KR20070076243A - Method of enhancing video quality of movie file in mobile communication network and apparatus implementing the same - Google Patents

Abstract

A method for enhancing low video quality of moving pictures in a mobile communication network, and an apparatus for realizing the same are provided to secure good quality by enhancing a specific part of a picture according to an object of moving pictures. A preprocessor(120) firstly performs a quality improving process in a whole picture region of inputted moving pictures. An ROI(Region Of Interest) filter(130) secondarily performs a quality improving process in a specific interested region of the moving pictures passing through the preprocessing. A decoder(110) decodes the inputted moving pictures for transmitting the decoded moving pictures to the preprocessor when the inputted moving pictures are encoded in a predetermined video standard.

Description

Method of enhancing video quality of movie file in mobile communication network and apparatus implementing the same

1 is a block diagram illustrating an internal configuration of a video quality improving apparatus according to the present invention.

2 (a) is a flowchart illustrating an algorithm for determining a deblocking region in a deblocking filter of a video quality improving apparatus according to the present invention.

FIG. 2 (b) is a graph showing the pixel value transition at the block boundary in the image which has passed the deblocking filter of the video quality improving apparatus according to the present invention. FIG.

3 (a) to 3 (c) are graphs showing the results of band-by-band filtering required for extracting high frequency components from an edge enhancement filter of a video quality improving apparatus according to the present invention.

Figure 3 (d) is a graph showing a high frequency region pixel value of the image passed through the edge enhancement filter of the video image quality improving apparatus according to the present invention.

4 (a) to 4 (b) are histograms of images before and after passing the contrast enhancement filter of the apparatus for improving video quality according to the present invention, respectively.

5 is a flowchart sequentially showing each step of the video quality improving method according to the present invention.

6 is a block diagram of a mobile communication system when the video quality improvement method according to the present invention is applied to a video ring back tone (RBT) service of a mobile communication network.

7 is a block diagram of a mobile communication system when the method for improving video quality according to the present invention is applied to an out of office video message service (VMS) of a mobile communication network.

The present invention relates to a method for improving the image quality of a video played in a mobile communication terminal and an apparatus for implementing the same, and more particularly, a pre-processing unit for performing a quality improvement process on an entire screen area of an input video. The present invention relates to an apparatus for improving video quality including a region of interest (ROI) filter which performs a quality improvement process on a specific region of interest of a video that has been preprocessed.

Currently, various codecs (Codec: quasi-composite of encoder / decoder) are used to compress / decompress video. These codecs compress video data into smaller amounts of data to reduce the load on storage space or network. Used. There are many kinds of codecs such as MPEG, H.26x, AVI, RM, etc. Among them, 3GPP or 3GPP2, which establishes 3G mobile communication standard, is defined by H.263 as defined by the International Standards Organization (ISO or ITU-T). MPEG-4 visual and H.264 (AVC) are adopted. Among them, H.263 is the most commonly used codec for real-time communication such as mobile communication network.

H.263 is an ITU-T recommendation to compress data at very low bit rates. H.263 is basically based on H.261, an existing recommendation, and adopts a hybrid compression method that removes temporal redundancy through prediction between frames and spatial redundancy using DCT (Discrete Cosine Transform). . In case of the video compressed by H.263 method, it was confirmed that the image quality is not very good at low bitrate such as 43kbps that can be serviced in the existing voice trunk. This is due to the fact that H.263 was produced for real-time encoding purposes, and especially in the case of low bitrates.

The present invention has been proposed to solve the above problems, the user of the final mobile communication terminal by applying a correction to intensively recover close to the original for a specific portion of the screen in a low-quality video file transmitted to the mobile communication terminal An object of the present invention is to provide a means or method for allowing a user to feel high quality video.

Another object of the present invention is to provide an effect of improving image quality by providing preprocessing means for performing overall correction on the entire video screen before correcting the specific portion.

Another object of the present invention is to distinguish between a video input encoded with a mobile communication network and an original video input as an original, but to improve the correction efficiency by performing a temporary decoding process only for the former. .

The present invention for achieving the above object relates to a device for improving the video quality in a mobile communication network, the pre-processing unit and the pre-processing to perform the image quality improvement process for the entire area of the screen of the input video first It includes a ROI (Region Of Interest) filter for performing a secondary image quality improvement process for a particular region of interest of the video.

In this case, when the input video is encoded in a predetermined video standard, a decoder may be further decoded according to the corresponding video standard and transmitted to the preprocessor.

Here, the ROI filter may encode the received video according to the H.263 standard, but may perform filtering by encoding quantization coefficients of the ROI and the ROI differently in the corresponding video.

The preprocessing unit may further include a deblocking filter for correcting a boundary generated in units of blocks on the video screen, an edge improvement filter for correcting the boundary of each image constituting the video screen, and a brightness of the video screen. One or more of the contrast enhancement filter may be included. If the original video is not encoded, it may be input to any one selected from an edge enhancement filter, a contrast enhancement filter, or the ROI filter of the preprocessor.

On the other hand, the present invention for achieving the above object relates to a method for improving the video quality in a mobile communication network, the pre-processing step for performing the quality improvement process for the entire area of the input video screen and the corresponding video And a region of interest (ROI) filtering step of performing an image quality improvement process on a specific ROI.

In this case, before the preprocessing step, a decoding step of decoding the input video according to a corresponding video standard may be further included.

Here, the ROI filtering step may be performed by encoding the received video according to the H.263 standard, but encoding the quantized coefficients of the ROI and the uninterested region differently from each other in the corresponding video.

The preprocessing step may include a deblocking filtering step of correcting a boundary generated in units of blocks on the video screen, an edge improvement filtering step of correcting the boundary of each image constituting the video screen, and an increase of brightness of the video screen. It includes one or more of the contrast enhancement filtering step to correct.

On the other hand, when an unencoded original video is input, it may be performed from any one selected from an edge improvement step, a contrast improvement step, or the ROI filtering step of the preprocessing step.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. First, in adding reference numerals to the elements of each drawing, it should be noted that the same elements are denoted by the same reference numerals as much as possible even if they are shown in different drawings. In the following description of the present invention, if it is determined that a detailed description of a related known function or configuration may unnecessarily obscure the subject matter of the present invention, the detailed description thereof will be omitted.

1 is a block diagram showing an internal configuration of a video image quality improving apparatus 100 according to the present invention. Such an image quality improvement apparatus 100 includes a preprocessing unit 120 and a ROI (Region Of Interest) filter 130, and a decoder 110 may be additionally added thereto.

Meanwhile, the decoder 110 applies the image quality improvement process according to the present invention to the input video when the video inputted to the video quality improvement apparatus is encoded in a predetermined video standard such as H.263 or MPEG. It is responsible for decoding the input video before. Of course, if the input video is the original video in the unencoded state, it is not necessary to go through such a decoder 110.

The preprocessor 120 performs a primary image quality improvement process on the entire screen area of the video input through an online mobile communication network or an offline recording medium.

The input video in the present invention is a video file that has been encoded and compressed once by a mobile phone or a wired video terminal, or the original video file before encoding. Among them, the encoded file has already lost a lot of information about the image. Typically, compression causes block boundaries in the image, blurring of the edges of the image, color corruption, brightness Falling phenomenon; and the like. The preprocessor 120 plays a role of correcting the image lost due to the compression as close to the original as possible before performing ROI filtering, which is a main step for improving image quality.

In this embodiment, the deblocking filtering, edge improvement filtering, and contrast improvement filtering are sequentially performed as the preprocessing process, and the deblocking filter 121 and the edge improvement are configured to implement the respective processes. An Edge Enhancement Filter 122 and a Contrast Enhancement Filter 133 are provided. Here, not all three processes are necessarily required, and only a process required by a designer may be selected. In addition, the pre-processing process is not necessarily limited to the three types of filtering processes, and some of the processes may be replaced with or added to various conventional image quality improvement processes for the entire screen of the video. It is obvious to those of ordinary skill in the art.

The deblocking filter 121 plays a role of correcting that a boundary is generated in units of blocks on the video screen.

Video encoding (compression) algorithms such as H.263 or MPEG, which are used in mobile communication networks, divide one screen (or picture) into a number of blocks to improve speed. For H.263 QCIF, 176x144 pixels It forms a screen and blocks it in 16x16 units. As a result of the block-based compression process, when the video is played through the decoding process, it can be seen that the boundary of the image is formed in the block unit. The deblocking filter 131 applies a low pass filter between blocks to remove such boundaries. However, low pass filtering should be performed selectively because low pass filtering of the actual edges in the block may cause the image to be blurred overall. Therefore, we find the edge of each block and apply the low pass filter only to the part where no edge is applied.

The deblocking process in the present invention is performed in two parts.

The first part is the step of determining run which is an area in which a deblocking filter operates in a block, and an algorithm thereof is shown in the flowchart of FIG. Here, right_run [i] means the right run value in the i-th block, and right [j] means the j-th pixel value to the right in the block. The reason j should not exceed 8 is because the block that is the final image processing unit is 8x8. The alpha value is the threshold value of the value that is considered to be an edge in the image. The right_run, left_run, top_run, and bottom_run are obtained for every block boundary through the above-described detection operation on the block boundary, and then the deblocking execution region is determined.

The second part of the deblocking is the step of performing lowpass filtering on the region determined by the above algorithm. To this end, a sine waveform filter of the following equation is used.

In Equation 1, Mean means average of two pixel values between block boundaries and Amplitude means difference between boundary values of Mean and deblocking region, that is, the size of a sinusoidal filter to be applied. The variable t represents the position of the pixel about the boundary between blocks. The above filter is applied to each block in the horizontal and vertical directions to eliminate blocking. For reference, FIG. 2 (b) shows a trend of pixel values at a block boundary in a deblocking filter image, where a white point is a pixel value of a decoded input image and a black point is after a deblocking filter. Pointer to a pixel value. In a black and white image, pixel value 0 means black and pixel value 255 means white. In FIG. 2B, the input image (transition of white points) has a boundary based on the boundary between blocks, but the output image (transition of black points) that has passed through the deblocking filter is smoothly bordered.

Next, the edge improvement filter 122 serves to correct the boundary of each image constituting the video screen to be clear.

H.263 uses a lossy compression method, which is a technique of converting pixel information of an image into a frequency to reduce the capacity by removing high frequencies that do not have a significant effect on the human eye. In this case, the capacity is reduced, but the boundary of the image is blurred. The edge improvement filter 122 restores the damaged portion closer to the original for the same reason. In this way, if the edge portion becomes clearer, it is more clearly visible to the human eye, and thus the image quality can be improved subjectively.

In this embodiment, an Unsharp masking technique is used for edge improvement. This technique applies the following formula to increase the sharpness of the boundary.

In Equation 2, v (m, n) is a signal after the edge is improved, u (m, n) is an input signal, g (m, n) is a high frequency component obtained from coordinates (m, n) of the image, and λ. Is a constant that determines how much to strengthen the boundary. The unsharp masking technique is an algorithm that reinforces the boundary of an image by adding high frequency components to an input signal as shown in Equation 2. Where g (m, n) is obtained as follows.

In general, in the signal processing, low frequency components are obtained by averaging signals and high frequency components are obtained by removing low frequency components from current pixel values. In Equation 3, g (m, n) is a value obtained by subtracting low frequency components from an input signal at a position of (m, n), which is equivalent to obtaining high frequency components.

3 (a) to 3 (c) illustrate a process of extracting a high frequency component g (m, n) for a one-dimensional signal which is a relatively simple signal through the algorithm described above, and FIG. 3 (d) shows FIG. By adding the high frequency components obtained in a) to 3 (c) to the input signal, the result of the edge value at the boundary is finally highlighted.

The contrast enhancement filter 123, which is the last component of the preprocessor 120, serves to correct the brightness of the video screen.

In the present invention, since the decoded video of the object to be improved in image quality is generally low in contrast, an automatic histogram warping technique is applied as an algorithm for increasing it. Automatic histogram warping ensures that when a histogram of an image is concentrated in a particular part, it is evenly distributed over the entire range. Since there may be several places where the histogram is concentrated, this case is automatically detected and distributed evenly. The histogram of the input image illustrated in FIG. 4A shows a case where pixels are concentrated in a specific portion. If this is processed using the automatic histogram warping technique, as can be seen in the histogram of FIG. 4 (b), it can be seen that the pixels are distributed in a wider area.

Next, the ROI (Region Of Interest) filter 130 performs a secondary image quality improvement process for a specific region of interest of the video, and specifically encodes the received video according to the H.263 standard. In this video, filtering is performed by encoding quantization coefficients of the ROI and the ROI differently from each other.

Generally, when people watch a video, they tend to focus on a certain part rather than focusing on the whole screen. For example, when sending a video letter, the sender will look at the camera and speak, while the recipient will focus on the sender's face. At this time, even if the resolution of the background is lower than the resolution of the image as a whole, if the resolution of the face is high, subjectively, the image quality is improved.

Accordingly, the ROI filter 130 uses the principle of allocating more bits to the ROI than the uninterested region. Specifically, the ROI filter 130 is implemented by assigning different quantization coefficients to the ROI region and the non-ROI region. To this end, the ROI filter 130 is composed of a ROI determination algorithm for dividing the ROI region and a bit rate control algorithm for allocating respective quantization coefficients to the ROI region and the non-ROI region. For reference, if the quantization coefficient is large, the compression rate is high, but the image quality is low. If the quantization coefficient is small, the compression rate is low, but the image quality is improved.

Now, a method of correcting an image quality of a video input from a mobile communication network using the image quality improving apparatus having the above configuration will be described. For reference, FIG. 5 sequentially illustrates each step of the video quality improving method according to the present invention.

When a video is input from the mobile communication network to the image quality improving apparatus, the video is input to the decoder only when the video is encoded in a video standard such as H.263, MPEG, or the like (S501).

The decoder decodes the corresponding video according to the corresponding video standard and transmits the decoded video to the preprocessor after the preprocessing process and the ROI correction process are performed smoothly.

The preprocessing unit primarily performs an image quality improvement process on the entire area of the video screen, specifically, a deblocking filtering process (S505) for correcting a boundary occurring in units of blocks on the video screen, and each image constituting the video screen. An edge enhancement filtering process (S507) for correcting the boundary and a contrast enhancement filtering process (S509) for correcting the brightness of the moving picture are increased. At this time, it is preferable to proceed with the process only when there is a need for correction after analyzing the input video at each process step to determine whether it is necessary to go through the corresponding process, and go through only a specific process desired by the designer. It is also possible to do so.

The video, which has undergone the preprocessing process, performs a second quality improvement process on a particular ROI of the corresponding video screen through the ROI filter to subjectively feel a high quality screen (S511). In this case, the secondary image quality improvement process encodes the video according to the H.263 standard, but encodes the video by differently encoding quantization coefficients for the ROI and the ROI in the video.

On the other hand, if the video inputted to the video quality improving apparatus is an original video not encoded in a video standard such as H.263, MPEG, etc., step S503 should be omitted, and steps S505 or S505 to S507 are optional depending on the characteristics of the video. May be omitted.

Hereinafter, an embodiment in which the video quality improving method according to the present invention is applied to a mobile communication service will be described. For reference, FIG. 6 illustrates a case of applying to a video ring back tone (RBT) service of a mobile communication network, and FIG. 7 illustrates a case of applying to a missed video messaging service (VMS) of a mobile communication network. It is shown.

As the video call using mobile communication, the most popular services are video ring back tone (RBT) and video messaging service (VMS). Among these, video RBT compresses a user's desired video to a predetermined server. Is a service that allows a video RBT to run on the sender's terminal until the receiver receives a call when the sender calls the recipient, and VMS sends a message as a video that the sender wants to leave if the receiver does not answer the call. Say service.

First, an application example of the video ringback tone service will be described. When the sending mobile communication terminal 10 makes a call to the other mobile communication terminal 15 through the mobile communication network, the receiving station (not shown in the drawing) is received by the receiving side. The ringback tone signal is transmitted to the transmitting side 10 until the call connection is accepted at step 15. At this time, the RBT server 20 of the content provider provides the video content to the switching center to replace the simple ringback tone signal. The ring back tone is transmitted to the transmitting side (10).

Here, since the RBT server 20 is equipped with a video quality improving apparatus according to the present invention, it is not simply providing video content but is a low quality video, but the necessary portion of the screen is provided after intensive image quality correction is performed in a mobile communication network. It will increase the value of the video.

Next, an application example of the missed video message service will be described. If the sender 10 makes a call to the receiver 15 through the mobile communication network, but the other party is busy or does not answer the call for more than a predetermined time, or video If the call is impossible because the call is transferred to the mailbox, the sender records the message to be communicated to the other party through his terminal 10 and transmits the message to the VMS server 30 of the switching center for storage.

At this time, the VMS server 30 is also equipped with a video quality improvement apparatus according to the present invention, when the other party later tries to access the VMS server 30 and checks his message, the stored video is low quality, but the sender of the screen Necessary parts such as face are provided to the other party after intensive image quality correction so that the sender can be identified more clearly.

In the foregoing detailed description of the present invention, specific embodiments have been described, but various modifications are possible without departing from the scope of the present invention. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be defined not only by the scope of the following claims, but also by those equivalent to the scope of the claims.

According to the present invention having such a configuration, the image quality in a mobile communication network in which transmission is performed within a limited bandwidth overcomes the conventional problem of low capacity and low quality, and intensively improves image quality only for a specific part of the screen according to the purpose of the video. By doing so, a satisfactory level of image quality can be obtained at low cost. In addition, overall image quality improvement can be expected by performing a certain level of image quality correction on the entire video screen through a predetermined preprocessing process before the specific partial correction is performed, and the video inputted while being encoded according to the standard of the mobile communication network It is possible to improve the correction efficiency by distinguishing between the input video and the original, but performing a temporary decoding process only for the former.

In addition, by applying the present invention to services that use ringback tone service, missed video message service, and other video as next generation mobile communication service, high-quality service can be achieved through mobile communication network of limited bandwidth. do.

Claims (10)

An apparatus for improving video quality in a mobile communication network, A pre-processing unit performing a quality improvement process on the entire screen area of the input video firstly; And A region of interest (ROI) filter that performs an image quality improvement process on a specific region of interest of the pre-processed video; Device for improving video quality comprising a. The method of claim 1, And a decoder which decodes the input video in a predetermined video standard according to the video standard and transmits the decoder to the preprocessor. The method of claim 1 or 2, wherein the ROI filter An apparatus for improving video quality by encoding an input video according to the H.263 standard, but performing encoding by encoding quantization coefficients of a region of interest and an uninterested region differently in the corresponding video. The method of claim 3, wherein the pretreatment unit Deblocking filter for correcting the boundary that occurs in the unit of block on the video screen, Edge improvement filter for correcting the boundary of each image constituting the video screen to be clear, and Contrast enhancement filter to correct the brightness of the video screen, Motion picture quality improving device, characterized in that it comprises any one or more of. The method of claim 4, wherein When the original video is not encoded is input, any one of the edge enhancement filter, the contrast enhancement filter, or the ROI filter of the pre-processing unit is input to any one of the video quality improvement device. In a method for improving video quality in a mobile communication network, A preprocessing step of performing an image quality improvement process on the entire area of the input video screen; And ROI (Region Of Interest) filtering step that performs the quality improvement process on the specific area of interest of the video Video quality improvement method comprising a. 7. The method according to claim 6, wherein prior to said pretreatment step And decoding the input video according to a video standard if the input video is encoded according to a predetermined video standard. 8. The method of claim 6 or 7, wherein the ROI filtering step A method of improving video quality by encoding an input video according to the H.263 standard, but performing encoding by encoding quantization coefficients of a region of interest and an uninterested region differently in the corresponding video. The method of claim 8, wherein the pretreatment step Deblocking filtering step of correcting the boundary that occurs in the unit of block on the video screen, Edge improvement filtering step of correcting the boundary of each image constituting the video screen to be clear, and Contrast enhancement filtering to increase the brightness of the video screen, The video quality improvement method characterized in that it comprises any one or more of. The method of claim 9, If the original video is not encoded, the video quality improvement method, characterized in that performed from any one selected from the edge improvement step, the contrast improvement step, or the ROI filtering step of the preprocessing step.

Images