JPH0851628A - Moving picture processor, its control method and moving picture transmission system - Google Patents

Abstract

PURPOSE:To provide a moving picture processor, its control method and a moving picture transmission system capable of eliminating the need of a movement compensation processing, reducing an arithmetic operation amount and a memory amount, simplifying a device and further, reducing the transmission amount of data and transmitting moving picture data even in the line of a low bit rate. CONSTITUTION:A picture group (GOP) composed of plural frames is divided into an I-picture and a C-picture. On a transmission side, scale information is prepared from the I-picture, the C-picture is quantized based on the scale information in a contour processing part 50, contour information for which picture element positions are classified for respective quantization values is prepared and encoding is performed in a run length encoding part 36. On a reception side, the encoded C-picture is decoded in a run length decoding part 46, a processing for storing the quantization value in the picture element position inside the frame is performed in an inverse contour processing part 60, the quantization value is inversely quantized based on the scale information and pictures are restored.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a moving picture processing apparatus for compressing and expanding moving picture data, a control method therefor, and a moving picture for transmitting and expanding compressed moving picture data using the moving picture processing apparatus. The present invention relates to an image transmission system, and in particular, it can reduce the amount of calculation by omitting the motion compensation process, simplify the device by reducing the amount of memory, and reduce the amount of data transmission to provide a low bit rate line. Also relates to a moving image processing apparatus capable of transmitting moving image data, a control method thereof, and a moving image transmission system.

[0002]

2. Description of the Related Art As a conventional MPEG1 system moving image processing apparatus for performing encoding for image compression and decoding for image expansion, there is one as shown in FIG. FIG.
FIG. 6 is a block diagram of a configuration of a conventional moving image processing apparatus.

As shown in FIG. 8, a conventional moving image processing apparatus is composed of an encoder section 10 for encoding image data and a decoder section 20 for decoding. Further, the encoder unit 10 performs inter-frame prediction from the pre-processing unit 11 that takes in the original image data from the input video signal, and inter-frame prediction from previous and next frames, and obtains a motion vector for each motion block in the frame. The compensation unit 12 includes a DCT unit 13 that performs a DCT (Discrete Cosine Transform) operation, a quantization unit 14 that performs quantization, and a variable length coding unit 15 that performs variable length coding.

The decoder unit 20 also includes a variable length decoding unit 25 for performing variable length decoding of the transmitted encoded data,
Inverse quantizer 24, which performs inverse quantization, and inverse DCT (IDC
T) An inverse DCT unit 23 that performs a calculation, a motion compensation unit 22 that performs motion compensation by moving a motion block based on a motion vector for an image that has already been decoded and stored in a memory, and an image from decoded data. And a post-processing section 21 that reproduces and outputs.

Next, each component will be specifically described. The pre-processing unit 11 of the encoder unit 10 is a unit that reads image data from a video signal of a VTR component. One frame of the input video signal is an odd number
It consists of an even number of two field signals, and the field signal period is 60 Hz. The pre-processing unit 11 of the conventional moving image processing apparatus of the MPEG1 system takes out one of these field signals and reads the image data.

The inter-frame prediction / motion compensation unit 12
This is a part that performs inter-frame prediction based on either the back frame or the previous or next frame, and further performs motion compensation processing for obtaining a motion vector for each motion block in the frame. The motion compensation process will be described later.

Further, in the moving image processing apparatus of FIG. 8, the DCT unit 13, the quantization unit 14, the variable length coding unit 15 of the encoder unit 10, the variable length decoding unit 25 and the dequantization unit of the decoder unit 20. The inverse DCT unit 24 and the inverse DCT unit 23 use the same principle as the JPEG encoding method. That is, an image divided into 8 × 8 pixel blocks is transmitted from the inter-frame prediction / motion compensation unit 12 of the encoder unit 10 on the transmission side, the image is subjected to DCT conversion in the DCT unit 13, and quantized in the quantization unit 14. Quantization is performed based on the table, and the variable length coding unit 15 performs 0 by zigzag scanning.
Of the quantized value and the level of the non-zero quantized value are encoded based on a two-dimensional VLC (Variable Length Coder) table, and the encoded data and the motion vector are transmitted to the transmission path.

In the decoder 20 on the receiving side, the variable-length decoding unit 25 performs variable-length decoding on the received encoded data, the inverse quantization unit 24 performs inverse quantization, and the inverse DCT unit 23 performs inverse quantization. The DCT calculation is performed, the motion compensation unit 22 reproduces an image from the motion vector and the decoded data, and the reproduced image is output from the post-processing unit 21.

Next, the relationship of prediction between frames will be described with reference to FIG. FIG. 9 is a schematic explanatory diagram showing a prediction relationship between frames. In MPEG1, some image frames are skipped and encoded, and images sandwiched between them are predicted from the front and the back and encoded. Therefore, there are three types of coded images, I-picture, P-picture, and B-picture, depending on the degree of prediction.

Here, the types of images will be briefly described. An I-picture is an image in which only that frame is compressed, and is an intra-frame encoded image that has no correlation with the preceding and following frames. The P-picture is an image created by performing prediction including motion compensation prediction based on the I-picture, and is a predictive-coded image related to the I-picture. Also, B-pictures are I-pictures and P-pictures.
Frames sandwiched between pictures, I-picture and P-
It is an image created by predicting from a picture, and is a bidirectional predictive encoded image using both past and future images.

As shown in FIG. 9, eight frames here form one group (GOP: Group Of Pictures), and the first frame (frame 1) is an I-picture and the fifth frame (frame 5) is It is a P-picture. The input image is frame 1, frame 2, frame 3
The input order is frame 1
(I-picture) and frame 5 (P-picture) are encoded, and then B-pictures of frame 2, frame 3 and frame 4 are predicted from frame 1 and frame 5 and encoded. Then, after the next I-picture is encoded, the frame 5, the frame 6, the frame 7, and the frame 8 sandwiched between the next I-pictures are predicted and encoded.

Next, the motion compensation processing in the motion compensation / interframe prediction unit 12 of the conventional encoder unit 10 and the motion compensation unit 22 of the decoder unit 20 will be described with reference to FIG. FIG. 10 is a schematic explanatory view showing a conventional motion compensation process. First, an input image of a frame to be an I-picture is divided into, for example, a block of 8 × 8 pixels, and this block is set to a 2 × 2 16 × 16 pixel macroblock (motion block). This frame serves as a unit of motion compensation and serves as the basic data for the motion vector calculation to be processed thereafter.

Next, in the next frame which becomes the P-picture, first, it is divided into motion blocks in the same manner as the previous I-picture, and each motion block is compared with the motion block of the I-picture. As a result of the comparison, if there are blocks having different pixel values, it means that some motion exists between the two frames.

When there are different motion blocks between the two frames, the search is performed to find where the motion block in the previous frame (I-picture) exists in the current frame. Then, when a motion block identical or similar to the motion block in the I-picture is detected in the current frame, the position of the motion block in the current frame and I
-The difference from the position of the motion block in the picture is obtained as a motion vector and sent as a P-picture.

Then, the decoder unit 20 decodes and reproduces the received encoded data of the I-picture, and temporarily stores the I-picture in the memory. Then, the motion compensation unit 22 moves each motion block of the I-picture in the memory to reproduce the P-picture based on the motion vector transmitted as the P-picture, and the post-processing unit 2
Send to 1. In this way, the motion compensation process is performed.

As for B-pictures, the preceding and following I
-A motion vector is obtained by referring to a picture and a P-picture, but there are three types of predictions: forward prediction predicted from past images, backward prediction predicted from future images, and interpolation prediction predicted from past and future images. There is a prediction mode, and the optimum one is selected and predicted.

[0017]

However, in the above-described conventional moving image processing apparatus, control method therefor, and moving image transmission system, when the transmission data of P-picture and B-picture is created in the encoder section Since a complicated motion compensation process for detecting a vector is performed, there is a problem that the amount of calculation and the amount of memory increase, the device becomes complicated, and the device cost increases.

The present invention has been made in view of the above-mentioned circumstances. The motion compensating process is unnecessary, the amount of calculation and the amount of memory are reduced to simplify the device, and the amount of data transmission is reduced to reduce the number of bits. An object of the present invention is to provide a moving image processing apparatus capable of transmitting moving image data even on a high-rate line, a control method thereof, and a moving image transmission system.

[0019]

The invention according to claim 1 for solving the problems of the above-mentioned conventional example is a moving image processing apparatus for encoding a moving image for each image data of a frame and outputting it to a transmission path. A scale information creation unit that creates scale information of pixel values in a frame of the I-picture, with a plurality of consecutive frames as one image group, and a first frame of the image group as an I-picture, A first encoding unit that compresses and encodes image data of a frame of an I-picture to generate frame information of the I-picture, and a frame other than the I-picture in the image group as a C-picture. Pixel values in the frame of the C-picture are quantized based on the scale information, pixel positions are classified for each quantized quantized value, and contour line data is obtained. And Contour processing unit for forming, is characterized by a second encoding unit for creating frame information of the contour data by encoding C- picture.

According to a second aspect of the present invention for solving the above-mentioned problems of the conventional example, in the control method of the moving image processing apparatus according to the first aspect, a pixel value is generated by a scale information creating unit for an I-picture of an image group. Scale information of
The first encoding unit compresses and encodes the image data of the I-picture frame to create I-picture frame information, and the contour line processing unit scales the pixel values of the C-picture of the image group. Quantizing based on information, classifying pixel positions for each of the quantized values to create contour line data, and coding the contour line data by a second coding unit to create frame information of a C-picture. It has a feature.

According to a third aspect of the present invention for solving the above-mentioned problems of the conventional example, a moving image processing apparatus according to the first aspect is provided.
Regarding the frame information of the I-picture and the frame information of the C-picture output from the moving image processing apparatus described in the above,
A first decoding unit that decodes and expands the frame information of the I-picture, and an I that is expanded by the first decoding unit.
A scale information creating unit that creates scale information of pixel values in the frame from image data of the frame of the picture; and a second decoding unit that decodes frame information of the C-picture.
And a process of storing the quantized value corresponding to each pixel position in the frame of the C-picture from the quantized value and the pixel position of the contour line data decoded by the second decoding unit. , And an inverse contour line processing unit that dequantizes the quantized value of each pixel based on the scale information to restore the image data of the frame of the C-picture.

According to a fourth aspect of the present invention for solving the problems of the conventional example, in the control method of the moving image processing apparatus according to the third aspect, I outputted from the moving image processing apparatus according to the first aspect. A frame information of a picture is decoded and expanded by a first decoding unit, and scale information creation unit creates scale information of pixel values in the frame from image data of the expanded I-picture frame; The frame information of a C-picture output from the moving image processing apparatus according to claim 1 is decoded by a second decoding unit, and the inverse contour processing unit decodes the decoded contour data by the quantization value and the pixel position. The process of storing the quantized value corresponding to each pixel position in the frame of the C-picture is performed, the quantized value of each pixel is dequantized based on the scale information, and the frame of the C-picture is processed. It is characterized in that to restore the image data.

According to a fifth aspect of the present invention for solving the problems of the conventional example, a moving image processing apparatus according to the first aspect and a moving image processing apparatus according to the third aspect are transmitted in a moving image transmission system. It is characterized by being connected by road.

[0024]

According to the first and second aspects of the invention, the pixels of the C-picture, which is another frame in the image group, based on the scale information created by the scale information creating unit from the I-picture in the image group. Quantize the values in the contour line processing unit,
Further, the contour image data in which the pixel positions are classified for each quantized value is created, and the contour image data is encoded by the second encoding unit and output to the transmission path. The moving image compression process can be performed more easily than the moving image data compression process.

According to the third and fourth aspects of the invention, the scale information creating unit creates the scale information of the pixel values in the frame from the image data of the I-picture frame decoded by the first decoding unit. , The inverse contour processing unit stores the quantized value corresponding to the pixel position in the frame from the contour data decoded by the second decoding unit, and further dequantizes the quantized value based on the scale information. Since the moving image processing apparatus and the control method therefor restore the image data of the frame of the C-picture, it is possible to easily perform the moving image expansion processing as compared with the moving image data expansion processing by MPEG.

According to the invention described in claim 5, the moving image processing apparatus according to claim 1 is provided on the transmitting side and the moving image processing apparatus according to claim 3 is provided on the receiving side, and both apparatuses are connected by a transmission line. Since the moving image transmission system is configured as described above, the amount of data transmission can be reduced.

[0027]

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described with reference to the drawings. FIG. 1 is an explanatory diagram showing the structure of an image frame in a moving image processing apparatus according to an embodiment of the present invention. In the moving image processing apparatus of the present invention, a moving image is composed of a plurality of GOPs (Group Of Pictures) as shown in FIG. 1, and one GOP is composed of a plurality of frames. The first frame in the GOP is called an I-picture, and the second and subsequent frames are called C-pictures. Then, in the control method of the moving image processing apparatus according to the present embodiment, all pixel value data for I-pictures are compressed / decompressed by the JPEG method similar to the conventional one, while for C-pictures, I-pictures in the same GOP -Based on the range of the pixel value of the picture, quantized data is created from all pixel value data forming the C-picture, and further contour line processing is performed to create contour line data from the quantized data to encode the contour line data. By doing so, the image of the C-picture is compressed, inversely the quantized data is created from the decoded contour line data, and the inverse contour line processing is performed to restore the pixel value data from the quantized data. The image is expanded.

Here, the data obtained by compressing the image of the I-picture has a sufficient amount of information to express one image by itself, and has a large data amount as compared with that of the C-picture. On the other hand, the image of the C-picture is quantized based on the range of pixel values of the I-picture in the same GOP, contour-line-processed, and compressed, so that the information amount is compressed. -Pixel values on the C-picture that are not included in the pixel value range of the picture will be ignored.

As a result, a moving image close to the original image can be restored by reducing the number of C-pictures in one GOP, but the compression rate of image data does not improve and the transmission efficiency does not improve. On the other hand, if the number of C-pictures in one GOP is increased, the reproducibility of the image with respect to the original image is reduced, but the compression rate of the image data is improved and the transmission efficiency is improved. Therefore, the number of C-pictures for one I-picture, that is, the number of frames forming a GOP is taken into consideration in view of the restored image quality and the transmission efficiency required for the moving image transmission system using the moving image processing apparatus of this embodiment. To decide. In the example of FIG. 1, the GOP is composed of one I-picture and three C-pictures.

Next, the configuration of the moving picture transmission system according to the embodiment of the present invention will be described with reference to FIG. Figure 2
FIG. 1 is a configuration block diagram of a moving image transmission system according to an embodiment of the present invention. It should be noted that portions having the same configurations as those in FIG.

The moving picture transmission system of this embodiment is simply a moving picture compression apparatus for compressing and transmitting image data of each frame constituting a moving picture, and a moving picture for receiving and expanding compressed image data. It is composed of a decompression device and a transmission line. A moving image compression device and a moving image decompression device that are configured as one device is called a moving image processing device, and a moving image processing device is provided on the transmitting side and the receiving side and connected by a transmission line. This is a general moving image transmission system. In FIG. 2, for simplification of explanation, the moving image compression apparatus on the transmission side is shown in the upper part, and the moving image expansion apparatus on the reception side is shown in the lower part.

First, the basic structure of the moving picture compression apparatus is shown in FIG.
As shown in, the preprocessing unit 30 that performs preprocessing of encoding
An I-picture encoding unit 31 that compresses (encodes) an I-picture, a C-picture encoding unit 32 that encodes a C-picture, and a transmission unit 33.
And a memory 34.

Then, the I-picture encoding unit 31
Is composed of a DCT unit 13 which is the same as the conventional unit, a quantization unit 14, a variable length coding unit 15, and a scale information creation unit 35 which is a characteristic part of the present embodiment. The C-picture encoding unit 32 is composed of a contour line processing unit 50 and a run-length encoding unit 36, which are characteristic parts of this embodiment. In addition, in the memory 34,
Image data area 34a and scale information area 34b
And a contour line information area 34c are provided.

Similarly, the basic structure of the moving image decompressing device is, as shown in FIG. 2, a post-processing unit 40 for performing post-processing of decoded image data corresponding to each component of the moving image compressing device. , I-picture decoding unit 41 for decoding I-pictures, C-picture decoding unit 42 for decoding C-pictures, receiving unit 43, and memory 44.

Then, the I-picture decoding unit 41
The inverse DCT unit 23, which is the same as the conventional one, the inverse quantization unit 24, the variable length decoding unit 25, and the scale information creation unit 45, which is a characteristic part of the present embodiment, are configured. Further, the C-picture decoding unit 42 is composed of an inverse contour line processing unit 60 and a run length decoding unit 46, which are characteristic parts of this embodiment. In the memory 44,
Image data area 44a and scale information area 44b
And a contour line information area 44c.

Next, the function of each part will be specifically described.
The DCT unit 13, the quantization unit 14, and the variable length coding unit 15 in the I-picture encoding unit 31, the inverse DCT unit 23, the inverse quantization unit 24, and the variable length decoding unit 25 in the I-picture decoding unit 41. Performs the same JPEG encoding and decoding processing as that of the conventional art, the description thereof is omitted here, and only the characteristic part of the present embodiment will be described.

The image data area 34a in the memory 34 of the moving image compression apparatus is an area for storing one screen (frame) of digital image data obtained by A / D converting the input video signal. In addition, the scale information area 34b is
This is an area for storing scale information including a minimum value, a maximum value, and a scale width of pixel values forming image data of an I-picture. Here, the scale width is a value indicating the width of pixel values that can be taken by one quantized value when the pixel values of the minimum value to the maximum value are quantized into the quantized values of 0 to n. The quantized value is a value obtained by quantizing the pixel value of the C-picture based on the scale information of the I-picture. Further, the contour line information area 34c is an area for storing contour line information obtained by converting the contour line data of the C-picture into a transmission data format. Here, the contour line data is data in which the pixel position is classified for each quantized value.

On the other hand, the image data area 44a in the memory 44 of the moving picture decompression device stores the I-picture image data decoded by the JPEG method or the C-picture image data subjected to the inverse contour line processing. This is the area to store. In addition, the scale information area 44b includes the received I-
This is an area for storing scale information including a minimum value, a maximum value, and a scale width of pixel values forming image data of a picture. The contour line information area 44c is an area for storing contour line information of the received and run-length decoded C-picture.

Then, the preprocessing unit 30 of the moving picture compression apparatus is
The input video signal is A / D converted to create digital image data, the created digital image data is stored in frame units in the image data area 34a in the memory 34, and the stored frames are referred to as I-pictures. Do you C
-A frame determination process for determining whether to use a picture is performed, and if it is an I-picture, the I-picture encoding unit 3
1 is started, and if it is a C-picture, the C-picture encoding unit 32 is started.

The input video signal is composed of data obtained by interlacing and scanning an image of one screen called interlace. That is, the even-numbered signal (field 2) is input to the preprocessor 30 after the odd-numbered signal (field 1) of the screen. Therefore, in the preprocessing unit 30 of the present embodiment, the input video signal is converted into digital image data and the image data area 34a in the memory 34 is converted.
When storing in, the image data of the field 1 previously input is skipped and stored in the address position corresponding to the odd row, and the image data of the next input field 2 is stored in the address position between the data of the field 1 By doing so, the image data for one screen is stored in the image data area 34a as the image of the screen.

Further, in the frame discrimination processing of the input video signal, specifically, the number of input frames is counted, and 1 of GOP is calculated based on the preset number of GOP frames (F). For the second frame,
In other words, if the frame count is a multiple of F + 1, then I
-As a picture, in the case of the second and subsequent frames, that is, when the frame count number is other than a multiple of F + 1, C-
It is designed to be a picture.

If the number of C-pictures in one GOP is reduced, the quality of the restored image is improved, but the compression ratio is not improved and the transmission efficiency is not improved. On the other hand, the C-picture in one GOP is not improved. If the number is increased, the quality of the restored image is lowered, but the compression rate is improved and the transmission efficiency is improved. Therefore, the number of GOP frames is set in advance in consideration of the required restored image quality and the transmission efficiency.

Next, the scale information creating unit 35 is activated when the preprocessing unit 30 determines that the frame of the image data stored in the image data area 34a is an I-picture, and the scale information creating process is performed. Is to do. Here, the scale information creation processing is to read the image data stored in the image data area 34a in DCT transform blocks (usually 8 × 8 pixels)
The image data area 34a is output while being output to the CT unit 13.
Is a process of creating scale information of the image data stored in and storing it in the scale information area 34b.

Here, the scale information is the maximum value and the minimum value of the pixel values forming the image data of the I-picture, and the scale width. Then, the value of n of the quantized value (0 to n) is
It is a value set in advance depending on the quality of the image to be compressed / decompressed. For example, if each pixel value is quantized with an 8-bit quantized value (0 to 255) and the pixel value is 100 to 500, the scale width is about (500-100) / 256. Become.

Next, the contour line processing unit 50 includes the preprocessing unit 30.
This is activated when it is determined that the frame of the image data stored in the image data area 34a is the C-picture in. Then, in the contour line processing unit 50, based on the scale information of the I-picture stored in the scale information area 34b, the image data area 3
The contour data is created by subjecting the image data stored in 4a to contour contour processing and stored in the contour contour information area 34c. The details will be described later.

The run-length coding unit 36 performs the run-length coding process on the contour information of the C-picture stored in the contour-line information area 34c. The run-length coding method is the same. This is according to the general method of coding the length of a code (0 or 1) in succession.

Then, the transmitting unit 33 uses the I-picture frame information of the I-picture encoded by the I-picture encoding unit 31 or the frame information of the C-picture encoded by the C-picture encoding unit 32 as an I-picture or a C-picture. A transmission data is created by adding a header containing a picture flag indicating the frame type of a picture, and is output to the transmission path.

On the other hand, the receiving unit 43 of the moving image decompressing device performs a receiving process for receiving the data transmitted from the transmitting unit 33 of the moving image compressing device, and receives from the value of the picture flag included in the header of the received data. A frame determination process is performed to determine whether the frame information is an I-picture or a C-picture. If the frame information is an I-picture, the I-picture decoding unit 41
Is started, and if it is a C-picture, the C-picture decoding unit 42 is started.

In the I-picture decoding unit 41, the variable length decoding unit 25 decodes the received I-picture frame information, the inverse quantization unit 24 inversely quantizes the frame information, and the inverse DCT unit 23 further inverse DCT. The image data of one block is converted and output to the scale information creating unit 45.

The scale information creating section 45 includes an inverse DCT section 2
Image data area 44
In addition to storing in a, scale information is created from the received image data of the I-picture and stored in the scale information area 44b.

Note that the scale information referred to here is exactly the same as the scale information in the above-described moving picture compression apparatus.
-The maximum and minimum values of the pixel values forming the image data of the picture, and the scale width in the case of quantizing the minimum value to the maximum value of the pixel values into the quantized values of 0 to n, and the value of n is It is a value that is set in advance depending on the quality of the image to be compressed / decompressed and must match the setting value on the moving image compression apparatus side.

The run length decoding unit 46 is activated when it is determined that the frame information received by the receiving unit 43 is a C-picture, and the run length decoding process is performed on the received frame information. The contour line information is created and the contour line information is stored in the contour line information area 44c. The method of run length decoding processing is a general method.

Next, the inverse contour line processing unit 60 is activated when the run length decoding unit 46 completes the storage of the contour line information, and is stored in the scale information area 44b.
-Based on the scale information of the picture, the contour information stored in the contour information area 44c is subjected to inverse contour processing to create C-picture image data, and the created image data is stored in the image data area 44a. It is a thing. The details will be described later.

The post-processing unit 40 is started when the scale information creating unit 45 completes the storage of the scale information or when the inverse contour processing unit 60 completes the storage of the image data by performing the inverse contour processing. And post-processing is performed. The specific contents of the post-processing are as follows. First, the image data of the odd-numbered rows is taken out from the image data area 44a, D / A converted and outputted, and then the image data of the even-numbered rows is taken out D
A / A-converted and output to create and output an interlaced video signal.

Next, the internal structure of the contour line processing section 50 will be described in detail with reference to FIG. FIG. 3 is a block diagram showing a configuration inside and around the contour line processing unit 50 of the moving picture compression apparatus according to the present embodiment. As shown in FIG. 3, the contour line processing unit 50 of the moving picture compression apparatus according to the present embodiment includes a quantizing unit 51, a contour line data creating unit 52, a data converting unit 53, and a quantum as a data area used by each unit. It is composed of a digitized data area 55 and a contour line data table 56.

Explaining the configuration of each part in detail, the quantized data area 55 stores in the scale information area 34b each pixel value data constituting the image data of the C-picture stored in the image data area 34a. In the area for storing a value (quantized data) quantized on the basis of the scale information of the I-picture being displayed, the quantized value corresponding to the pixel position of one screen (one frame) is stored as in the image data area 34a. It is stored in a matrix.

The contour line data table 56 is an area of a table format for storing the contour line data of the quantized data stored in the quantized data area 55, and is in the form of a position number indicating the pixel position corresponding to the quantized value. The data is stored. Specifically, the value of the quantized value of the contour line data table 56 is set in advance to a value of 0 to n determined by how many bits the quantized value is represented, and the position number is the quantized value. For example, if the pixel position has a pixel position of (5, 5), then the pixel position is stored as indicated by the converted number (numerical value) of “0505”.

Then, the quantizing means 51 is first activated in the contour line processing section 50, and each pixel value of the image data stored in the image data area 34a is converted into the scale information stored in the scale information area 34b. Quantized based on the above, and stored in the quantized data area 55. Quantized processing is performed.
Is a means of activating.

The specific contents of the quantization process are as follows: where the minimum value of the scale information stored in the scale information area 34b is min, the maximum value is max, and the scale width is s.
The pixel value (x) at each pixel position is sequentially read from left to right and from top to bottom of the image data area 34a, and the quantized value (y) corresponding to the read pixel value (x) is calculated by the following formula. Then, it is converted into an integer value, and the obtained quantized value is stored in the position of the quantized data table 34b corresponding to the read pixel position. y = (x-min) / s (Equation 1) However, the method of obtaining an integer value from the calculation result of (Equation 1) may be rounding down or rounding down.

Further, in the image data stored in the image data area 34a, the pixel value smaller than the value of min is
Pixel values that are unconditionally quantized to a quantized value “0” and larger than the value of max are unconditionally quantized to a quantized value “n”. That is, the pixel value portion originally not included in the I-picture is treated as a value of min or max when the quantization processing of the C-picture is performed, and cannot be represented accurately in the restored image of the C-picture. Become.
Therefore, in a moving image that changes drastically, the number of frames that make up one GOP is reduced, and C for an I-picture is reduced.
-It is necessary to reduce the number of pictures.

Next, the contour line data creating means 52 is activated after the quantizing processing is completed in the quantizing means 51, and it quantizes from one screen of quantized data stored in the quantized data area 55. This is a means for performing contour line data creation processing for creating contour line data in which pixel positions having the quantized value are classified for each value, and activating the data conversion means 53 after the contour line data creation processing is completed.

The specific contents of the contour line data creation processing are as follows:
The quantized value of each pixel position in the quantized data area 55 is sequentially read, the quantized value 56a of the contour data table 56 that matches the read quantized value is searched, and the position number 56b corresponding to the quantized value 56a that matches. The position number of the quantized value read in is stored.

Next, the data conversion means 53 is started after the contour line data generation processing in the contour line data generation means 52, and performs data conversion processing for converting the contour line data into contour line information which is a transmission format, and after the data conversion processing is completed. This is means for activating the run length encoding processing 36 shown in FIG. The contour line information format is
As shown in FIG. 3, data indicating a quantized value delimiter,
The value of the quantized value and the position number having the quantized value are set as one set and stored in order from the quantized values "0" to "n". In principle, the position number is stored one or more after the data indicating the quantized value delimiter and the value of the quantized value, but if there is no corresponding position number, the quantized value delimiter is set. Only the indicated data and the quantized value are stored.

The specific contents of the data conversion process are as follows. The quantized value 56a and the position number 56b corresponding to the quantized value 56a are read from the contour line data table 56, and the data indicating the delimiter of the quantized value and the value of the quantized value are read. , And the position number are sequentially written in the contour line information area 34c.

In the case of creating contour line information by the above method, when the image to be compressed is an image with little variation in pixel values, there is no corresponding position number in the contour line information, and there is no quantized value delimiter. Since a number of pairs of only the data indicating the value and the value of the quantized value are included, the transmission data is wasteful. Therefore, in the data conversion processing of the data conversion means 53, it is judged whether or not each quantized value has a position number, and the contour line information is created only for the quantized value having the position number. The amount of data to be transmitted can be reduced although it will drop slightly.

Next, the internal structure of the inverse contour processing unit 60 will be described in detail with reference to FIG. FIG. 4 is a block diagram showing the configuration inside and around the inverse contour line processing unit 60 of the moving picture decompression device of this embodiment. The inverse contour processing unit 60 of the moving picture decompression device of the present embodiment corresponds to each means and each data area inside the contour processing unit 50 of the moving picture compression device shown in FIG. 3, as shown in FIG. Inverse quantizer 61
, Quantized data creation means 62, and data inverse conversion means 6
3, a quantized data area 65 as a data area used by each means, and a contour line data table 66.

The structure of each unit will be described in detail. The contour line data table 66 is an area of a table format for storing each quantized value and the position number having the quantized value. The quantized data area 65 is an area for storing quantized values corresponding to pixel positions of one screen (one frame) in a matrix.

Next, the data inverse conversion means 63 is activated when the decoding processing of the received data is completed by the run length decoding section 46, and the received and decoded contour line information stored in the contour line information area 44c is received. Is a means for performing a data inverse conversion process of inversely converting the data into contour data and creating the contour data table 66.

Here, the format of the contour line information is exactly the same as the contour line information of the moving picture compression apparatus. And
The specific contents of the data inverse conversion processing are as follows: the quantized value and its position number are read from the contour line information area 44c, the quantized value is stored in the quantized value 66a of the contour line data table 66, and the position number is written into the position number 66b. To be stored in.

Next, the quantized data creating means 62 is activated after the data inverse conversion processing is completed by the data inverse conversion means 63, and the quantized data area 63 is stored in the quantized data area 63 from the contour data stored in the contour data table 66. To 1
Quantized data creation processing for creating quantized data for the screen is performed, and after the quantized data creation processing is completed, the inverse quantization means 61.
Is a means of activating. The specific content of the quantized data creation processing is as follows:
a and the position number 66b corresponding thereto are read, and the quantized data area 6 corresponding to the read position number 66b is read.
The value of the quantized value 66a is stored in the address position of 5.

The dequantization means 61 is activated after the quantized data creation processing in the quantized data creation means 62 is completed, and the quantized data stored in the quantized data area 63 is stored in the scale information area 44b. Image data is created by performing inverse quantization on the basis of the scale information stored, and the image data is stored in the image data area 44a. After the inverse quantization process, the post-processing unit 40 shown in FIG. It is a means to start.

The specific contents of the inverse quantization process are as follows: where the minimum value of the scale information stored in the scale information area 34b is min, the maximum value is max, and the scale width is s, the left side of the quantized data area 65. From the right to the right and from top to bottom, the quantized value (y) at each pixel position is sequentially read, and the pixel value (x) corresponding to the read quantized value (y) is calculated by the following formula to obtain an integer value. The converted and obtained pixel value is stored in the position of the pixel value data area 44a corresponding to the read pixel position. x = y × s + min (Equation 2) However, the method of obtaining the integer value from the calculation result of (Equation 2) may be rounding down or rounding down.

Next, the control method of the moving image processing apparatus of this embodiment will be described. First, a control method of the moving picture compression apparatus according to the present embodiment will be described with reference to FIGS. FIG. 5 is a flowchart showing an outline of the control flow of the moving picture compression apparatus of this embodiment. Note that FIG. 5 shows the order in which the processing for one frame is started. As shown in FIG. 5, the control flow of the moving picture compression apparatus of the present embodiment is as follows. First, as the operation of the preprocessing unit 30, a video signal is received (100), and the image data area 34a of the memory 34 is stored. It is stored (102) and the received frame is I-
A frame discrimination process of a picture or a C-picture is performed (104).

If the received frame is an I-picture, scale information creation processing is performed as the operation of the scale information creation unit 35 (110), and the operations of the DCT unit 13, the quantization unit 14, and the variable length coding unit 15 are performed. As a result, the JPEG encoding process is performed (112), the picture flag indicating that the I-picture is set in the header as the operation of the transmission unit 33 (120), the transmission process is performed (150), and the moving image compression process is ended. .

On the other hand, when the frame received in the process 104 is a C-picture, contour line processing is performed as the operation of the contour line processing unit 50 (130), and then run length encoding is performed as the operation of the run length encoding unit 36. Processing is performed (132), and the operation of the transmission unit 33 is to set a picture flag indicating a C-picture in the header (1
40), the transmission process is performed (150), and the moving image compression process ends.

Next, a method of controlling the moving picture decompression device of this embodiment will be described with reference to FIGS. FIG. 6 is a flowchart showing an outline of the control flow of the moving picture decompression device of this embodiment. Note that FIG. 6 shows the order in which the processing for one frame is started. As shown in FIG. 6, the outline of the control flow of the moving image decompressing apparatus of this embodiment is as follows.
0), the received frame is subjected to frame discrimination processing as to whether it is an I-picture or a C-picture (202).

If the received frame is an I-picture, the variable length decoding unit 25 and the inverse quantization unit 24
And the inverse DCT unit 23 performs JPEG decoding processing (210), and the scale information creation unit 45 performs scale information creation processing (212), and the post-processing unit 4
Post-processing is performed as an operation of 0 (240), and the moving image decompression processing is ended.

On the other hand, when the frame received in the process 202 is a C-picture, the run-length decoding unit 46 performs the run-length decoding process (22).
0), then inverse contour line processing is performed as the operation of the inverse contour line processing unit 60 (230), post-processing is performed as the operation of the post-processing unit 40 (240), and the moving image expansion processing is ended.

According to the moving picture processing apparatus and its control method of the present embodiment, in the moving picture compression apparatus, the I-picture which is the first frame of the GOP has JPEG for all pixel values.
The encoding process is applied to compress and transmit the second
The C-picture, which is the subsequent frame, is quantized based on the scale information that is the pixel value range of the I-picture, and is further compressed by creating contour line data in which pixel positions are classified for each quantized value. On the other hand, when the received frame is an I-picture, the moving image decompressing device decompresses it by performing JPEG decoding processing, and then C
-In the case of a picture, since each quantized value at each pixel position is obtained from the contour line data, and further dequantized based on the scale information that is the pixel value range of the received I-picture, it is expanded. The compression process is simple and can be realized with a small amount of memory. As a result, the CP of the moving image processing apparatus can be realized.
This has the effect of reducing the U load and improving the processing capacity.

Next, the outline of another embodiment (second embodiment) of the present invention will be described with reference to FIGS. 2, 3, 4, and 7. FIG. 7 is an explanatory diagram for explaining the outline of the moving image processing apparatus according to the second embodiment. The moving image processing apparatus of the second embodiment transmits only the contour line data of the quantized value at a specific interval from the contour line data of the C-picture that has been contoured as in the first embodiment. That is, FIG.
The original image of the C-picture shown in FIG.
Only contour data at specific intervals as shown in (b) are transmitted, and the contour lines are interpolated on the receiving side.

The structure of the moving picture compression apparatus of the second embodiment is based on the structure of the moving picture compression apparatus of the first embodiment shown in FIGS. 2 to 4, but shown in FIG. In the data converting means 53 in the contour line processing unit 50 of the moving image compression apparatus of the first embodiment, only the quantized values extracted at a specific interval from all the quantized values set in the contour line data table 56, Data indicating the delimiter of the quantized value,
The value of the quantized value and the position number are shown in the contour line information area 34c.
It will be written in sequence.

On the other hand, in the moving picture expansion apparatus of the second embodiment, the quantized data in the quantized data area 65 created by the quantized data creating means 62 of the first embodiment shown in FIG. It means that no data is stored at the pixel position having the quantized value that is not extracted by the moving image compression apparatus. Therefore, in the moving picture decompression device of the second embodiment,
A process for smoothing the quantized data is added between the quantized data creating means 62 and the inverse quantizing means 61.

Here, the smoothing process of the quantized data is
The quantized data area 65 is searched, and for pixel positions in which no quantized value is stored, interpolation calculation is performed from the value in which the quantized value in the horizontal direction or the vertical direction is stored to calculate the quantized value. . The interpolation calculation method includes linear interpolation and curve interpolation as examples.

According to the moving picture processing apparatus and the control method therefor of the second embodiment, based on the embodiment of the younger brother 1, the C-picture is extracted at a specific interval from the contour line data for all quantized values. And transmitting only the contour line data relating to the quantized value, while the moving picture decompression device stores the quantized value at each pixel position from the received contour line data of the C-picture and stores the quantized value Since the position is subjected to the smoothing process, the compression rate of the C-picture is improved, and as a result, the transmission efficiency of the moving image transmission system can be improved.

Next, the outline of another embodiment (third embodiment) of the present invention will be described. In the moving image processing apparatus of the third embodiment, C is used on the moving image compression apparatus side as in the first embodiment.
-When quantizing image data for a picture, only pixel values extracted at specific intervals are extracted and quantized, and thereafter, contour line data is created and compressed data is transmitted as in the first embodiment. . Then, on the side of the moving image decompressing device, the pixel value is interpolated with respect to the pixel position where the pixel value is not stored in the image data area 44c created in the same manner as in the first embodiment.

The structure of the moving picture compression apparatus of the third embodiment is based on the structure of the moving picture compression apparatus of the first embodiment shown in FIGS. 2 to 4, but shown in FIG. In the quantizing means 51 in the contour line processing unit 50 of the moving picture compression apparatus of the first embodiment, only the pixel values extracted at a specific interval from all the pixel values stored in the image data area 34a are quantized. Then, the contour data is created in the quantized data area 55, and the contour data creating means 52 creates the contour data only for the position where the quantized value is stored.

On the other hand, in the moving picture expansion apparatus of the third embodiment, no pixel value is stored in the image data area 44a created by the inverse quantizing means 61 of the first embodiment shown in FIG. The interpolation processing for the pixel position is added.

Here, the interpolation processing is to search the image data area 44a, and for pixel positions where pixel values are not stored, perform interpolation calculation from pixel values stored in the horizontal direction or the vertical direction to obtain pixel values. It is to be calculated. The interpolation calculation method includes linear interpolation and curve interpolation as examples.

According to the moving picture processing apparatus and its control method of the third embodiment, the C-picture is converted into the C-picture in the contour line processing unit 50 of the moving picture compression apparatus based on the embodiment of the younger brother 1. Quantization is performed only on pixel values extracted at a specific interval from all the constituent pixel values to form contour data, and the contour data is transmitted. On the other hand, in the moving picture decompression device, the contour data of the received C-picture is quantized. Image data is created, image data is created by inverse quantization,
Since the interpolation processing is performed on the pixel position where the pixel value is not stored, the efficiency of the compression processing of the C-picture is improved and the compression rate is improved, and as a result, the transmission efficiency of the moving image transmission system is improved. There is an effect that can be made.

[0090]

According to the first and second aspects of the present invention, a C-picture which becomes another frame in the image group based on the scale information created from the I-picture in the image group by the scale information creating unit. The pixel value of is quantized by the contour line processing unit, the contour line data in which the pixel position is classified for each quantized value is further created, and the contour line data is encoded by the second encoding unit and output to the transmission path. Since the apparatus and the control method thereof are used, it is possible to easily perform the compression processing of the moving image as compared with the compression processing of the moving image data by MPEG, and thus it is possible to reduce the calculation amount and the memory amount.
Further, there is an effect that the device configuration can be simplified and the data transmission amount can be further reduced.

According to the third and fourth aspects of the invention, the scale information creating unit creates scale information of pixel values in the frame from the image data of the I-picture frame decoded by the first decoding unit. , The inverse contour processing unit stores the quantized value corresponding to the pixel position in the frame from the contour data decoded by the second decoding unit, and further dequantizes the quantized value based on the scale information. Since the moving image processing apparatus and the control method thereof for restoring the image data of the C-picture frame are used, the moving image expansion processing can be performed more easily than the moving image data expansion processing by MPEG. The calculation amount and the memory amount can be reduced, and the device configuration can be further simplified.

According to the invention described in claim 5, the moving image processing apparatus according to claim 1 is provided on the transmitting side, and the moving image processing apparatus according to claim 3 is provided on the receiving side, and both apparatuses are connected by a transmission line. Since this is a moving image transmission system, it is possible to reduce the amount of data transmitted, and thus it is possible to facilitate the transmission of moving image data even on a transmission path of a low bit rate line.

[Brief description of drawings]

FIG. 1 is an explanatory diagram showing a configuration of an image frame in a moving image processing apparatus according to an embodiment of the present invention.

FIG. 2 is a configuration block diagram of a moving image transmission system according to an embodiment of the present invention.

FIG. 3 is a contour line processing unit 50 of the moving picture compression apparatus according to the present embodiment.
It is a block diagram which shows the structure inside and its periphery.

FIG. 4 is an inverse contour line processing unit 6 of the moving picture decompression device according to the present embodiment.
FIG. 3 is a block diagram showing a configuration inside 0 and its periphery.

FIG. 5 is a flowchart showing an outline of a control flow of the moving picture compression apparatus according to the present embodiment.

FIG. 6 is a flowchart showing an outline of a control flow of the moving picture decompression device of this embodiment.

FIG. 7 is an explanatory diagram illustrating an outline of a moving image processing apparatus according to a second embodiment.

FIG. 8 is a configuration block diagram of a conventional moving image processing apparatus.

FIG. 9 is a schematic explanatory diagram showing a prediction relationship between frames.

FIG. 10 is a schematic explanatory view showing a conventional motion compensation process.

[Explanation of symbols]

10 ... Encoder unit, 11 ... Pre-processing unit, 12 ... Inter-frame prediction / motion compensation unit, 13 ... DCT unit, 14 ...
Quantization unit, 15 ... Variable length coding unit, 20 ... Decoder unit, 21 ... Post-processing unit, 22 ... Motion compensation unit, 23 ...
Inverse DCT unit, 24 ... Inverse quantization unit, 25 ... Variable length decoding unit, 30 ... Preprocessing unit, 31 ... I-picture encoding unit, 32 ... C-picture encoding unit, 33 ... Transmission unit, 34 ... Memory, 34a ... Image data area,
34b ... Scale information area, 34c ... Contour line information area, 35 ... Scale information creating section, 36 ... Run length encoding section, 40 ... Post-processing section, 41 ... I-picture decoding section, 42 ... C-picture decoding section, 43
... reception section, 44 ... memory, 44a ... image data area, 44b ... scale information area, 44c ... contour line information area, 45 ... scale information creation section, 46 ... run length decoding section, 50 ... contour line processing section, 51 ... quantum Transforming means, 52 ... Contour line data creating means, 53 ... Data converting means, 55 ... Quantized data area, 56 ...
Contour data table, 60 ... Inverse contour processing unit, 6
1 ... Inverse quantization means, 62 ... Quantized data creation means, 6
3 ... Data inversion means, 65 ... Quantized data area,
66 ... Contour line data table

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI Technical indication location H03M 7/40 9382-5K H04N 5/92 H04N 5/92 H

Claims (5)

[Claims] 1. A moving image processing apparatus for encoding a moving image for each image data of a frame and outputting the encoded data to a transmission path, wherein a plurality of consecutive frames are set as one image group, and the first frame of the image group is A scale information creating unit that creates scale information of pixel values in the frame of the I-picture as an I-picture, and image data of the frame of the I-picture is compressed and encoded to create frame information of the I-picture. And a first encoding unit for performing C-frame conversion for other frames except I-pictures in the image group.
A contour line processing unit that quantizes pixel values in the frame of the C-picture as a picture based on the scale information, classifies pixel positions for each quantized quantized value, and creates contour line data; and the contour line data. To encode C
And a second encoding unit that creates frame information of a picture. 2. A scale information creating unit creates scale information of pixel values for I-pictures in an image group, and a first encoding unit compresses and encodes image data of a frame of the I-picture to obtain I-pictures. -Create frame information of a picture, quantize pixel values of the C-picture of the image group based on the scale information in a contour line processing unit, and classify pixel positions for each of the quantized values to create contour line data. The method of controlling a moving image processing apparatus according to claim 1, wherein the contour data is encoded by a second encoding unit to create frame information of a C-picture. 3. The first information for decoding and expanding the I-picture frame information for the I-picture frame information and the C-picture frame information output from the moving image processing apparatus according to claim 1. A decoding unit, a scale information creation unit that creates scale information of pixel values in the frame from the image data of the I-picture frame expanded by the first decoding unit, and frame information of the C-picture And a quantized value corresponding to each pixel position in the frame of the C-picture from the quantized value and pixel position of the contour line data decoded by the second decoding unit. Is stored, the quantized value of each pixel is dequantized based on the scale information, and the C-
A moving image processing apparatus, comprising: an inverse contour line processing unit that restores image data of a frame of a picture. 4. The frame information of an I-picture output from the moving image processing apparatus according to claim 1, is decoded and expanded by a first decoding unit, and the expanded I-picture is expanded by a scale information creating unit. Scale information of pixel values in the frame is created from the image data of the frame, and the frame information of the C-picture output from the moving image processing apparatus according to claim 1 is decoded by the second decoding unit, and inversed. The contour line processing unit performs a process of storing the quantized value corresponding to each pixel position in the frame of the C-picture from the quantized value of the decoded contour line data and the pixel position, and the quantized value of each pixel is stored. The method of controlling a moving image processing apparatus according to claim 3, wherein dequantization is performed based on the scale information to restore the image data of the frame of the C-picture. 5. A moving image transmission system comprising the moving image processing apparatus according to claim 1 and the moving image processing apparatus according to claim 3 connected by a transmission path.