KR0174443B1 - Method and apparatus for encoding still image using adaptive vector quantization - Google Patents

Abstract

The present invention relates to a still image encoding technique using adaptive vector quantization, which enables to extract and compress only a certain portion of a video signal of one screen of a still image in response to a request from a remote user terminal. To this end, the present invention determines the quantization step size determined in consideration of the complexity of the input still image signal received from the remote user terminal via the ATM network in consideration of the complexity of the input still image signal, and the determined quantization step size Based on the quantization and variable length coding of the DCT transform coefficients corresponding to the still image, it is transmitted to the remote user terminal through the ATM network, and the image information transmission request signal, location information and size from the remote user terminal through the ATM network. Of all the still picture signals stored in the frame memory when received Determining and extracting a predetermined image portion, and determining a new quantization step size in consideration of the image complexity, image size, and pre-allocated bandwidth of the specific image portion extracted, and corresponding to the specific image portion based on this new quantization step size The DCT transform coefficients are quantized and variable length coded and then transmitted to a remote user terminal through an ATM network. Accordingly, the present invention can compress and transmit images of specific parts desired by a user while minimizing image quality deterioration due to quantization in consideration of allocated bandwidth as well as encoding transmission of all still images.

Description

Method and Apparatus for Still Image Coding Using Adaptive Vector Quantization

1 is a block diagram of a still picture coding apparatus using adaptive vector quantization according to a preferred embodiment of the present invention.

2 is a flowchart illustrating a process of compressing and encoding a still image according to the present invention, and compressing and encoding only a predetermined portion of image information in response to a user's request.

3 is a diagram illustrating an example of subdividing still image data of one frame into subblocks of a predetermined size according to the present invention.

* Explanation of symbols for main parts of the drawings

10 controller 20 frame memory

30: DCT block 40: quantization block

50: VLC block

The present invention relates to a technique for compressing and encoding an image signal, and more particularly, to a still image encoding method and apparatus for effectively compressing and encoding a still image signal using an adaptive vector quantization technique.

As is well known in the art, the transmission of discrete video signals can maintain better image quality than analog signals. When a video signal composed of a series of image frames is represented in digital form, a significant amount of data must be transmitted. However, since the usable frequency range of the conventional transmission channel is limited, in order to transmit a large amount of digital data, it is necessary to compress the data to be transmitted and reduce the amount of transmission.

On the other hand, the encoding of video signals is currently represented by encoding of moving images and encoding of still images. Here, for example, MPEG (Moving Picture Experts Group) is a mechanism that leads the decision of encoding standard of moving images. An example of a mechanism for determining the encoding standard of an image is JPEG (Joint Photographic Experts Group). Here, the present invention relates to a still picture coding algorithm driven by JPEG.

In this case, among the various compression techniques for compressing and encoding a video signal, a hybrid coding scheme combining a stochastic coding technique and a temporal and spatial compression technique is known to be the most efficient. The hybrid coding technique compresses and encodes a video signal using MC-DCT (Motion Compansation-Discrete Cosine Transform) and then transmits a video signal encoded through a transmission channel to a receiver.

On the other hand, in the case of a still image according to the present invention, when the transmission side transmits a still image signal, a discrete cosine transform (hereinafter abbreviated as DCT) and quantization is used to reduce the amount of data transmitted. After compression coding using the spatial correlation of the video signal, and then transmitting the compression-coded still image signal to the receiving side (user terminal) through the transmission channel.

Application of the still image coding algorithm as described above includes color facsimile, still picture telephony / conference, photo printing for printing and newspaper, transmission of medical field and design, and the like.

On the other hand, for a still image of a screen received through a transmission channel and displayed on a monitor, a user wants to enlarge only a specific part of a screen according to the purpose of use, especially an X-ray picture or a blueprint in a medical field. In the case of the conventional still image coding system, this may not be considered at all, and thus, there is a problem in that a desire according to a user's needs cannot be satisfied.

Accordingly, the present invention has been devised in view of the above-described needs, and adaptive vector quantization capable of extracting and compressing only the video signal of a specific portion of a video signal of one screen that forms a still picture in response to a request from a receiver. It is an object of the present invention to provide a still image encoding method and apparatus using the same.

In accordance with an aspect of the present invention, there is provided an apparatus for compressing and encoding a still image signal of a frame unit to be transmitted / received through an ATM network through a predetermined signal processing process. On the basis of the video information transmission request signal, the location information, and the size information received from the terminal, taking into account the write enable signal, the read address signal, and the complexity of the video required for compression encoding the corresponding still picture or the specified video part System control means for generating a quantization parameter for determining a quantization step size of said still image; A frame memory for storing a still image signal of one frame provided from an input side in accordance with a write enable signal from the system control means; A DCT block for dividing, by the system control means, pixel data of all still images or specified image portions drawn out from the frame memory into predetermined block units based on a read address signal and converting the pixel data into a DCT conversion coefficient; A quantization block for quantizing the DCT transform coefficients provided from the DCT block to a finite number of values based on a quantization step size determined according to a quantization parameter considering the complexity of the corresponding image provided from the system control means; Adaptive vector quantization consisting of a VLC block provided to the transmitter by variable length coding by allocating a short length code or a long length code according to the frequency of occurrence of each code using a code table for the quantized DCT transform coefficients. Provided is a still image encoding apparatus using.

According to another aspect of the present invention, there is provided a method of compression encoding a still image signal of a unit of a frame stored in a frame memory to be transmitted and received through an ATM network through DCT, quantization, and variable length encoding. The quantization step size is determined in consideration of the complexity of the still image signal input according to the video information transmission request signal received from the remote user terminal through the ATM network, and the stop of one frame drawn out from the frame memory. Converting pixel data of a predetermined block unit with respect to a video signal into DCT conversion coefficients of a predetermined block unit; Quantizing and variable length encoding the DCT transform coefficients based on the quantization step size, and then transmitting the DCT transform coefficients to the remote user terminal through the ATM network; Determining and fetching a specific video part of all still picture signals stored in the frame memory based on the video information transmission request signal, location information, and size information received from a remote user terminal through the ATM network; The new quantization step size is determined in consideration of the image complexity, image size, and pre-allocated bandwidth of the extracted specific image portion, and the pixel data of the predetermined block unit is converted into DCT conversion coefficients of the predetermined block unit. Converting process; And quantizing and variable length encoding the DCT transform coefficients corresponding to the specific video portion based on the new quantization step size, and transmitting the same to the remote user terminal through the ATM network. A video encoding method is provided.

The above and other objects and various advantages of the present invention will become more apparent from the preferred embodiments of the present invention described below with reference to the accompanying drawings by those skilled in the art.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram of a still picture coding apparatus using adaptive vector quantization according to the present invention. As shown in the figure, the encoding apparatus of the present invention includes a controller 10, a frame memory 20, a DCT block 30, a quantization block 40, and a variable length coding (VLC) block 50. .

In FIG. 1, the controller 10 is a control signal input from a receiving terminal of a remote place through an Asynchronous Transfer Mode (ATM) network (not shown), for example, a transmission request signal (REQ), a screen of one screen. After inputting position information (PI: Position Information) for expressing a specific portion and size information (SI: Size Information) of the specified screen portion, the frame memory (described later through line L12) corresponding to the input control signal ( 20, a read address signal for fetching image data is generated, and a quantization parameter QP for determining the quantization step size is provided to the quantization block 40 described later through the line L14. At this time, the controller 10 determines the quantization step size for quantizing the frame data in consideration of the complexity of the current input image provided through the line L11.

In addition, the frame memory 20 stores a still image signal of one frame provided from the input side based on the write enable signal provided from the controller 10 through the line L16. The signal is provided to the DCT block 30 of the next stage.

Meanwhile, in the DCT block 30 and the quantization block 40, still image data provided from the frame memory 20 is encoded by a series of quantized DCT transform coefficients. Then, the quantized transform coefficients are output to the VLC block 50 of the next stage.

More specifically, the DCT block 30 divides the input image into blocks of a predetermined size, for example, 8 × 8 pixel blocks, and then uses a cosine function to divide the image signal (pixel data) in the spatial domain. Pixel data of 8x8 unit is converted into DCT conversion coefficient of 8x8 unit. In addition, the quantization block 40 quantizes the DCT conversion coefficient from the DCT block 30 to a finite number of values through nonlinear operations. In such quantization, the quantization step size in the quantization block 40 is determined by the quantization parameter QP from the controller 10 according to the complexity of the input still image.

Therefore, the VLC block 50 variably encodes the DCT transform coefficient quantized through the quantization block 40 into run and coefficient through, for example, zigzag scanning. In more detail, the VLC block 50 is variable according to the frequency of occurrence of each code by using one code table, that is, a code having a high frequency of occurrence of a code is a short code, and a code having a low frequency of code is a long length. It is encoded by the code of and then provided to the output buffer (not shown) for transmission to the receiving side. The reason for allocating different lengths to all codes through the VLC block 50 is to increase the coding efficiency by substantially reducing the average value of the code lengths.

Next, the operation process of the still picture coding apparatus according to the present invention, which is composed of the above-described members, will be described in detail with reference to FIGS. 2 and 3 showing the flow thereof.

First, based on the user's video information request signal REQ, the controller 10 provides a write enable signal to the frame memory 20 via the line L16, thereby providing one frame for the desired still image in the frame memory 20. FIG. Data is stored, and thus, one frame signal stored in the frame memory 20 is read based on the read address signal from the controller 10 provided through the line L12, so that the DCT block 30, the quantization block 40, and It is assumed that the state is provided to the corresponding user terminal through the ATM network since it is compressed and coded through the VLC block 50 and then sent to a transmitter not shown.

That is, when the video information request signal REQ from the user terminal received through the ATM network (not shown) is input to the controller 10 (step 100), recording from the controller 10 provided through the line L16 is performed. The corresponding still image of one frame provided from the input side based on the enable signal is stored in the frame memory 20 (step 110). In this case, the controller 10 determines the quantization parameter QP for quantization of the frame data based on the complexity of the still image currently input through the line L11 and provides it to the quantization block 40 (step 120).

Accordingly, the still image signal converted into a DCT conversion coefficient of, for example, 8 × 8 block units through the DCT block 30 is quantized to a step size determined according to the quantization parameter QP from the controller 10. In addition, the variable length is encoded (step 130), and then sent to a transmitter not shown so as to be transmitted to a remote user terminal through an ATM network (step 140).

Therefore, as described above, in the situation where the desired still image desired by the user is displayed on the monitor of the user's terminal, the user needs to enlarge the specific part of the screen of one frame as necessary and request it through the ATM network. When a signal is sent, that is, when a specific video transmission request signal REQ, position information PI, and size information SI are input to the controller 10 (step 150), the controller 10 transmits each of these signal information ( On the line L16, a read address signal for extracting from the frame memory 20 only a predetermined portion specified by the user based on REQ, PI, and SI is generated.

At this time, when a part of the image is specified by the user on one screen according to the present invention, as shown in FIG. 3, for example, as shown in FIG. (Step 160). That is, as an example, when the user specifies the portion indicated by # 42 on the screen shown in FIG. 3, it is withdrawn from the frame memory 20 based on the read address signal provided from the controller 10 via the line L16. The image data provided to the DCT block 30 will be the production data of the portion indicated by # 42.

At this time, the controller 10 determines a new quantization parameter QP in consideration of the complexity of the image of the portion indicated by the specified # 42 input through the line L18. Accordingly, the quantization block 40 quantizes the DCT transform coefficients for the image of the portion indicated by the specified # 42 as described above with the quantization step size according to this newly determined quantization parameter QP, and then variable length coded (step 130, it is sent to the transmitter not shown so that it can be transmitted to the remote user terminal via the ATM network (step 140). Therefore, through this process, the remote user can continuously view only the images of the desired specific parts of the entire screen. In addition, after the controller 10 encodes and transmits the entire image or a specified portion of the image, the controller 10 maintains a standby state for request signal REQ (step 170).

On the other hand, when the controller 10 determines the quantization parameter (QP) for the specified image portion, the amount of encoded data for the specified portion of the image is at least four times smaller than the amount of image data when the entire image is encoded. Therefore, the quantization parameter (QP) is determined to fully utilize the bandwidth allocated to the entire image and to minimize image quality deterioration due to quantization. That is, assuming that the first quantization parameter (QP) of the entire image is 20, as in the above case, when only the # 42 part of the image specified by the user is encoded and transmitted, the quantization parameter (QP) is approximately 2.5. Will be. That is, the quantization step size is reduced by about 1/4 compared to when quantizing the entire image, which makes it possible to minimize the deterioration of image quality due to quantization by fully utilizing the allocated bandwidth when transmitting only a certain portion of the image.

In this case, considering that the entire image of one screen does not have an even distribution of complexity, the quantization step size is not necessarily reduced to 1 / N, and more or more depending on the image complexity of the part specified by the user. It may be reduced to less than that.

Therefore, when a user receives a specific part of the image and restores it to the original signal before the encoding is performed, the receiving side (remote terminal of the remote place) can display the specific part of the image on the entire monitor while minimizing image quality deterioration. In addition, specific image information may be extended through methods such as interpolation, which are well known in the art.

As described above, according to the present invention, in consideration of the allocated bandwidth as well as the entire image for the still image transmitted and received through the ATM network, while compressing the image for the specific parts desired by the user while minimizing image quality degradation due to quantization Can be encoded and transmitted.

Claims (5)

An apparatus for compressing and encoding a still image signal of a frame unit to be transmitted and received through an ATM network through a predetermined signal processing process, comprising: image information transmission request signal, location information, and the like received from a remote user terminal through the ATM network; A quantization parameter for determining the quantization step size of the still image in consideration of the write enable signal, the read address signal, and the complexity of the corresponding image required for compression encoding the corresponding still image or the specified image portion thereof based on the size information. Generating system control means (10); A frame memory 20 for storing a still image signal of one frame provided from an input side according to the write enable signal from the system control means; A DCT block (30) for dividing the pixel data of the whole still image or the specified image portion drawn out from the frame memory by a predetermined block unit by the system control means into a DCT conversion coefficient; A quantization block 40 for quantizing the DCT transform coefficients provided from the DCT block to a finite number of values based on the quantization step size determined according to the quantization parameter considering the complexity of the corresponding image provided from the system control means. ); And a VLC block 50 which is variable length coded and provided to the transmitter by allocating a short length code or a long length code according to a frequency of occurrence of each code using a code table for the quantized DCT transform coefficients. Image Coding Apparatus Using Regressive Vector Quantization. The apparatus of claim 1, wherein the size information is an integer value determined by continuously dividing the entire still image by a predetermined number of predetermined times. The method according to claim 1, wherein the system control means generates a corresponding new quantization parameter in consideration of the complexity and the allocated bandwidth of the specified image when only the specified image portion is extracted from the frame memory. Still image encoding apparatus using adaptive vector quantization. A method of compressing and encoding a still image signal of a unit of a frame stored in a frame memory to be transmitted and received through an ATM network through DCT, quantization, and variable length encoding, the image information received from a remote user terminal through the ATM network. A quantization step size is determined in consideration of the complexity of an input still picture signal according to a transmission request signal, and pixel data of a predetermined block unit for the still picture signal of one frame drawn out from the frame memory is determined in a predetermined block unit. Converting into DCT transform coefficients of the; Quantizing and variable length encoding the DCT transform coefficients based on the quantization step size, and then transmitting the DCT transform coefficients to the remote user terminal through the ATM network; Determining and fetching a specific video part of all still picture signals stored in the frame memory based on the video information transmission request signal, location information, and size information received from a remote user terminal through the ATM network; The new quantization step size is determined in consideration of the image complexity, image size, and pre-allocated bandwidth of the extracted specific image portion, and the pixel data of the predetermined block unit is converted into DCT conversion coefficients of the predetermined block unit. Converting process; And quantizing and variable length encoding the DCT transform coefficients corresponding to the specific video portion based on the new quantization step size, and transmitting the same to the remote user terminal through the ATM network. Image coding method. 5. The method of claim 4, wherein the size information is an integer value determined by continuously dividing the entire still image by a predetermined number of predetermined times.