JP4295861B2 - Transcoder device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention is applied to a serial transmission path switching system that selectively switches video signals of various transmission rates with one switching device, and more particularly to a transcoder device that arbitrarily changes the transmission rate of an encoded video signal.
[0002]
[Prior art]
Recently, with the advancement of digital technology, not only standard standard television signals but also high-definition television signals have been digitized. Also, video compression technology has been actively developed by digitizing such video signals. As this video compression technique, for example, there is a coding compression system standardized by MPEG (Moving Picture Coding Experts Group).
[0003]
From the above, for example, in the studio equipment of a broadcasting station, video signals of various transmission rates are handled in the station. In this case, a serial transmission path switching device is used in which a plurality of serial transmission paths in which a plurality of transmission rates are mixed are concentrated in one place and selectively exchanged.
[0004]
By the way, in the serial transmission path switching device, when a plurality of serial transmission paths are exchanged and connected, it is necessary to have the same transmission rate. Therefore, a transcoder device that arbitrarily changes the transmission rate of a bit stream encoded by the MPEG standard is used.
[0005]
This type of transcoder device includes a video decoder and a video encoder, decodes a bit stream obtained by encoding a video signal by the video decoder, and re-encodes the decoded video signal at a transmission rate specified by the video encoder. And output to the serial transmission line switching device.
[0006]
However, the conventional transcoder apparatus as described above cannot determine the encoding structure of the bit stream at the time of input, and therefore, the mismatch between the bit stream at the time of input and the bit stream at the time of re-encoding. Will occur and the image quality will change stochastically, leading to degradation of the image quality. In addition, the bitstream may be pulled down at an arbitrary ratio in order to absorb the difference between the television system and the film system, for example, and the conventional transcoder device may determine the bitstream pulldown structure. Can not.
[0007]
Therefore, in the transcoder apparatus, auxiliary information indicating the bitstream encoding structure and pull-down structure is generated by the video decoder and multiplexed on the decoded video signal, and the auxiliary information is extracted from the video signal by the subsequent video encoder. Therefore, the mol (mol) system that inherits the coding structure and the pull-down structure is used. However, in this mol method, detection of the coding structure and pull-down structure, generation of auxiliary information indicating the coding structure and pull-down structure, and multiplexing processing are performed by one video decoder, and extraction and coding control of auxiliary information are performed. Since the processing is performed by one video encoder, a high processing capability is required for each of the video decoder and the video encoder, and therefore the inside of the video decoder and the video encoder must be modified.
[0008]
[Problems to be solved by the invention]
As described above, in the conventional transcoder apparatus, image quality degradation or the like occurs due to mismatch between the encoding structure and pull-down structure of the bit stream at the time of input and the encoding structure and pull-down structure of the bit stream at the time of re-encoding. Has the problem. In addition, in order to match the coding structure and the pull-down structure, detection of the coding structure and the pull-down structure, generation of auxiliary information indicating the coding structure and the pull-down structure, and a process of multiplexing the auxiliary information on the decoded video signal are performed. A single video decoder performs auxiliary information extraction and encoding control processing with a single video encoder, which requires high processing capabilities for each video decoder and video encoder, and requires modification of internal circuitry. Has the disadvantage of becoming.
[0009]
The object of the present invention is to be able to arbitrarily change the transmission rate of an encoded video signal using an existing decoder and encoder, and to flexibly handle encoded video signals having different structures. An object of the present invention is to provide a transcoder device that can be used.
[0010]
[Means for Solving the Problems]
The transcoder apparatus according to the present invention is an apparatus that handles an encoded video signal that has been pulled down to M (M is a natural number): N (N is a natural number) and encoded by inserting a dummy field. Decoding means for decoding the encoded video signal, encoding means for encoding the decoded signal decoded by the decoding means at an externally designated transmission rate, and the encoded video signal input to the decoding means Detection means for detecting a pull-down structure and encoding control means for controlling the encoding process of the encoding means based on the detection result of the detection means are provided. The encoding control means controls the encoding means so as to execute encoding processing in accordance with the pull-down structure and encoding structure of the encoded video signal input to the decoding means from the detection result of the detecting means. It is characterized by.
[0011]
According to this configuration, since the encoded video signal is decoded once and returned to the original, and encoded again at an arbitrary transmission rate, the transmission rate of the encoded signal can be easily changed. The pull-down structure of the encoded video signal at the time of input is detected in a system different from the decoding process and the encoding process of the encoded video signal, and the decoded signal is matched with the pull-down structure of the encoded video signal at the input based on this detection result. Thus, the encoding process is controlled so as to be re-encoded.
[0012]
For this reason, since the pull-down structure of the encoded video signal at the time of input is inherited at the time of re-encoding in a different system from the decoding process and the encoding process, the encoded video signal at the time of input and the encoded video that has been re-encoded The fluctuation and deterioration of image quality due to mismatch with the signal can be minimized, and it is possible to flexibly cope with encoded video signals having different structures, as well as a decoder and an encoder. In this case, since it is not necessary to have a high processing capability for inheriting the pull-down structure of the encoded video signal at the time of re-encoding, it is not necessary to change or modify the internal circuit.
[0013]
Therefore, it is suitable for a switching system that handles various transmission rates by simply using an existing decoder and encoder, and it becomes highly versatile if detection and encoding control processing is executed by a computer.
[0014]
The above-described configuration further includes delay means interposed in the signal path between the decoding means and the encoding means for delaying the output of the decoding means.
[0015]
According to this configuration, since it takes a processing time to change the pull-down structure or the encoding structure in the re-encoding process, the processing time required for the change can be absorbed by delaying the decoded signal by a predetermined time.
[0016]
In the above configuration, the encoding means skips a dummy field included in the input decoded signal under the control of the encoding control means.
[0017]
According to this configuration, the encoding unit determines whether or not the pull-down structure of the encoded video signal at the time of input matches the pull-down structure of the decoded signal under the control of the encoding control unit. When the determination is made, the dummy field included in the decoded signal is skipped and the encoding process is performed. Therefore, even if an encoded video signal having a different pull-down structure arrives, the encoded video signal It is possible to forcibly match the pull-down structure between the encoded video signal at the time of input and the re-encoded encoded video signal without degrading the quality of the image.
[0018]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
FIG. 1 is a circuit block diagram showing an embodiment of a transcoder device according to the present invention. Here, an example of handling a video ES (Elementary Stream) signal of the MPEG-2 standard having a GOP (Group Of Picture) structure having a 3: 2 pull-down structure and being encoded by at least two kinds of encoding methods as a bit stream. Explained.
In the figure, reference numeral 101 denotes a video decoder, which decodes an input MPEG-2 video ES signal into an original video signal. The decoded video signal is delayed for a predetermined time by the delay unit 102 and supplied to the video encoder 103. The video encoder 103 can arbitrarily set an encoding structure for encoding an input signal or a pull-down structure for encoding. The video encoder 103 encodes the output of the delay unit 102 at an externally designated transmission rate and is not shown as a video ES signal. Output to the serial transmission line switching device. The transmission rate instructed to the video encoder 103 can be set to an arbitrary value from the outside.
[0019]
On the other hand, in the input video ES signal, the GOP header information is detected by the GOP detection unit 104, and at the same time, the pull-down structure is detected by the pull-down detection unit 105. Each detection result is supplied to the encoding control unit 106.
[0020]
The encoding control unit 106 analyzes the GOP structure and pull-down structure of the video ES signal at the time of input from the input detection result, and the encoding method of the video encoder 103 and the pull-down structure to be encoded of the video ES signal at the time of input. Control is performed so as to match the GOP structure and the pull-down structure.
[0021]
Next, the operation in the above configuration will be described with reference to the structure of the MPEG-2 video ES signal shown in FIGS.
That is, as shown in FIG. 2, the video ES signal handled in the above configuration is encoded using an I frame obtained by directly encoding one frame itself by an intra-frame encoding method and prediction from a past frame. Is a GOP structure consisting of P frames and B frames encoded using predictions from both the past and future frames, and a 15-frame repeating pattern consisting of IBBPBBPBBPBBPBB. A GOP header is added.
[0022]
Therefore, in order to determine the encoding method of the video encoder 103, GOP header information is detected from the video ES signal at the time of input, and this detection result is given to the encoding control unit 106. In response to this detection result, the encoding control unit 106 applies a start trigger signal to the video encoder 103 to reset the encoding process of the video encoder 103. Note that the video encoder 103 encodes and outputs the input signal by the intra-frame coding method at the initial time, and then encodes and outputs the signal by the inter-frame coding method, and then resets until the reset is applied. This is a circuit that maintains this state.
[0023]
As a result, the picture code type of each frame output from the video encoder 103 matches the pattern of the picture code type at the time of input to the video decoder 101.
[0024]
As shown in FIG. 3, the video ES signal has a 3: 2 pull-down structure in which each of a plurality of frames is divided into two fields and an invalid dummy field is inserted between the fields, for example, in units of two frames. It is. Here, the same field as the first field is created and inserted between the fields as a dummy field.
[0025]
That is, even when the GOP does not change, a video ES signal different from the normal encoding process may arrive due to the occurrence of 3: 2 pull-down. When the decoding error is small, the video ES signal encoded by the video encoder 103 has almost the same 3: 2 pull-down structure as the video ES signal at the time of input, but the decoding error is actually large and the same. The 3: 2 pull-down structure may not be achieved.
[0026]
Therefore, in order to determine the pull-down structure to be encoded by the video encoder 103, the pull-down detection unit 105 detects the pull-down structure from the video ES signal at the time of input, and gives this detection result to the encoding control unit 106. In response to this detection result, the encoding control unit 106 gives the video encoder 103 control information necessary for setting such as even / odd field encoding and skipping in the 3: 2 pull-down structure. The video encoder 103 grasps the 3: 2 pull-down structure of the input video ES signal based on the given control information, and re-encodes the output of the delay unit 102 with the same pattern as this 3: 2 pull-down structure. The 3: 2 pull-down structure is forcibly matched between input and re-encoding.
[0027]
FIG. 4 is a flowchart showing the operation of the video encoder 103 for achieving a match of the 3: 2 pull-down structure between input and re-encoding.
[0028]
That is, when control information is notified from the encoding control unit 106 (step S11), the video encoder 103 grasps and encodes the 3: 2 pull-down structure of the video ES signal at the input based on this control information. A pull-down structure is determined (step S12). Then, the pull-down structure to be encoded is compared with the pull-down structure of the video signal output from the delay unit 102 (step S13), and it is determined whether or not they match (step S14). Here, when it is determined that they match (YES), a normal encoding process is executed (step S15). When they do not match (NO), the dummy field included in the video signal is skipped and the encoding process is executed. (Step S16).
[0029]
Note that the insertion position of the dummy field to be skipped is detected by the pull-down detection unit 105, and the encoding control unit 106 notifies the video encoder 103 of information indicating the insertion position of the dummy field as control information.
[0030]
As described above, in the above embodiment, the video ES signal is decoded once by the video decoder 101 and returned to the original state, and is encoded again by the video encoder 103 at an arbitrary transmission rate. The GOP detection unit 104 and the pull-down detection unit 105 can detect the GOP structure and 3: 2 pull-down structure of the video ES signal at the time of input in a separate system from the video decoder 101 and the video encoder 103. Then, the encoding control unit 106 receives the detection result and controls the encoding process of the video encoder 103 so as to re-encode the video signal according to the GOP structure and 3: 2 pull-down structure of the video ES signal at the time of input. I am doing so.
[0031]
For this reason, the GOP structure and 3: 2 pull-down structure of the video ES signal at the time of input are inherited at the time of re-encoding by a system different from the video decoder 101 and the video encoder 103. It is possible to minimize image quality fluctuations and deterioration due to inconsistency with the video ES signal, and the video decoder 101 and the video encoder 103 inherit the characteristics of the video ES signal at the time of re-encoding. Therefore, it is not necessary to change or modify the internal circuit. That is, the video decoder 101 and the video encoder 103 do not need to detect the encoding structure of the video ES signal, generate auxiliary information indicating the encoding structure, extract auxiliary information, and perform encoding control processing.
[0032]
Therefore, it is suitable for a switching system that handles various transmission rates only by using the existing video decoder 101 and video encoder 103, and each process of the GOP detection unit 104, pull-down detection unit 105, and encoding control unit 106 is 1 It will be highly versatile if run on one computer.
[0033]
Also, control information for determining a pull-down structure to be encoded by the video encoder 103 is notified from the encoding control unit 106 to the video encoder 103, and the video encoder 103 should be encoded based on the notified control information. A pull-down structure is set, and whether or not the set pull-down structure, that is, the 3: 2 pull-down structure of the video ES signal at the time of input, matches the pull-down structure of the video signal, that is, the 3: 2 pull-down structure that is actually encoded. If it is determined that the two do not match, the dummy field included in the video signal is skipped and the encoding process is performed. Therefore, the video ES signal having a pull-down structure different from the 3: 2 pull-down structure. When the video ES signal at the time of input does not deteriorate, It is possible to match the pull-down structure of the encoded video ES signal. Note that the field to be skipped is an invalid dummy field and does not affect the quality of the video ES signal.
[0034]
In the above embodiment, an example in which the delay unit 102 is interposed in the signal path between the video decoder 101 and the video encoder 103 has been described. The delay unit 102 has a GOP structure and a 3: 2 pull-down structure. The delay time and the presence / absence thereof are determined according to the length of processing time required for detection. Therefore, the delay unit 102 is not necessary for a detection circuit that is sufficiently fast to process. However, in general, it often takes a sufficient processing time to change the GOP structure and pull-down structure of the video encoder 103, so it is better to delay by 0.5 to several frames at present. In this case, by delaying the video signal by the delay unit 102, the encoding method in the video encoder 103 and the processing time required for changing the pull-down structure to be encoded can be absorbed.
[0035]
In the above embodiment, the example in which the encoding scheme of the video encoder 103 is controlled based on the GOP header information included in the video ES signal has been described. However, when the GES header information is not included in the video ES signal. May check the IPB pattern and detect the identification code added to the head of the I frame as the head of the GOP.
[0036]
In addition to GOP, identification information for identifying at least two types of frames encoded by encoding methods with different modes and identifying the type of encoding method corresponding to the own frame at the head of each frame You may make it handle the encoding signal which added. In this case, identification information is detected from the encoded video signal, and the encoding method of the video encoder 103 is controlled based on the content of the detected identification information.
[0037]
Furthermore, in the above-described embodiment, an example in which the video ES signal has a 3: 2 pull-down structure has been described. However, in addition to the 3: 2 pull-down structure, each of a plurality of frames is divided into two fields, and a plurality of frames is obtained. In this case, an encoded video signal having an M: N (M and N are natural numbers) pull-down structure in which dummy fields are inserted between the fields may be handled.
[0038]
In addition, the configuration of the video decoder, video encoder, GOP detection unit, pull-down detection unit, and encoding control unit, the types of bitstreams to be handled, and the like can be variously modified without departing from the scope of the present invention.
[0039]
【The invention's effect】
As described above in detail, according to the present invention, the transmission rate of the encoded signal to be handled can be arbitrarily changed using an existing decoder and encoder, and the encoded video signal has a different structure. In contrast, it is possible to provide a transcoder device that can be flexibly dealt with and can be suitably used for a serial transmission path switching system that handles serial transmission paths of various transmission rates.
[Brief description of the drawings]
FIG. 1 is a circuit block diagram showing an embodiment of a transcoder device according to the present invention.
FIG. 2 is a view for explaining the GOP structure of a video ES signal handled in the embodiment.
FIG. 3 is an exemplary view for explaining a 3: 2 pull-down structure of a video ES signal handled in the embodiment;
FIG. 4 is a flowchart for explaining the operation of the video encoder shown in FIG. 1;
[Explanation of symbols]
101 ... Video decoder,
102 ... delay part,
103 ... Video encoder,
104 ... GOP detection unit,
105 ... pull-down detection unit,
106: Encoding control unit.

Claims (5)

An apparatus for handling an encoded video signal that has been pulled down to M (M is a natural number): N (N is a natural number) and subjected to an encoding process by inserting a dummy field,
Decoding means for decoding the encoded video signal;
Encoding means for encoding the decoded signal decoded by the decoding means at a transmission rate designated from the outside;
Detecting means for detecting a pull-down structure of the encoded video signal input to the decoding means;
Coding control means for controlling the coding process of the coding means based on the detection result of the detection means,
The encoding control unit notifies the encoding unit of control information indicating an insertion position of a dummy field to be skipped even, odd field encoding, and skipping in the pull-down structure from the detection result of the detection unit,
The encoding means grasps the pull-down structure of the encoded video signal at the time of input to the decoding means based on the control information, and compares the grasped pull-down structure with the pull-down structure of the output of the decoding means. A transcoder device that determines whether or not they match, and performs coding processing by skipping a dummy field included in the output of the decoding means if they do not match . Coding structure of the encoded video signal, the transcoder device according to claim 1, characterized in that the Group of Picture structure. The transcoder apparatus according to claim 2 , wherein the detection unit further detects a coding structure from the coded video signal.   2. The transcoder apparatus according to claim 1, further comprising delay means interposed in a signal path between the decoding means and the encoding means for delaying an output of the decoding means.   The encoding control unit controls the encoding unit to execute encoding processing in accordance with a pull-down structure and an encoding structure of an encoded video signal input to the decoding unit based on a detection result of the detection unit. The transcoder apparatus according to claim 1 or 2, characterized in that

Images