JPH05235985A - Video signal packet transmitter and receiver - Google Patents

Abstract

(57) [Abstract] [Purpose] ATM for video signal packet transmission equipment
Minimize the image quality deterioration of the detailed video when the packet is discarded due to network congestion. A transmitting apparatus includes a high-order component energy calculating unit for calculating the energy of a DCT coefficient high-order component and comparing it with a threshold value, and a priority packet transfer for converting a DCT coefficient low-order component and a high-order component having large energy into priority packets. And a non-priority packet transfer unit that packetizes high-order components of small energy into non-priority packets. Then, if it is necessary to discard packets due to ATM network congestion, the non-priority packets are discarded. The receiving device includes a non-priority packet compensating unit, and when a packet is discarded, outputs a predetermined compensation signal instead of the signal of the discarded non-priority packet, and includes the output signal and the priority packet. Video is played with the signal that is present.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for packetizing and transmitting a digital video signal and an apparatus for receiving such a packet.

[0002]

2. Description of the Related Art An example of a conventional video signal packet transmitting / receiving apparatus will be described with reference to FIG. In this figure, (A) is
The configuration of the transmitting device includes a memory 300, a coding unit 301, a priority packet transfer unit 302, and a non-priority packet transfer unit 303. (B) is a configuration of the receiving device, in which 304 is a packet receiving unit, 305 is a non-priority packet interpolating unit, 306 is a decoding unit, and 307 is a memory.

The operation of the above video signal packet transmitting / receiving apparatus will be described. On the transmitting side, the video signal is
Reading is performed in units of 8 × 8 pixel blocks recorded in the memory 300 and having 8 lines in the horizontal direction and 8 lines in the vertical direction. The encoding unit 301 performs a discrete cosine transform (hereinafter referred to as “DCT”) on the pixel block, and 8 ×
Output the 8DCT coefficient block. At this time, the DCT coefficient block is classified into low-order components and high-order components shown in FIG.
The low-order component is converted into a priority packet by the priority packet transfer unit 302 and transferred to the ATM transmission line, and the high-order component is converted into a non-priority packet by the non-priority packet transfer unit 303.
It is transferred to the TM transmission line. Also, due to sudden changes in the scene, AT
When the amount of information to be transmitted on the M transmission path increases rapidly, and the amount of generated information exceeds the processing capacity of the transmission device, the transmission capacity of the line and the capacity of the exchange, congestion occurs and packet discarding becomes necessary. , Non-priority packets are discarded, and priority packets are transmitted with priority. On the receiving side, the packet receiving unit 304 classifies the received packets into priority packets and non-priority packets. The non-priority packet interpolating unit 305 detects whether or not the non-priority packet is discarded, and sets the information of the DCT coefficient higher-order component lost due to the packet discard to “0”.
Output with supplement. The decoding unit 306 determines that the priority packet D
An inverse discrete cosine transform (hereinafter referred to as “IDCT”) is applied to a DCT coefficient block composed of a CT coefficient low-order component and a DCT coefficient high-order component of a non-priority packet to decode a pixel block. The memory 307 records a pixel block and reproduces a video signal.

The low-order DCT coefficient component is a component that reproduces the rough characteristics of an image, and the image quality deterioration when lost due to packet discard is so large that it is transferred as a priority packet. On the other hand, the DCT coefficient higher-order component is
In general, the energy is smaller than that of the low-order component, and when the energy is lost, the influence on the image quality is not so large as that of the low-order component.
Therefore, in the conventional video signal packet transmission device, the DCT
Higher order components of the coefficients are transferred as non-priority packets.

However, the energy of the higher order components is
It varies depending on the type of image and may become large in the case of a fine image. In such an image, if a high-order component is transmitted as a non-priority packet, if the packet should be discarded, the image quality will be greatly deteriorated.

[0006]

As described above, in the conventional video signal packet transmission / reception, the DCT coefficient high-order component is always transferred as a non-priority packet regardless of the type of image. Therefore, when a non-priority packet is discarded in the transmission of a complicated and precise image, the adverse effect on the image quality due to this is very large. However, while viewers' demands for detailed, high-quality images are increasing, that is, the amount of information to be transmitted is increasing, the processing capability of transmitting and receiving devices as well as other information to be transmitted is increasing. As the transmission capacity of the ATM transmission network itself may be insufficient in terms of the number of lines, the allocation time, the capacity of the light emitting element and the exchange, etc., a lot of such congestion may occur. Conceivable.

In view of the above-mentioned problems, the present invention provides a D image with a complicated image.
When the energy of higher order CT coefficient component is large,
By transferring higher-order components as priority packets,
The present invention has been made for the purpose of providing a video signal packet transmitting / receiving apparatus capable of minimizing deterioration of image quality when a packet is discarded. Next, other transforms such as Hadamard transform and KL transform may be adopted in addition to the discrete cosine transform depending on the application of the current video signal, the convenience of the transmitting and receiving facilities, etc. Challenges are occurring.

[0008]

In order to achieve the above object, in the invention of claim 1, an image reading means for recording a video signal and reading out an image in units of a pixel block consisting of a predetermined number of pixels; In a video signal packet transmission device having a coding means for subjecting the outputted pixel block to discrete cosine transform and outputting a DCT coefficient block, the energy of the higher order component of the DCT coefficient block is calculated and compared with a predetermined threshold value. High-order component energy calculation means for performing the above, comparison information transfer means for transferring the comparison result, priority packet transfer for forming and transferring a priority packet from the DCT coefficient low-order component and the DCT coefficient high-order component having energy larger than the threshold value And a non-priority packet composed of high-order DCT coefficient components having energy smaller than the threshold And a video signal packet transmitting apparatus is characterized in that a non-priority packet transfer means that.

According to the second aspect of the present invention, an image reading means for recording a video signal and reading an image in units of a pixel block consisting of a predetermined number of pixels, and a DCT coefficient block for performing a discrete cosine transform on the read pixel block. In a video signal packet transmission device having an encoding means for outputting a packet, a packet receiving means for classifying a received packet into a priority packet and a non-priority packet, and information on a result of comparison between the energy of the DCT coefficient higher-order component and a predetermined threshold value. Comparison information receiving means for receiving, priority packet classifying means for decomposing the classified prioritized packets and classifying and outputting them into a DCT coefficient low-order component and a DCT coefficient high-order component, and discarding detection of non-priority packets and only when discarding Non-priority packet compensating means for complementing and outputting the information of the DCT coefficient high-order component with a predetermined signal;
Higher order component selecting means for selecting higher order DCT coefficient components from the priority packet classifying means or the non-priority packet compensating means according to the information of the comparison result, and an inverse discrete cosine transform for the lower order DCT coefficient components and higher order components. The video signal packet receiving device is characterized by comprising a decoding means for decoding the applied pixel block and a memory for recording the decoded pixel block and reproducing the video signal.

In the invention of claim 3, an orthogonal transformation such as Hadamard transformation or KL transformation is adopted in place of the discrete cosine transformation, and the DCT coefficient block is a matrix corresponding to the transformation adopted. The video signal packet transmitting device and the receiving device according to claim 1 or 2.

[0011]

With the above structure, in the invention of claim 1,
At the time of transmission, the DCT coefficient high-order component whose energy calculated by the high-order component energy calculation means is larger than the threshold value is transferred as a priority packet together with the DCT coefficient low-order component under the function of the priority packet transfer means. Then, when congestion occurs on the ATM transmission line and packet discarding is required, non-priority packets are discarded, and priority packet transmission is prioritized.

According to the second aspect of the present invention, when the cell discard of the non-priority packet due to congestion occurs, the non-priority packet compensating means that detects this cell outputs a predetermined compensation signal for the discarded portion. .. Then, based on the received information about the threshold value from the comparison information receiving means, the higher order component selecting means selects the higher order DCT coefficient higher component from the priority packet classifying means and the non-priority packet compensating means. The DCT coefficient higher-order components having smaller energy are selected, and these signals are input to the decoding means together with the DCT coefficient lower-order components output by the priority packet classification means. Then, the received image is reproduced on it.

In the third aspect of the present invention, the above-mentioned operation is performed by another transform instead of the Fourier transform.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A video signal packet transmitting device according to the invention of claim 1 and a video signal packet receiving device according to the invention of claim 2 will be described below based on embodiments. FIG. 1 is a block diagram of an embodiment of a video signal packet transmission device employing a transmission device and a reception device according to the present invention. (A) of the figure shows a transmitter according to the invention of claim 1,
Is a memory, 101 is a coding unit, 102 is a higher-order component energy calculation unit, 103 is a comparison information transfer unit, 104 is a priority packet transfer unit, and 105 is a non-priority packet transfer unit.
(B) is a receiving device according to the invention of claim 2,
Is a packet receiving unit, 107 is a priority packet classifying unit, 10
8 is a non-priority packet compensator, 109 is a comparison information receiver,
Reference numeral 110 is a high-order component selection unit, 111 is a decoding unit, and 112 is a memory.

Next, the operation according to the gist of the present invention will be described. Note that the transfer of the comparison result with the energy threshold, the classification and detection of the priority packet and the non-priority packet, the detection of the packet discard, etc. are performed by the information added separately to the video signal in the cell forming the packet. The operation, the cell data storage structure, the cell manufacturing method, the video pixel block division method, the DCT coefficient block generation method, etc. are described in, for example, the Institute of Electronics and Communication Engineers, Ohmsha "Electronic Communication Handbook", Kashimura et al. Ohm. Published by "Digital PBX Picture Book in the ISDN Era", published by Ohno et al., "OSI &I"
It has been published in a large number of printed materials such as "SDN Picture and Glossary of Terms", and the applicant of the present invention also separately discloses related arts, for example, Japanese Patent Application Nos. 1-157758, 1-157760 and 1-157760.
Although this is a well-known technique in terms of hardware and software as filed in Japanese Patent Application No. 157761, its description is omitted. Similarly, the circuit structure of the memory, the encoding unit, etc. is a well-known technique, and therefore its explanation is omitted.

First, the transmission operation according to the invention of claim 1 will be described. The memory 100 records a video signal and reads it out in units of an 8 × 8 pixel block composed of 8 lines in the horizontal direction and 8 lines in the vertical direction. The encoding unit 101
DCT is applied to this pixel block, and 2 × 8 × 8 DCT
Output the coefficient block of the dimensional frequency component. The DCT coefficient block is divided into low-order components and high-order components as shown in the coefficient matrix of FIG. Higher-order component energy calculation unit 102
Compares the energy Eh of the higher-order component with the set threshold Vth. At this time, the energy Eh of the higher-order component is represented by the following (Equation 1) when Cij is a coefficient of the coefficient matrix.

[0017]

[Equation 1]

The comparison information transfer unit 103 transfers this comparison result. On the other hand, the priority packet transfer unit 104 prioritizes and transfers the DCT coefficient low-order component and the high-order component with Eh ≧ Vth in the priority packet based on the comparison result. The non-priority packet transfer unit 105 converts the higher-order DCT coefficient components with Eh <Vth into non-priority packets and transfers them. At this time, if congestion occurs on the ATM transmission line and packet discarding becomes necessary, the non-priority packets are discarded, and the priority packets are preferentially transmitted.

Next, the receiving operation will be described. The packet receiving unit 106 classifies the received packets into priority packets and non-priority packets. The priority packet classification unit 107 classifies the priority packets into information of DCT coefficient low-order components and information of high-order components. The non-priority packet compensating unit 108 monitors whether or not the non-priority packet has been discarded, and when the non-priority packet is discarded, the information of the relevant unit is complemented with “0” and output. On the other hand, the comparison information receiving unit 109 receives the comparison information between the higher-order component energy Eh and the threshold value Vth, which is separately transmitted from the comparison information transfer unit 103. In accordance with this comparison information, the higher-order component selection unit 110 selects and inputs the DCT coefficient higher-order component information from the priority packet classification unit when Eh ≧ Vth and from the non-priority packet compensation unit when Eh <Vth. To do. The decoding unit 111 outputs the DCT output from the priority packet classification unit.
DCT output from coefficient low-order component and high-order component selector
IDCT is applied to the higher-order coefficient components to decode the pixel block. The memory 112 records this pixel block and then reproduces a video signal.

Next, the invention of claim 3 will be described. In the inventions of claims 1 and 2, the discrete cosine transform is adopted for the conversion of the video signal, but in the invention of this claim 3, other transforms such as the Hadamard transform and the K-L transform are adopted to transform the DCT coefficient. Instead of blocks, we use the corresponding matrix of the adopted transformation. The other aspects, the processing of the video signal, the basic aspects of the equation for calculating the energy Eh, etc. are the same as those of the embodiments of the invention of claims 1 and 2. For this reason,
Further, the advantages and disadvantages compared with Fourier transforms such as Hadamard transform and K-L transform are well-known techniques as described in, for example, "Electronic Communication Handbook", 1st edition, page 1793, published by Ohmsha Co., Ltd. Omit it.

Although the present invention has been described above based on the embodiments, it goes without saying that the present invention is not limited to the above embodiments. That is, specifically, the output of the non-priority packet compensating means is not the 0 signal but the image signal of the portion of the screen generated immediately before, so that the adverse effect of the non-priority packet discard on the image is further reduced.

The configuration of the pixel block is N lines in the horizontal direction and M lines in the vertical direction (where N ≠ M) according to the image forming apparatus on the receiver side. In calculating the higher-order component energy Eh, a coefficient related to image sharpening (secondary coefficient of change) is also taken into consideration. In order to effectively deal with cell discard caused by the ATM network side, the transmitter side and the receiver side have some communication lines for information exchange separately.

Calculation of the energy Eh having a higher order component is represented by the following equation. Eh = Σ | C _ij |, where 0 ≦ i ≦ 7, 0 ≦ j ≦ 7, i +
j ≧ 5. In calculating the energy Eh of the higher order components,
i + j is not ≧ 5, but i + j ≧ 6. Needless to say, it also includes the above.

[0024]

As described above, according to the present invention, when transmitting a complex image, the DCT coefficient higher-order component is transferred as a priority packet, and when transmitting a simple image, the higher-order component is transferred as a non-priority packet. By transferring, it is possible to minimize deterioration of image quality when discarding non-priority packets.

[Brief description of drawings]

FIG. 1 is a configuration diagram of an embodiment of a video signal packet transmission device according to the present invention, (A) showing a transmitting device and (B) showing a receiving device.

FIG. 2 is a diagram showing a concept of a method of calculating an energy of a DCT coefficient higher-order component in the embodiment.

3A and 3B are configuration diagrams of a video signal packet transmission device according to a related art; FIG. 3A shows a transmission device, and FIG. 3B shows a reception device.

FIG. 4 is a schematic diagram of a DCT coefficient block.

[Explanation of symbols]

 100 Memory 101 Encoding Unit 102 Higher-Order Component Energy Calculation Unit 103 Comparison Information Transfer Unit 104 Priority Packet Transfer Unit 105 Non-Priority Packet Transfer Unit 106 Packet Reception Unit 107 Priority Packet Classification Unit 108 Non-Priority Packet Compensation Unit 109 Comparison Information Reception Unit 110 Higher order component selection unit 111 Decoding unit 112 Memory

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁵ Identification code Office reference number FI technical display location H04N 7/133 Z 4228-5C

Claims (3)

[Claims] 1. An image reading means for recording a video signal and reading out an image in a pixel block composed of a predetermined number of pixels, and an encoding means for performing a discrete cosine transform on the read pixel block and outputting a DCT coefficient block. A high-order component energy calculating means for calculating the energy of the high-order component of the DCT coefficient block and performing a magnitude comparison with a predetermined threshold value, and a comparison information transferring means for transferring the comparison result. Priority packet transfer means for forming and transferring a priority packet from a DCT coefficient low-order component and a DCT coefficient high-order component having energy larger than the threshold, and a non-priority packet forming and transferring a DCT coefficient high-order component having energy smaller than the threshold Video signal packet transmission, characterized by comprising: Apparatus. 2. An image reading means for recording a video signal and reading an image in a pixel block composed of a predetermined number of pixels, and an encoding means for performing a discrete cosine transform on the read pixel block and outputting a DCT coefficient block. In the video signal packet transmission device having the above, the packet receiving means for classifying the received packet into the priority packet and the non-priority packet, and the comparison information receiving means for receiving the information of the comparison result between the energy of the DCT coefficient higher order component and the predetermined threshold And a priority packet classifying means for decomposing the classified priority packets and classifying and outputting them into a DCT coefficient low-order component and a DCT coefficient high-order component, and discarding non-priority packets to detect DCT coefficient high-order components only when discarding. Non-priority packet compensating means for complementing and outputting information with a predetermined signal, and the priority packet according to the information of the comparison result. Higher-order component selecting means for selecting higher-order DCT coefficient components from the classifying means or the non-priority packet compensating means, and decoding for performing inverse discrete cosine transform on the lower-order and higher-order DCT coefficient components to decode pixel blocks A video signal packet receiving apparatus comprising: means and a memory that reproduces a video signal after recording the decoded pixel block. 3. The DCT coefficient block is a matrix corresponding to the transform adopted by adopting an orthogonal transform such as Hadamard transform or KL transform instead of the discrete cosine transform. Item 2. The video signal packet transmitter and receiver according to item 2.