JP3494595B2 - Image transmission method and image transmission device - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image transmitting method and an image transmitting apparatus for transmitting a still image between computers.

[0002]

2. Description of the Related Art Recently, a data conference system has been proposed in which data on a computer is used as conference material. This data conferencing system is equipped with a plurality of computers equipped with communication devices, an external storage device for storing computer screens as still images, an input device for writing comments on still images, and conference software for controlling conferences. During a conference, a still image is transmitted to a partner terminal as necessary, and all participants use the input device to write comments while referring to the same still image to advance the conference. When using such a data image conferencing system on a low-speed line, it is necessary to devise a method of transmitting a still image in a short time because the conference cannot proceed smoothly because it takes a long time to transmit the still image.

A technique disclosed in Japanese Patent Laid-Open No. 5-316109 is known as a technique for smoothly proceeding with a conference using still images. This is, for example, a call function, a facsimile communication function for transmitting and receiving conference materials, a telewriting communication function for appropriately drawing still images such as conference materials, a still image conference function for proceedings of meetings,
The present invention also relates to a method of controlling a still image conference device having a document storage function of storing document information such as conference material.
This control method is such that the conference material is stored in advance in the server before the connection and the insufficient material is received from the server when the participants are connected by using the facsimile communication function.

It is also known that image transmission is performed in order to reduce the transmission time. Still image compression methods include lossy compression and lossless compression. Generally, the lossy compression method has an average compression rate for any image. It is possible to adjust the quality, and if the quality is lowered, the data size becomes smaller, but there is a problem that the image restoration property becomes low. On the other hand, the lossless compression method has a problem that the data size of a solid picture image is significantly smaller than that of the lossless compression, but there is a problem that the compression is hardly effective for a photographic image. As described above, each compression method has its own characteristics.If the compression process is fixed, the data size may not be reduced depending on the type of image, and the minute characters contained in the computer screen may not be readable due to the reduced restoration. Occur.

Japanese Unexamined Patent Publication No. 10-243131 discloses an image forming apparatus which has a condition setting function and a priority processing function and compresses an image when the image cannot be recorded and stores it in an image storage means. Have been described. In order to prevent moire at the time of lossy compression, lossy compression method and lossless compression means are provided so as to be switchable, and the image quality (binary image or multi-valued image) is judged to determine the lossless compression means. The type is automatically switched. In addition, as a method for determining the image quality of a still image, Japanese Patent Laid-Open No. 6-284281
In the publication, there is proposed a method of determining a character image and a photographic image using the peak value of the image quality histogram.

[0006]

As described above, in order to smoothly proceed with a conference using data images, still images must be transmitted in a short time. Japanese Patent Laid-Open No. 10-243
The automatic switching of the compression method described in Japanese Patent No. 131 is performed from the reversible method, and the image size at the time of transmitting the photographic image is not considered. Further, the determination is to determine whether it is a binary image or a multi-valued image, and if the still image is a color image, the image quality cannot be determined. Further, it is not possible to take advantage of the fact that reversible compression is effective for solid images even if they are not binary images.

The method disclosed in Japanese Patent Laid-Open No. 6-284281 uses a peak value of an image quality histogram to determine a character image and a photographic image. The lossless compression method does not work well.
An object of the present invention is to provide a method for transmitting a still image in a short time while suppressing deterioration of the still image.

[0008]

In order to achieve the above object, an image transmitting method according to the present invention comprises a step of determining the image quality of a still image, and an image quality among lossless compression method and lossy compression method depending on the image quality. The image compression method having the higher compression ratio with less deterioration of the image quality, the step of compressing the still image with the selected image compression method, and the step of transmitting the compressed still image. To do. The lossless compression method is, for example, PNG, and the lossy compression method is, for example, JPEG.

The step of determining the image quality of the still image can determine the image quality based on a function of the frequency difference between adjacent luminances of the luminance histogram of the still image, for example, standard deviation or average value. Here, the luminance histogram means the colors of all pixels in the image from RGB (red, green, blue) to HS.
It is a histogram that is converted into L (hue / sharpness / brightness), the maximum value of the brightness is set to 255, the minimum value is set to 0, and the appearance frequency of each brightness is arranged in order from the lowest brightness. In determining the image quality, it is typically determined whether the image is a solid picture image or a photographic image. Further, typically, a solid picture image is image-compressed by a reversible compression method, and a photographic image is image-compressed by an irreversible compression method, which is an image compression method with little deterioration of image quality and a high compression rate.

In the image quality judgment of a still image, even if a solid image is erroneously judged as a photographic image, only a slight amount of image deterioration is caused. Since this is not possible, the image transmission time is greatly affected. Therefore, when the lossless compression method is selected as the image compression method, the size of the compressed image is compared with the limit value, and if it exceeds the limit value, the image is re-imaged by the lossy compression image compression method. It is desirable to perform compression. By providing this image compression step again, the influence of the error in the image quality judgment is reduced.

The image transmitting apparatus according to the present invention is an image transmitting apparatus for compressing and transmitting a still image, which is obtained from a luminance histogram generating means for generating a luminance histogram of an image and a luminance histogram generated by the luminance histogram generating means. Solid image degree calculating means for calculating a solid image degree based on absolute value data of the difference in frequency of adjacent luminances, and lossless compression method and lossy compression method as image compression processing methods by comparing the solid image degree with a threshold value. It is characterized by comprising a compression processing selection means for selecting any one of the above, a lossless compression processing means and an irreversible compression processing means for performing image compression according to an instruction from the compression processing means. The solid image degree calculating means can determine the value obtained by dividing the standard deviation or the average value of the absolute values of the frequency difference of the adjacent brightness by the number of pixels of the still image as the solid image degree.

[0012]

BEST MODE FOR CARRYING OUT THE INVENTION An embodiment for carrying out the present invention will be specifically described below with reference to the drawings. Hereinafter, the PNG compression method will be described as an example of the reversible compression method of an image, and the JPEG compression method will be described as an example of the lossy compression method. However, the present invention is not limited to the case where the PNG compression method and the JPEG compression method are adopted as the combination of the lossless compression method and the lossy compression method, but the GIF method and the P
The present invention is also applicable to the case where a method other than the PNG compression method such as the IC2 method is adopted and a compression method other than the JPEG compression method such as the wavelet method and the fractal method is adopted as the lossy compression method.

FIG. 1 is a block diagram showing a schematic configuration of a data conference system to which an image transmitting method according to the present invention is applied. The data conference system includes a host terminal 10, a server 12, and a plurality of participant terminals 14, 16 and 18. The organizer of the conference who operates the organizer terminal 10 creates the conference material and stores it in the external storage device 11 connected to the organizer terminal 10 in advance. The conference materials also include color still images. The organizer terminal 10 transmits the meeting material stored in the external storage device 11 to the server 12 when the organizer terminal 10 is connected to the server terminal 12, and the server 12 connects the received meeting material to the server. It is stored in the external storage device 13. The organizer terminal 10 includes the image transmission processing unit shown in FIG. 2, and when transmitting the conference material to the server 12, the image transmission processing unit compresses the color still image included in the conference material and transmits it to the server 12. The participants of the conference in which the participant terminals 14, 16 and 18 are connected to the server 12 receive the conference materials from the server 12 and store them in the external storage devices 15, 17 and 19, respectively.

FIG. 2 is a block diagram for explaining the image transmission processing means included in the organizer terminal 10. The image transmission processing means 20 includes a luminance histogram creating means 22, a solid image degree calculating means 23 for calculating a solid image degree, which will be described later, a compression processing selecting means 24, an irreversible compression means 25, and a reversible compression means 2.
6. A transmission buffer 27 for storing the compressed image is provided. The image data 21 as the conference material is analyzed by the brightness histogram creating means 22 of the image transmission processing means 20, and the brightness histogram of the image data 21 is created. The solid image degree calculating means 23 calculates the solid image degree from the created brightness histogram.

The compression processing selection means 24 is an image compression processing means having a higher compression rate, out of the JPEG compression processing means (JPEG compression program) 25 and the PNG compression processing means (PNG compression program) 26, based on the calculated solid image degree. To transfer the image data to the selected image compression processing means. The selected image compression processing means, that is, the JPEG compression processing means 25 or the PNG compression processing means 26 compresses the delivered image data and stores the compressed image data in the transmission buffer 27. Transmission buffer 27
The image data stored in the transmission means 28 is operated by the organizer when the organizer terminal 10 is connected to the server 12.
Is transmitted to the server 12 via.

FIG. 3 is a diagram showing a comparison of typical compression rates and compression times of the PNG compression processing which is a reversible compression method and the JPEG compression processing which is a lossy compression method. As can be seen from FIG. 3, although PNG, which is a reversible compression method, shows a high compression rate for a solid image, it can hardly compress a photographic image. The compression time is relatively long, around 5 seconds. On the other hand, the lossy compression method JPE
G shows the same compression ratio for both a photographic image and a solid image, but the compression ratio of JPEG is only about half the compression ratio of PNG for a solid image. The compression time of JPEG is as short as 1 second. From this, it can be seen that it is advantageous to perform image compression using a lossy compression method such as JPEG when the image is a photographic image, and to perform lossless compression method such as PNG when the image is a solid image. .

FIG. 4 shows the image transmission processing means 2 shown in FIG.
6 is a flowchart for explaining a compression method switching process performed at 0 according to a procedure. First, in step 11, the image transmission processing means 20 acquires image data, and in step 12, the brightness histogram creation means 22 arranges the appearance frequencies of the brightness of each pixel in the image data in the order of decreasing brightness to create a brightness histogram. And calculate the number of colors. The number of colors appearing in the luminance histogram created in step 13 is determined. When the number of colors is extremely small, for example, 4 or less, the process proceeds from step 13 to step 17 to perform image compression by the PNG compression program. If not, proceed from step 13 to step 14,
The solid image degree is calculated based on the luminance histogram as described later.

If the solid image degree is larger than the threshold value, the process proceeds to step 17 to perform image compression with the PNG compression program, and if it is smaller, the process proceeds to step 16 to perform compression with the JPEG compression program. The threshold value in step 15 may be determined based on sample data, but is preferably about 40. In this way, the PNG compression program or the JPEG compression program having the characteristics shown in FIG. 3 is selected automatically with the higher compression rate.

However, it cannot be said that there is a case where the judgment is wrong in step 15. For example, assuming that the image to be determined to be a solid image is a photographic image in the determination in step 15, in this case, the solid image to be PNG-compressed is JPEG-compressed. But J
Although PEG has a lower compression rate than PNG, it can compress images with a relatively high compression rate of 1/12, and the processing time is as short as about 1 second, so it is not very useful for data conference management. No major impact. However, if it is determined that a solid image is a photographic image, JP
A photographic image to be EG-compressed will be PNG-compressed.
In this case, it takes about 5 seconds for the compression process, and since PNG can hardly compress the photographic image, the image transmission process by the organizer terminal 10 to the server 12 and the participant terminals 14, 16, 18 are performed. The image reception processing from the server 12 is burdened and the conference management is greatly affected.

Therefore, when the PNG compression processing is performed in step 17, the process proceeds to step 18 and it is determined whether or not the data size is reduced by the compression processing. If the data size is smaller than the predetermined limit value in the determination in step 18, it is determined that the PNG compression has been successfully performed, and the process ends. On the other hand, step 1
If the determination in step 5 fails, and as a result it is determined in step 18 that the data size is not reduced by PNG compression, the process proceeds to step 16 as post-processing to perform recompression using JPEG.

Using FIG. 5 to FIG. 7, step 1 of FIG.
The contents of the solid image degree calculation processing shown in 4 will be described. Figure 5
6 is a flow chart for explaining the procedure of the solid image degree calculation process, FIG. 6 is a schematic diagram showing an example of a luminance histogram, and FIG. 7 is a diagram showing an example of array data obtained from the luminance histogram shown in FIG.

In step 12 of FIG. 4, it is assumed that the luminance histogram as shown in FIG. 6 is created. Figure 6
In, the horizontal axis is the brightness when the darkest color is 0 and the brightest color is 255, and the vertical axis is the frequency of occurrence. Then,
In steps 21 and 22 of FIG. 5, the frequencies of the luminance histogram are acquired in order from the one with the lowest luminance or the one with the highest luminance, and the part where the frequency 0 is not continuous is searched. In steps 23 to 24, the absolute value of the difference between the frequencies of adjacent luminances is taken from that point to the point where the frequency becomes 0 next time, and the absolute values are stored in the array as shown in FIG. This is repeated until the end of the brightness (from the end of the horizontal axis in FIG. 6). By not counting the number of consecutive zeros in this way, it is possible to calculate continuity only where there are frequencies. Next, in step 25, the standard deviation of the array data created as shown in FIG. 7 is calculated, and in step 26, the obtained standard deviation is divided by the number of pixels of the entire image so as not to depend on the image size. . The value thus calculated is used as the solid image degree. In step 25, instead of the standard deviation of the array data, the average value of the array data, the root mean square of the array data, and the like, the larger the absolute value of the frequency difference between adjacent brightness histograms, the larger the value. Any function that changes to may be used.

Next, as an application example of the present invention, FIGS. 8 to 15 show various still images and luminance histograms for the images. 8 to 11 are examples of the photographic image and its luminance histogram. The luminance histogram for the photographic image of FIG. 8 is shown in FIG. 9, and the luminance histogram for the photographic image of FIG. 10 is shown in FIG. 12 to 15 are examples of images of documents and charts created by a computer, and FIG. 13 shows a luminance histogram for the chart of FIG.
The luminance histogram for the image of No. 4 is shown in FIG. In FIG. 9, FIG. 11, FIG. 13, and FIG. 15 showing the luminance histogram, the horizontal axis is the luminance and the vertical axis is the frequency of appearance.

As is clear from each image and the brightness histogram for that image, the photographic image is composed of a great number of colors having slightly different brightness, and in the brightness histogram, the frequencies of adjacent brightness are continuous. Is changing. On the other hand, the displayed images of documents and charts are
There are few types of color brightness that form an image, and the frequency of specific color brightness is by far the highest. The solid image degree of the present invention quantifies such a difference in the luminance histogram between a photographic image and another image.

FIG. 16, FIG. 8, FIG. 12, FIG.
The solid image degree of the image shown in FIG. 4 and the data size of each compression method are compared and shown. As shown in FIG. 16, for example, the image of FIG. 8 is an image of 1024 × 768 pixels, the solid image degree is 10.3, and the bitmap data size is 2.
The data size is 359.4 kB, the PNG-compressed data size is 1731.1 kB, and the JPEG-compressed data size is 52.6 kB. Similarly, the image of FIG.
An image of 24 × 768 pixels and a solid image degree of 184.
8, the data size of the bitmap is 2359.4 kB,
Data size when PNG compressed is 27.3 kB, J
The data size when PEG-compressed is 90.5 kB.

For example, if the threshold in step 15 of FIG. 4 is set to 40, the images of FIGS.
Since the solid image degree is smaller than the threshold value, it is compressed by JPEG. On the other hand, the images of FIGS. 12 and 14 are compressed by PNG because the solid image degree is larger than the threshold value. Figure 1
As can be seen by comparing the PNG size and the JPEG size listed in No. 6, both images are PNG and JP in this case.
The image is compressed by the compression method of the smaller size of EG.

[0027]

According to the present invention, the optimum image compression method can be automatically selected when transmitting a still image, and when the image is decompressed, characters are blurred.
It is possible to prevent the case where the data size does not become small.

[Brief description of drawings]

FIG. 1 is a block diagram showing a schematic configuration of a data conference system to which an image transmission method according to the present invention is applied.

FIG. 2 is a schematic diagram of image transmission processing means according to the present invention.

FIG. 3 is a typical compression rate of PNG compression and JPEG compression,
The figure which compared and showed the compression time.

FIG. 4 is a flowchart for explaining a compression method switching process step by step.

FIG. 5 is a flowchart illustrating a procedure of solid image degree calculation processing.

FIG. 6 is a schematic diagram showing an example of a luminance histogram.

FIG. 7 is a diagram showing an example of array data obtained from a luminance histogram.

FIG. 8 is a diagram showing an example of a still image.

9 is a diagram showing a luminance histogram for the image of FIG.

FIG. 10 is a diagram showing an example of a still image.

11 is a diagram showing a luminance histogram for the image of FIG.

FIG. 12 is a diagram showing an example of a still image.

13 is a diagram showing a luminance histogram for the image of FIG.

FIG. 14 is a diagram showing an example of a still image.

15 is a diagram showing a luminance histogram for the image of FIG.

FIG. 16 is a comparative diagram showing the solid image degree of a still image and the data size of each compression method.

[Explanation of symbols]

10 ... Sponsor terminal, 11 ... External storage device, 12 ... Server, 14, 16, 18 ... Participant terminal, 20 ... Image transmission processing means, 21 ... Image data, 22 ... Luminance histogram creating means, 23 ... Solid image degree Calculation means, 24 ... compression processing selection means, 25 ... lossy compression means, 26 ... lossless compression means,
27 ... Transmission buffer, 28 ... Transmission means

   ─────────────────────────────────────────────────── ─── Continued front page       (56) Reference JP-A-6-152982 (JP, A)                 JP-A-6-225160 (JP, A)                 JP-A-8-36635 (JP, A)                 Japanese Patent Laid-Open No. 10-105698 (JP, A)

Claims (2)

(57) [Claims] 1. A step of determining the image quality of a still image,
A step of selecting an image compression method with less deterioration of the image quality and a higher compression rate according to the image quality, and a step of compressing the still image with the selected image compression method. And a step of transmitting a compressed still image, wherein the step of determining the image quality of the still image includes a standard of difference in frequency of adjacent luminance of a luminance histogram of the still image.
An image transmitting method characterized in that the image quality is judged based on a deviation, an average value or a root mean square value . 2. An image transmitting apparatus for compressing and transmitting a still image, wherein a brightness histogram creating means for creating a brightness histogram of an image, and a frequency of adjacent brightness calculated from the brightness histogram created by the brightness histogram creating means. Standard deviation of absolute value of difference , mean value or square root
A solid image degree calculating means for calculating a solid image degree based on a mean value, and a compression process for comparing the solid image degree with a threshold value and selecting one of a lossless compression method and a lossy compression method as an image compression method. An image transmitting apparatus comprising: a selection unit, a lossless compression processing unit and an irreversible compression processing unit that perform image compression according to an instruction from the compression processing unit.