KR100556857B1 - A method of improvement video coding with partial enhancement for rse-fgs video streaming - Google Patents

Abstract

The present invention provides a method for transmitting a video signal improved in an optimal state by adapting to a change in a state of a channel, and determining whether to transmit a video signal to an assigned channel; If it is determined that the video signal is transmitted in the above process, encoding the video signal; The standard converts the encoded video signal into a reference bit plane in the above process, selects a critical area, raises the selected critical area from the reference bit plane to a set level, and performs zero symbol processing on the upper part of the background area and the lower part of the critical area. A STANDARD SELECTIVE ENHANCEMENT process; A RECTANGULAR REGION BASED SELECTIVE ENHANCEMENT PARTIAL ENHANCEMENT process of changing a part of the lower bit plane of the critical region signal raised by the standard S process to the lowest bit plane in the process; And compressing the video signal processed in the RS process and transmitting the compressed video signal to the allocated channel, so that the clear video signal can be transmitted even when the transmittable bandwidth of the allocated channel is low and the available bandwidth is available. By utilizing the optimally has the effect of transmitting the image data signal.

Description

A METHOD OF IMPROVEMENT VIDEO CODING WITH PARTIAL ENHANCEMENT FOR RSE-FGS VIDEO STREAMING}

1 is a block diagram of a general MPEG-4 FG coder function,

2 is an explanatory diagram of data transmission by conventional source coding;

3 is an explanatory diagram of a method for encoding a video signal according to the prior art;

4 is an explanatory diagram of a S-video signal encoding concept according to the prior art;

5 is a flowchart illustrating a method for improving performance of a selection portion of a video signal blind region of the present invention;

6 is an explanatory diagram of a method for improving performance of a selection portion of a video signal blind region of the present invention;

7 is a state diagram showing the number of unit bits per bit plane of an encoded Akiyo image frame according to an embodiment of the present invention;

8 is a state diagram showing the number of unit bits per bit plane in which the standard S processing and the RSS processing of the embodiment of the present invention are performed;

9 is a state diagram of the PSN of the critical region according to an embodiment of the present invention;

10 is a state diagram of the PSN of the background region according to an embodiment of the present invention;

11 is a state diagram of a PNS of the entire video signal according to an embodiment of the present invention;

12 is a comparative quality photograph of a standard S and RS model in 20 KBPS of an image signal according to an embodiment of the present invention;

FIG. 13 is a comparative quality photograph of a standard S and RS P scheme at 40 KBPS of an image signal according to an embodiment of the present invention; FIG.

FIG. 14 is a comparative quality photograph of a standard S and RS P scheme at 60 KBPS of an image signal according to an embodiment of the present invention; FIG.

          ** Explanation of symbols on the main parts of the drawing **

20: discrete time conversion unit 30: quantum unit

40: compression section 130: encoding enhancement section

The present invention adapts to the state change of the channel in the equipment for transmitting the video signal through the Internet and transmits to the optimal state. In particular, the important part of the video signal is designated as the rectangular area, and further divided into several higher areas. Therefore, the present invention relates to a method for improving the partial area selection performance of a video signal that transmits an optimum video signal even when the line condition is poor, since the partial selection processing is appropriately adapted to the channel condition.

Means of communicating with the other party include transmitting a sound, transmitting an image, transmitting a sign, transmitting a sense, etc. In general, in a communication, audible sound Is commonly used, and in recent third generation communication systems, multi-media (MULTIMEDIA) communication to which video signals and texts are added is possible.

The multimedia communication, using a still image or a moving picture, in particular, the development of the fourth generation communication system capable of multimedia communication by a real-time video signal is currently in progress, when the real-time video communication as described above, video conferencing Services such as, remote lectures, telemedicine, and video on demand (VOD).

The video signal, especially the multimedia signal, has a large amount of data, and the system for multimedia communication, in particular, the Internet, a wireless mobile communication system, etc., has a limited bandwidth for each communication channel so that a plurality of subscribers can communicate at the same time. And a compression scheme for transmitting a multimedia signal through the limited bandwidth as described above.

In the Internet or wireless mobile communication network transmitting the video signal, the bandwidth of the corresponding channel or line is frequently changed sensitively by time and surrounding environment changes, and even if the transmitted video signal is compressed, the multimedia signal having a wide bandwidth is transmitted. Since it is difficult to transmit in real time, there is a problem that it is difficult to process reliable multimedia communication in real time.

Therefore, there is a need for the development of a method for transmitting an optimal video signal suitable for the current line or channel condition.

Hereinafter, a transmission method of a video signal according to the prior art will be described with reference to the accompanying drawings.

It is attached to explain the prior art, Figure 1 is a general diagram of the MPEG-4 F S coder functional configuration, Figure 2 is a diagram illustrating the data transmission by the conventional source coding, Figure 3 is a S video signal according to the prior art 4 is an explanatory diagram of an encoding method, and FIG. 4 is an explanatory diagram of a S-video signal encoding concept according to the prior art.

Referring to FIG. 1, a general MPEG-4 FGS encoder includes: a first adder 10 for adding an input video signal to a self-processed motion compensation signal and outputting the same; A discrete time transform unit (DCT) for converting an image signal of a space applied from the first adder 10 into an image signal having a frequency by time; A quantum unit 30 for quantizing the frequency signal applied from the discrete time conversion unit 20; A compression unit (VLC) for compressing a signal quantized and output from the quantum unit 30 and outputting it as a base video signal (BASE LAYER BIT STREAM); An inverse quantum unit (50) for inputting and inverse quantizing a signal quantized and output from the quantum unit (30); An inverse discrete time conversion unit (60) for inputting a video signal converted into a frequency signal as a signal output from the inverse quantum unit (50), converting the video signal into a spatial video signal, and outputting the converted video signal; A second adder 70 for adding and outputting a signal of the inverse discrete time converter 60 and a signal whose motion is compensated for; A limiting unit (80) for inputting the signal of the second adder (70) to suppress (CLIPPING) a signal of a predetermined magnitude or more; A memory (90) for inputting the signal of the limiting unit (80) and storing the signal as a frame unit video signal; A motion estimation unit (100) for estimating a motion vector value by processing a video signal in a frame unit applied from the memory 90 and an original video signal input; Motion compensation is performed by processing image signals in units of frames applied from the motion estimation unit 100 and the memory 90 and output to the first and second adders 10 and 70, respectively. A motion compensation unit 110; A third adder 120 for adding the signal output from the limiting unit 80 and the input original video signal; Encoding enhancement unit 130 for outputting an enhanced image signal (ENHANCEMENT BIT-STREAM) by processing the signal of the third adder 120 by BIT PLANE SHIFT, FIND MAXIMUM, and BIT PLANE VLC. It is composed of

Hereinafter, a general MPEG-4 FGS encoder having the above configuration will be described in detail with reference to the accompanying drawings.

The video signal input to the first adder 10, the third adder 120, and the motion estimator 100 is a digital video signal, and the first adder 10 converts a spatial video signal into a discrete time conversion unit ( 20) is converted into a video signal of the frequency, and the output signal of the discrete time converter 20 is input to the quantum unit 30 and quantized, so that the high frequency part is blocked or removed, and the processed signal is a compression unit. 40 and inverse quantum part 50, respectively.

The compression unit 50 compresses the digital image signal input from the quantum unit 30 by a variable length conversion (VLC) algorithm (ALGORITHM) to reduce the data amount or size and convert the base image signal (BASE). LAYER BIT-STREAM).

The inverse quantum unit 50 inversely quantizes the image signal quantized by the quantum unit 30, thereby restoring a high frequency portion, and the image signal by the frequency is applied to the inverse discrete time conversion unit 60 to generate an image in space. After the signal is converted into a signal, the motion of the frame unit applied to the second adder 70 and output from the motion compensator 110 is added to the compensated video signal and is equal to or less than a predetermined size set by the limiter 80. The frame unit video signal of which size is limited as described above is applied to the memory 90 and the third adder 120, respectively.

The memory 90 stores an image signal in a frame unit and applies the motion estimation unit 100 and the motion compensator 110 to the motion estimation unit 100, respectively. The video signal inputted in the frame unit and the original video signal are directly input and processed to estimate a motion vector value and apply the motion vector to the motion compensator 110. The motion compensator 110 provides the motion. The first adder adds a motion-compensated video signal by processing the motion vector value applied from the estimator 100 and the video signal in a frame unit applied from the memory 90. Output to 10 and the 2nd adder 70, respectively.

That is, the first adder 10 adds the video signal originally input and the signal compensated for the motion input from the motion compensator 110 and outputs the discrete signal to the discrete time converter 20. The unit 20 inputs and processes a signal with improved motion, and the second adder 70 outputs a signal with improved motion because the motion-compensated video signal is added to the video signal with improved motion. As described above, the video signal having further improved motion is suppressed to a certain size or less by the limiting unit 80 and applied to the third adder 120.

The third adder 120 adds the original video signal and the video signal having the improved motion to the encoding enhancer 130, and the encoding enhancer 130 transfers the input signal to the bit plane. (BIT PLANE SHIFT) and outputs enhanced video bit-level by means of maximum value detection and encryption.

The video signal improved by being output from the encoding enhancement unit 130 is to improve the image quality of the basic video signal output from the compression unit 40, and to improve the quality improvement function of the encoding enhancement unit 130. Therefore, the quality of the final video signal transmitted is further improved.

As described above, the general MPEG-4 FGS encoder processes the input video signal by FGS and outputs the basic video signal and the enhanced video signal, respectively. Synthetic processing is performed by the function unit connected in the following order to output the improved video signal.

Referring to FIG. 2, the transmission curve shows that the bandwidth BANDWIDTH is set to be wider or narrower due to a change in the use environment or the surrounding environment. The correlation between the signal transmission or the data transmission rate change is displayed as a curve according to the set state. The source coding indicates a stepwise state in which the data signal is transmitted according to the set bandwidth. Only shows that the data signal is transmitted.

Referring to FIG. 3, the spatial signal of the digital image signal input in the frame unit in the discrete time conversion unit (DCT) 20 is a macroblock of 16 * 16 pixels (PIXEL). MACRO BLOCK) is shown in the order of encoding or encoding in the order of the upper left to the lower right in order, the description of the encoding process as described above is shown in detail on the right.

The 16 * 16 pixel macroblock (MB) video signal is output by converting the spatial video signal into a frequency video signal by a corresponding process of the discrete time converter 20, and converting the video signal into a frequency component as described above. The signal shows that the 8 * 8 pixel block (B: BLOCK) is mainly concentrated or biased in the low frequency band.

As described above, the ZIGZAG scanning method is used in order to process the signal concentrated in the low frequency band of the upper left corner as follows, and the digital signals of the zigzag scanned blocks as described above are displayed in the respective bit planes. ) Is shown in the following sequence, and the symbols of the result of processing the digital data of each bit plane in the RUN, END OF PLANE (SYMBOL) method are shown in sequence in the following sequence. It is.

The RUN, END OF PLANE symbol processing method will be described in more detail. How many digital 0 values are detected before the first digital 1 value is detected on the corresponding BIT PLANE? Is written as the RUN value before the symbol's comma, and if a value of 1 is detected again after the detected value 1 on the corresponding bit plane, 0 is recorded after the symbol's comma. If a value of 1 is not detected again, 1 is recorded as an EOP value after the comma (,) of the symbol, and it is represented as a single symbol. Repeat until 1 is not redetected on BIT PLANE.

For example, when describing the MOST SIGNIFICANT BIT (MSB) bit plane among the symbols shown in FIG. 3, since a value of 0 is not detected before the first detected value, a RUN of the most significant bit plane symbol is performed. Since 0 is recorded as a value and 1 is not detected again on the corresponding bit plane after the detected 1, 1 is recorded as an EOP value of the symbol and represented as a symbol of (0, 1).

As another example, when describing the LAST SIGNIFICANT BIT (LSB) bit plane, since the value of 0 is not detected before the first detected value, 0 is recorded as the RUN value of the least significant bit plane symbol. Since the second 1 value is detected again after the detected 1 value, 0 is recorded as the EOP value of the symbol and the first symbol is recorded as (0,0).

Since the value of 0 is not detected before the second detected 1 value on the same least significant bit plane, the value of 0 is recorded as the run value of the second symbol, and the value of 1 again after the second detected 1 value. Since 0 is detected, the second symbol is recorded as (0,0) by recording 0 as the opi value of the second symbol.

In addition, since 26 values of 0 are detected before the third detected value 1 on the same least significant bit plane, a value of 26 is recorded as the run value of the third symbol, and a value of 1 after the third detected value 1 is obtained. Since the value is not detected, 1 is recorded as the opi value of the third symbol and the third symbol is recorded as (26,1).

Each bit plane as described above may be represented at each level.

Hereinafter, with reference to the accompanying Figure 4, the conventional SE (SE) (SELECTIVE ENHANCEMENT) video signal encoding concept will be described.

4 illustrates a concept of an SE video signal encoding process. In the upper left corner, each pixel PIXEL constituting a frame unit video signal is subjected to discrete time conversion (DCT) processing and quantum ( Q) In one embodiment, each pixel is represented by a maximum of five digits (DIGIT) or bits (BIT).

As described above, when each pixel constituting the frame-by-frame video signal is represented by five digits or bits, the frame video signal is composed of five bit planes, and the five bit planes are divided into five levels. The rectangle drawn in the middle indicates the area to be processed as an important area, and the remaining area becomes the background area.

As shown in the above, a cross-sectional view of a frame unit video signal represented by levels of five bit planes is shown at the lower left, and the inner part squared with a thick line is an important area to be processed by SE. The rest is the background area.

In the method of processing SSI, the above-described critical region is selected and transmitted before the background region, and the SE is processed, for example, to increase four levels, and constitutes a bit plane in units of frames. An ALL-ZERO SYMBOL process that fills the rest with zeros, and since the critical area has been increased by 4 levels, an all-zero symbol process is also shown in the upper right corner of the empty area at the bottom.

In the upper right figure of FIG. 4, a portion indicated by a thick line indicates an all zero symbolized portion or region, and a lower right portion of FIG. 4 shows a cross-sectional view thereof, and a background processed portion is an all zero symbol treatment. In this figure, the rectangular areas are marked as important areas, and the areas excluded from the rectangular and background processes are the background areas.

As described above, since the SE processing is performed to increase the 4 levels in order to transmit the important region first, it can be seen that the video signal becomes the bit plane (BIT-PLANE) of 9 levels (LEVEL) as a whole.

As described above, the nine-level video signal processed by the SE is sequentially transmitted from the most significant bit (MSB) or the higher level, and the video signal of the portion selected as the important region is transmitted first, and the remaining region or the background region is later If the bandwidth (BANDWIDTH), which is a case where the allocated transmission line or channel is in good condition, is transmitted, both the video signals of the critical area and the background area are transmitted, and when the bandwidth is narrow when the channel state is bad. In the above 9 levels, data that can be transmitted by the corresponding bandwidth is transmitted from a higher level.

The video signal of the first bit plane, which is the upper level, shows the largest value in the digital number while having a smaller amount of data than the video signal of the bit plane of the remaining lower level, which is the most important data for determining the quality of the video signal. The lower the level, the lower the importance.

In addition, since the four-level S (SE) process as described above, since the first four levels are all zero symbols without the important area signal and data, the data amount is smaller, the lower the importance, the greater the amount of data do.

Therefore, when the bandwidth of the allocated transmission line or the channel is poor, the signal of the critical area is transmitted first, so that the optimal video signal, that is, the range of data transmission, can be transmitted according to the channel state. At first, the most important video signal is selected and transmitted first, and the remaining signals are transmitted only when they can be transmitted, and the amount of data of the SE-processed video signal is larger than the original video signal.

Therefore, the conventional technique as described above has a problem in that the effect of improving and transmitting the video signal is not large when the bandwidth of the allocated channel is wide, and all the signals of the selected critical region are transmitted first, so that the bandwidth of the corresponding channel is sufficient. There is a problem that cannot be utilized.

In addition, when the allocated bandwidth of the channel is narrow, only an important part of the video signal is transmitted intensively, so that the overall improvement effect of the frame-by-frame video signal is not large.

The present invention improves the encoding enhancement function of equipment for transmitting video signals, such as the Internet, to make the best use of the bandwidth available for data transmission according to the state of a channel allocated for video signal transmission, and to transmit data at low speed. It is an object of the present invention to provide a method for improving the quality of the partial region partial selection of a video signal which sufficiently improves the overall picture quality of the video signal.

The present invention has been made in order to achieve the above object, the process of determining whether to transmit the video signal to the assigned channel; If it is determined that the video signal is transmitted in the above process, encoding the video signal; The standard converts the encoded video signal into a reference bit plane in the above process, selects a critical area, raises the selected critical area from the reference bit plane to a set level, and performs zero symbol processing on the upper part of the background area and the lower part of the critical area. A STANDARD SELECTIVE ENHANCEMENT process; A RECTANGULAR REGION BASED SELECTIVE ENHANCEMENT PARTIAL ENHANCEMENT process of changing a part of the lower bit plane of the critical region signal raised by the standard S process to the lowest bit plane in the process; And compressing the video signal processed in the RS process and transmitting the compressed video signal through the allocated channel.

Hereinafter, a method for improving the performance of selecting a rectangular region of a video signal according to the present invention will be described with reference to the accompanying drawings.

5 is a flowchart illustrating a method for improving the performance of the selection of the blind region of the video signal according to the present invention, and FIG. Is a state diagram showing the number of unit bits per bit plane of an encoded Akiyo video frame according to an embodiment of the present invention, and FIG. 8 is a diagram showing the unit bits per bit plane processed by the standard SY processing and the RS processing of the embodiment of the present invention. Is a state diagram of the PSN of the critical region according to an embodiment of the present invention, FIG. 10 is a state diagram of the PSN of the background region according to an embodiment of the present invention, and FIG. 11 is a video signal of an embodiment of the present invention. Fig. 12 is a comparative quality photograph of a standard S and RS method in 20 KBPS of a video signal according to an embodiment of the present invention. A comparative quality photograph of a standard S and RS sample system at 40 KBPS of an image signal according to an embodiment of the present invention, and FIG. 14 shows a standard S and RS sample at 60 KBPS of a video signal according to an embodiment of the present invention. This is a comparative picture of the method.

Hereinafter, a method for improving the performance of selecting a rectangular region of the video signal according to the present invention will be described with reference to FIG. 5.

A process of determining whether or not to transmit a video signal to the allocated channel by receiving a channel whose data transmission bandwidth changes according to a surrounding environment and a usage environment, and so on (S100);

If it is determined in step S100 that the video signal is to be transmitted, the input video signal is encoded. The video signal in a frame unit is inputted and the discrete time transformed (DCT: DISCREET COSINE TRANSFORM). Converting the spatial video signal into a video signal by frequency (FREQUENCY), removing the high frequency component of the frequency video signal by quantizing (Q: QUANTIZATION) (S110),

In the encoding process (S110), a critical region (ROI: REGION OF IMPORTANT) of the encoded video signal is processed in a standard SSE (STANDARD SE: SELECTIVE ENHANCEMENT) process, and a plurality of bit planes (referenced to the video signal) BIT-PLANE), selects and selects a critical area ROI in the form of a rectangular area, and increases the selected critical area by a predetermined level LEVEL from the reference bit plane. A process of processing all zero symbols to fill the upper part of the non-ROI or the background and the lower part of the ROI with zero digits or bits;

RSE-PE (RECTANGULAR REGION BASED SELECTIVE ENHANCEMENT PARTIAL ENHANCEMENT) processing of the standard S-processed critical area (ROI) signal in the standard S process (S120). Changing the lower bit plane (BIT-PLANE) from the lowest bit plane to the lowest bit plane (LSB) (S130),

The variable length conversion (VLE: VARIABLE LENGTH CODING) is performed by compressing the RSE-PE processed video signal in the allocated R PATH or channel CH in the S P process (S130). Compression is performed by the algorithm AGLORITHM (S140).

Hereinafter, the present invention having the above-described configuration will be described in detail with reference to FIGS. 5 to 14.

By using the encoding enhancement unit 130 of the MPEG 4 FGS (FINE GRANULARITY SCALABILITY) encoder which transmits a video signal to an internet network system or a wireless mobile communication network system, it is determined whether the video signal is input and transmitted (S100). When the video signal is input and transmitted in S100, the video signal input by the frame unit is discrete time-converted (DCT), so that the spatial video signal is converted into a video signal by frequency, and quantized to The encoding process of removing the high frequency component is performed (S110).

As an experimental example, an AKIYO image having a macroblock of 11 MB in width and 9 MB in length and a QCIF size of 176 pixels in width and 144 pixels in width is used for the experiment. In the same encoding process, six bit planes (BIT-PLANE) can be obtained, and the number of detected bits or data of each bit plane is shown in FIG.

By selecting the critical region (ROI) from the Akiyo video signal of the experimental example as described above, the selected critical region is improved by 4 levels for an experimental example, and the remaining region is all zero symbol (ALL ZERO SYMBOL). Standard S (S) processing (S120), and on the left side of the accompanying Figure 6, the standard plane is processed as shown above is shown.

A key region bit plane is partially selected from the standard S-processed video signal, and for example, two least significant bit planes are partially selected from the important region, and an RS is processed to replace the lowest zero symbol of the critical region. S130), as described above is shown in detail on the right side of Figure 6 attached to the state of the DS processing.

As described above, the video signal of which the PS processing is completed is compressed by a variable length conversion (VLE) algorithm (ALGORITHM) and transmitted to the channel (S140).

In the attached FIG. 8, a 5 MB and a 6 MB region is selected as an important region in the QCIF Akiyo image, and the number of bits or the data amount for each bit plane of the standard S processing and the RS processing is experimented. It is detected by means of a bar chart, and in more detail, the bit number or data amount which is almost the same from the first bit plane to the fourth bit plane, which is the most significant bit plane, by RSE-PE processing. Note that there is a significant difference in the number of bits or the amount of data for each of the fifth, sixth, and ninth and tenth bitplanes.

When the bandwidth of the path or channel allocated for transmitting the video signal is narrow, the data of the second bit plane is transmitted after transmitting the data of the first bit plane, and if the bandwidth of the channel is still available. If the bandwidth of the channel is still available, the process of transmitting the data of the fourth bitplane is repeated. If the channel environment is good and all the wide bandwidth is available, The tenth bitplane is sent.

In other words, the method of the present invention allows the lower bit planes having a large number of bits or data and low priority to be transmitted last, and the background area to be transmitted first, even if the critical area is allocated, so that the assigned channel changes the surrounding environment. Even in the case of using the bandwidth limited by the above, the entire high quality signal of the frame-by-frame video signal is transmitted to the maximum through the narrow bandwidth as described above.

The attached FIG. 9 shows a case in which a signal of a critical region (ROI) of a QCIF Akiyo image used in an experiment is transmitted to a corresponding channel by processing a standard SSE of the prior art and an RSE according to the present invention. -PE) In case of processing and transmitting, the available bandwidth or bitstream to power signal to noise ratio (PSNR) of the corresponding channel is measured by experiment and shown as a graph (GRAPH). It is.

Referring to FIG. 9, as an experimental example, a PSR of a channel or a bitstream of a corresponding channel is transmitted in a standard SII method according to the prior art and an RSE-PE method according to the present invention. PSNR), there is no significant difference, but from above 20 KBPS it is confirmed that the higher PSNR is detected in the standard SE method according to the prior art, that is, the RS method according to the present invention It is not able to transmit much data.

Referring to FIG. 10, the PSNR in which the data of the background region BACKGROUND, which is not selected as the ROI, is transmitted according to the bandwidth of the corresponding channel, is detected as an experimental example. There is no difference before 20 KBPS. From the above 20 KBPS or more, it can be seen that the RSE-PE method according to the present invention transmits more signals in the background region.

In other words, in the standard SE method according to the related art, all data signals of the selected critical area are transmitted first and the background area signal is transmitted later. Therefore, when the bandwidth of the allocated channel is narrow or bad, the entire frame unit The image quality of the video signal is not improved.

However, the method according to the present invention divides a signal of the ROI into an upper portion having a small amount of data and a lower portion having a large amount of data, and transmits the selected portion of the upper portion of the critical region first and then the background region in the following order. It transmits the data signal of (BACKGROUND) and transmits the lower part of the important area with a large amount of data and the lower part of the background area later.

That is, some data of the important area is transmitted first, some data of the background area is transmitted, and finally some data of the important area and some data of the background area are transmitted together.

The attached FIG. 11 shows the above-described PSNR according to the present invention and the prior art by experiment. When the bandwidth or the bitstream of the channel is in the range of 20 KBPS to 140 KBPS, the method according to the present invention provides a large amount of data. In other words, the present invention utilizes the transmittable and available bandwidth or bitstream of the corresponding channel.

12 to 14 are picture quality comparison photographs when the QCIF Akiyo image processed by the standard S method according to the related art and the RS P method according to the present invention are transmitted in respective bandwidths or bitstreams.

12 is a bandwidth or bitstream of 20 KBPS, and the conventional standard S method and the RS method of the present invention show similar quality, and FIG. 13 is a comparison of image quality in a 40 KBPS bandwidth. Although the RS method seems to be slightly inferior to the image quality of the critical region where the Akiyo character is located, it shows that the background region has been further improved and improved more clearly, and FIG. 14 shows the comparison of the image quality in the 60 KBPS bandwidth. In this important area and the background area, the image is more surely improved than the prior art.

That is, according to the present invention, the upper bit plane of the important region data signal is partially selected and transmitted first, and the upper bit plane of the background region is transmitted in the next order, and the remaining lower bit plane signal and the background region of the important region are in the following order. Since the lower bitplane signals are sequentially transmitted, the video signal is most appropriately adapted to the data transmission possible state of the allocated channel.

Therefore, the RS method according to the present invention utilizes the available bandwidth or bitstream of the allocated channel to the maximum and transmits the improved video signal as a whole.

According to the present invention having the above configuration, since the upper bit plane signals of the critical region and the background region are transmitted first, and the lower bit plane signals are transmitted later, a clear video signal can be transmitted even in a state where the transmittable bandwidth of the allocated channel is low. There is an industrial use effect.

In addition, since the image data signal is transmitted by optimally utilizing the available bandwidth of the allocated channel, there is a convenient effect of improving the image quality even at a low bandwidth at a low speed.

Claims (6)

Determining whether to transmit the video signal through the assigned channel; If it is determined that the video signal is transmitted in the above process, encoding the video signal; The standard converts the encoded video signal into a reference bit plane in the above process, selects a critical area, raises the selected critical area from the reference bit plane to a set level, and performs zero symbol processing on the upper part of the background area and the lower part of the critical area. STANDARD SELECTIVE ENHANCEMENT course, RECTANGULAR REGION BASED SELECTIVE ENHANCEMENT PARTIAL ENHANCEMENT process of changing a part of the lower bit plane of the critical region signal raised by the set level by the standard SS process to the lowest bit plane in the above process; And compressing the video signal processed in the RS process and transmitting the compressed video signal to the assigned channel. The method of claim 1, wherein the encoding process, A method for improving the quality of a partial region selection of a rectangular region of a video signal, comprising inputting a video signal, converting the discrete time, converting the video signal into a frequency video signal, and quantizing the high frequency component to remove the high frequency component. delete delete According to claim 1, Changing the lower bit plane of the critical region to the lowest bit plane. And selecting one or more bit planes from the lowest bit planes of the increased critical area and replacing them with the least significant bit planes. The method of claim 1, wherein the compressed transmission process, The method for improving partial selection quality of the rectangular region of the video signal, comprising compressing and transmitting the processed video signal by using a variable length conversion algorithm.

Images