable recording medium having recorded thereon a program for executing the above-described image compression method for a multicore processor in a computer.

According to the present invention, it is possible to determine whether to compress the image in consideration of the energy consumption. In addition, according to the present invention, an image may be efficiently compressed using a compression parameter that is efficient for energy consumption.

1 is a flowchart of an image compression method for a multicore processor according to an exemplary embodiment of the present invention.
2 is a flowchart of an image compression method for a multicore processor according to an exemplary embodiment of the present invention.
3 is a block diagram of an image processing apparatus for a multicore processor according to an exemplary embodiment.
4 is a block diagram of an image compression apparatus for a multicore processor, according to an exemplary embodiment.

Prior to the description of the specific contents of the present invention, for the convenience of understanding, the outline of the solution of the problem to be solved by the present invention or the core of the technical idea will be presented first.

An image compression method for a multicore processor according to an exemplary embodiment of the present invention may include calculating compression energy consumed to compress an image, calculating transmission energy consumed to transmit the compressed image, and not compressing the image. Computing the uncompressed transmission energy consumed to transmit without, Comparing the compressed transmission energy and the uncompressed transmission energy that is the sum of the compressed energy and the transmission energy, And as a result of the comparison, the compressed transmission energy is the uncompressed transmission Generating a compressed image when the energy is smaller than the energy, and generating a compressed image when the compressed transmission energy is greater than or equal to the uncompressed transmission energy.

The configuration of the invention for clarifying the solution to the problem to be solved by the present invention will be described in detail with reference to the accompanying drawings based on the preferred embodiment of the present invention, the same in the reference numerals to the components of the drawings The same reference numerals are given to the components even though they are on different drawings, and it is to be noted that in the description of the drawings, components of other drawings may be cited if necessary. In addition, in describing the operation principle of the preferred embodiment of the present invention in detail, when it is determined that the detailed description of the known function or configuration and other matters related to the present invention may unnecessarily obscure the subject matter of the present invention, The detailed description is omitted.

1 is a flowchart of an image compression method for a multicore processor according to an exemplary embodiment of the present invention.

The image compression method according to an embodiment of the present invention may be performed in a multicore processor. Choose an efficient compression method for performing compression on multicore processors.

Step 110 is a step of calculating the compression energy consumed to compress the image.

More specifically, the energy consumption is calculated when the image is compressed to determine whether compressing the image consumes less energy or not compressing the image consumes less energy. The energy consumed when compressing an image includes the compression energy consumed to compress the image and the transmission energy consumed to transmit the compressed image. In this step, the compression energy consumed when compressing the image is calculated.

Compression energy can be calculated by considering the compression performance time and compression power consumption according to the image size and compression parameters. Energy can be calculated as the product of power consumption and time. Therefore, the compression energy can be calculated in consideration of the compression power consumption and the compression execution time. Compression execution time can be calculated using the image size and compression parameters. The compression parameter may be a quantization parameter. The quantization parameter can be set from 1 to 100. If the quantization parameter is close to 1, the compression rate and compression time increase, and the image quality and transmission time decrease. Conversely, if the quantization parameter is close to 100, the compression rate and compression time are reduced and the image quality and transmission time are increased. The quantization parameter may be 30 or more for a certain level or more of image quality. Compression execution time can be calculated by trend analysis of execution time. Trend analysis is a future prediction method that analyzes past time series data and explores the direction of change, assuming that past trends will continue. Extrapolation or projection can be used.

The compressive energy can be obtained by the following equation.

Ecomp = Ftime (IS, QP) x Wcomp

(IS: Input image size, QP: Compression parameter, Ftime (IS, QP): Trend of execution time according to IS and QP, Wcomp: Compression power consumption)

That is, the compression energy may be calculated by multiplying the execution time trend equation according to the input image size and the compression parameter with the compression power consumption. The execution time trend equation may be calculated by analyzing the execution time according to the input image size and the compression parameter of the past by the trend analysis method.

Step 130 is a step of calculating the transmission energy consumed to transmit the compressed image.

More specifically, the transmission energy may be calculated in consideration of the data size after compression, the network bandwidth, and the transmission power consumed for transmitting the compressed image according to the size of the image and the compression parameter. As described above, energy can be calculated as the product of power consumption and time. Therefore, the transmission energy can be calculated in consideration of the transmission time and the transmission power consumption. The transmission time can be calculated by dividing the data size after compression by the network bandwidth. If the file size and the file size that can be transferred per hour are determined, the file size can be calculated by dividing the file size by the file size that can be transferred per hour. The data size after compression is determined by the compression parameters. The compression parameter may be a quantization parameter. As described above, the quantization parameter may be from 1 to 100, and the closer the quantization parameter is to 1, the more the compression rate increases, and thus the data size after compression decreases. Conversely, as the quantization parameter approaches 100, the compression rate decreases, thus increasing the data size after compression. The quantization parameter may be 30 or more for a certain level or more of image quality.

The transmission energy can be obtained by the following equation.

Etran = Fbit (IS, QP) / NB x Wtran

(Fbit (IS, QP): Data size after compression according to IS and QP, NB: Network bandwidth, Wtran: Power consumption of transmission)

That is, the data after compression according to the size of the input image and the compression parameters can be divided by the network bandwidth, and the transmission energy can be calculated by multiplying the transmission power.

Step 150 is a step of calculating the uncompressed transmission energy consumed to transmit the image without compressing it.

More specifically, the energy consumption is calculated when the image is not compressed to determine whether compressing the image consumes less energy or not compressing the image consumes less energy. When the image is not compressed, energy consumption may be referred to as uncompressed transmission energy. The uncompressed transmission energy can be known by calculating only transmission energy. There is no compression because there is no compression energy. The basic method is the same as the method for calculating the transmission energy. However, since there is no compression, compression parameters are not necessary.

The uncompressed transmission energy may be calculated in consideration of the size of the input image, the network bandwidth, and the transmission power required to transmit the image. Since there is no compression, the size of the input image itself must be transmitted, and the transmission time can be calculated by dividing the size of the input image by the network bandwidth. The uncompressed transmission energy can be calculated by multiplying the calculated transmission time by the transmission power consumption.

The uncompressed transmission energy can be obtained by the following equation.

Euncomp = IS / NB x Wtran

(Euncomp: uncompressed transmission energy, IS: input image size, Wtran: transmission power consumption, NB: network bandwidth)

That is, the uncompressed transmission energy can be obtained by dividing the size of the input image by the network bandwidth and multiplying the transmission power consumption.

Step 170 is a step of comparing the compressed transmission energy and the uncompressed transmission energy, which is the sum of the compressed energy and the transmission energy.

More specifically, when comparing the compressed transmission energy, which is the energy consumed when compressing the sum of the transmission energy calculated in step 110 and the transmission energy calculated in step 130, and the uncompressed transmission energy calculated in step 150, This is a step to compare energy consumption with and without compression.

The compressed transmission energy is the sum of the compression energy and the transmission energy, and can be obtained by the following equation.

Etotal = Ecomp + Etran = Ftime (IS, QP) x Wcomp + Fbit (IS, QP) / NB x Wtran

In operation 190, a compressed image is generated when the compressed transmission energy is less than the uncompressed transmission energy, and a compressed image is generated when the compressed transmission energy is greater than or equal to the uncompressed transmission energy.

More specifically, when the compressed transmission energy is greater than the uncompressed transmission energy, it is inefficient to compress and transmit the image in terms of energy. Therefore, when the compressed transmission energy is smaller than the uncompressed transmission energy, a compressed image is generated. When the compressed transmission energy is greater than or equal to the uncompressed transmission energy, an uncompressed image is generated without compressing the image in terms of energy. Efficient

When the compression is efficient, the expression according to the compression parameter is as follows.

Etotal <Euncomp

-> Ftime (IS, QP) x Wcomp + Fbit (IS, QP) / NB x Wtran <IS / NB x Wtran

-> NB x Wcomp / Wtran <(IS-Fbit (IS, QP)) / Ftime (IS, QP)

Whether or not compression can be determined using the above equation.

2 is a flowchart of an image compression method for a multicore processor according to an exemplary embodiment of the present invention.

The image compression method according to an embodiment of the present invention may be performed in a multicore processor. Choose an efficient compression method for performing compression on multicore processors.

Step 210 is a step of calculating the compression energy consumed to compress the image.

More specifically, the compression energy consumed for compressing the image is calculated in consideration of the compression performance time and the power consumption of compression according to the image size and compression parameters. In the case of compression, it is necessary to find a compression parameter that can provide efficient compression in terms of energy. In order to calculate the compression parameter, first, a compression energy and a transmission energy should be calculated. This step is to calculate the compression energy consumed to compress the image in consideration of the compression performance time and compression consumption time according to the size and compression parameters of the image corresponding to step 110 of FIG. The detailed description of this step replaces the detailed description of step 110 of FIG. 1.

Step 230 is a step of calculating the transmission energy consumed to transmit the compressed image.

More specifically, the transmission energy consumed to transmit the compressed image is determined in consideration of the data size after compression, the network bandwidth, and the transmission power consumed to transmit the compressed image according to the size of the image and the compression parameters. Calculation step. In this step, the transmission energy is calculated to calculate the total compressed transmission energy. This step calculates the transmission energy consumed to transmit the compressed image in consideration of the data size after compression, the network bandwidth, and the transmission power consumed to transmit the compressed image according to the size of the image and the compression parameters. 1, which corresponds to step 130 of FIG. 1, the detailed description of this step replaces the detailed description of step 130 of FIG. 1.

Step 250 is a step of calculating a compression parameter that the compression transmission energy that is the sum of the compression energy and the transmission energy is the minimum.

More specifically, it is a step of calculating a compression parameter that is the minimum compression transmission energy that is the sum of the compression energy and the transmission energy. As the compression parameters increase, the compression rate and compression energy increase and the transmission energy decreases. On the contrary, as the compression parameter becomes smaller, the compression ratio and compression energy decrease and the transmission energy increases. As described above, the compression transmission energy, which is the sum of the compression energy and the transmission energy, has a minimum value in a specific compression parameter. Therefore, by calculating and applying a compression parameter that minimizes the transmission energy of compression, it is possible to perform the most efficient image compression in terms of energy. However, the quantization parameter may be 30 or more for a certain level or more of image quality. The compression energy is inversely proportional to the compression parameters and the transmission energy is proportional to the compression parameters. If the compression energy equation according to the compression parameter and the transmission energy equation according to the compression parameter are linear equations, the minimum value of the compression transmission energy can be obtained using arithmetic geometry. For example, the compression energy equation is y1 = a / x1 (y1: compression energy, a: constant, x1: compression parameter) and the transmission energy equation is y2 = bx2 + c (y2: transmission energy, b, c: constant, x2: compression Parameter), the compression transmission energy y = y1 + y2 = a / x1 + bx1 + c. Where x and y are positive. Therefore, when r1> 0 and r2> 0, if (r1 + r2) / 2 ≥ √ (r1 x r2), it can be seen that y ≥ 2b x √ (a / b) + c. It can be seen that the minimum value is 2b x √ (a / b) + c.

In step 270, the image is compressed using the compression parameter calculated in step 250.

More specifically, when the most efficient compression parameter is determined, the compression parameter is set to compress the image. Furthermore, the image may be compressed using the compression parameter. Compression methods may use DCT (Discrete Cosine Transform) and DWT (Discrete Wavelet Transform, Discrete Wavelet Transform). The compressed image may be transmitted to a destination to be transmitted by the image compression method for the multicore processor.

3 is a block diagram of an image processing apparatus for a multicore processor according to an exemplary embodiment.

The image processing terminal 300 for a multicore processor according to an exemplary embodiment of the present invention includes a calculator 310 and a processor 330.

The calculation unit 310 calculates the compression energy consumed to compress the image and the transmission energy consumed to transmit the compressed image, and calculates the uncompressed transmission energy consumed to transmit the image without compressing the image. The calculator 310 corresponds to steps 110 to 150 of FIG. 1, and the detailed description of the calculator 310 is replaced with the detailed description of steps 110 to 150 of FIG. 1.

The processor 330 compares the compressed transmission energy, which is the sum of the compression energy and the transmission energy, and the uncompressed transmission energy, which are calculated by the calculation unit 310, and as a result of the comparison, the compression transmission energy is smaller than the uncompressed transmission energy. In this case, the compressed image is transmitted, and if the compressed transmission energy is greater than or equal to the uncompressed transmission energy, the image is determined to be transmitted without compression. The processor 330 corresponds to steps 170 to 190 of FIG. 1, and the detailed description of the processor 330 is replaced with the detailed description of steps 170 to 190 of FIG. 1.

4 is a block diagram of an image compression apparatus for a multicore processor, according to an exemplary embodiment.

The image compression apparatus 400 for a multicore processor according to an exemplary embodiment of the present invention includes a calculator 410 and an image compressor 430.

The calculation unit 410 calculates the compression energy consumed to compress the image in consideration of the compression performance time and compression consumption time according to the image size and compression parameters, and calculates the compression energy consumed by the image size and compression parameters. Calculates transmission energy consumed to transmit the compressed image, taking into account the data size after compression, network bandwidth, and transmission power consumed to transmit the compressed image. Calculate the compression parameter that minimizes the compression transmission energy. The calculation unit 410 corresponds to steps 210 to 250 of FIG. 2, and the detailed description of the calculation unit 410 is replaced with the detailed description of steps 210 to 250 of FIG. 2.

The image compressor 430 compresses the image by using the compression parameter calculated by the calculator 410. The image compression unit 430 corresponds to step 270 of FIG. 2, and the detailed description of the image compression unit 430 is replaced with the detailed description of step 270 of FIG. 2.

Embodiments of the present invention may be implemented in the form of program instructions that can be executed by various computer means and recorded in a computer readable medium. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. The program instructions recorded on the medium may be those specially designed and constructed for the present invention or may be available to those skilled in the art of computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tape, optical media such as CD-ROMs, DVDs, and magnetic disks, such as floppy disks. Magneto-optical media, and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine code generated by a compiler, but also high-level language code that can be executed by a computer using an interpreter or the like. The hardware device described above may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

The present invention has been described above with reference to various embodiments. It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims. Therefore, the disclosed embodiments should be considered in an illustrative rather than a restrictive sense. The scope of the present invention is defined by the appended claims rather than by the foregoing description, and all differences within the scope of equivalents thereof should be construed as being included in the present invention.

300: image processing device
310, 410: output unit
330: processing unit
400: video compression device
430: video compression unit

Claims (5)

Calculating compressive energy consumed to compress the image;
Calculating transmission energy consumed to transmit the compressed image;
Calculating uncompressed transmission energy consumed to transmit the image without compressing it;
Comparing the compressed transmission energy that is the sum of the compressed energy and the transmission energy with the uncompressed transmission energy; And
As a result of the comparison, generating a compressed image when the compressed transmission energy is less than the uncompressed transmission energy, and generating a uncompressed image when the compressed transmission energy is greater than or equal to the uncompressed transmission energy. Image compression method. The method of claim 1,
The compression energy is calculated in consideration of the compression performance time and compression power consumption according to the size of the image and the compression parameters,
The transmission energy is calculated in consideration of the data size after compression, the network bandwidth, and the transmission power consumed for transmitting the compressed image according to the size of the image and the compression parameters.
The uncompressed transmission energy is calculated in consideration of the size of the image, the network bandwidth, and the transmission power required to transmit the image. Calculating compressive energy consumed to compress the image;
Calculating transmission energy consumed to transmit the compressed image;
Calculating a compression parameter that minimizes a compression transmission energy that is a sum of the compression energy and the transmission energy; And
And compressing the image using the calculated compression parameter. The method of claim 3, wherein
The compression energy is calculated in consideration of the compression performance time and compression power consumption according to the size of the image and the compression parameters,
The transmission energy is calculated in consideration of the data size after compression according to the size and compression parameters of the image, the network bandwidth, and the transmission power consumed to transmit the compressed image. . A computer-readable recording medium having recorded thereon a program for executing the method of claim 1 on a computer.

Images