TWI558171B - The image coding apparatus and an image transform coding scheme converting method - Google Patents

Description

Image coding mode conversion device and image coding mode conversion method

The present invention relates to an image coding mode conversion apparatus for re-encoding an encoded image signal.

A technique of encoding a moving image to perform encoding has been widely used in recent years. The encoding method of the moving image is, for example, a MPEG-2 (Moving Picture Expert Group) method adopted by a DVD (Digital Versatile Disk)-VIDEO, or a portable type. Terrestrial digital broadcasting (One-segment broadcasting) of the terminal or MPEG-4 AVC (Advanced Video Coding)/ITU-T H.264 method used by Blu-ray (registered trademark) Disk (for example, Non-Patent Document 1).

Further, as disclosed in Patent Document 1, there is also an invention in which a B (Bi-Directional Predictive) image having no motion is deleted for the encoded image signal, thereby further compressing the data of the moving image.

Prior technical literature Patent literature

Patent Document 1: Japanese Laid-Open Patent Publication No. 2004-289876

Non-patent literature

Non-Patent Document 1: MPEG-4 AVC (ISO/IEC 14496-10) /ITU-T H.264 Specifications

In the conventional image coding mode conversion device, after the coded image signal is decoded into an image signal by the image decoding device, image analysis is performed on the decoded image signal, and the decoded image signal is decoded according to the analysis result. Recoding is performed, so there is a problem that a large amount of calculation is necessary.

Further, in the recording apparatus of Patent Document 1, although the B image having no motion is deleted, the B image has a smaller amount of data than the I (Intra) image or the P (Predictive) image, so that a high compression effect cannot be obtained. Question.

In order to solve the above problems, the present invention implements a GOP (Group of Pictures) as a processing unit and performs processing on the last P video of the GOP. Entropy decoding is performed on the encoded P video image signal, and coding information such as a macro block type or an action vector of the macro block unit is obtained. Based on the acquired coded information, it is determined whether there is motion in the image, and if there is no action, the matching image is deleted.

In addition, if all the P images subsequent to the I image are deleted, it is determined that the P image is re-encoded when the I image does not operate.

By performing entropy decoding on the P video and only the I video as the re-encoding target, the amount of calculation related to the image coding mode conversion is reduced.

The image coding mode conversion device of the present invention is characterized by comprising: a coding type determination unit that inputs an intra-coded image signal and an image between frames The coded image signal is a plurality of coded image signals, and the coded image signal is determined to be intra-picture coded or inter-picture coded; and the entropy decoding unit determines that the code type code is determined by the coding type determination unit. Performing entropy decoding processing on the encoded image signal of the signal, and outputting the encoded information of each region of the encoded image signal separated by a predetermined unit; the motion determining unit determining the inter-picture encoding based on the encoded information Whether there is an operation of the image signal; the image deletion unit deletes the inter-picture coded image signal that is determined to be inoperative by the operation determination unit; and the image re-encoding determination unit determines whether or not the coding type is to be determined. The unit determines that the coded image signal of the intra-coded image signal is re-encoded, and the decoding unit decodes the intra-frame coded signal that is determined to be re-encoded by the image re-encoding determination unit, and an encoding unit. The inter-picture coding is performed on the decoded image signal decoded by the decoding unit.

According to the present invention, the present invention includes: an entropy decoding unit that performs entropy decoding processing on the coded P video; an operation determination unit that determines whether or not there is an operation by the coding information output from the entropy decoding unit; and if the operation determination unit determines that there is no action , an image deletion unit that deletes the encoded P image; if the entire encoded P image subsequent to the encoded I image is deleted, the image re-encoding determination unit that determines the re-encoding of the encoded I image; and the image that decodes the encoded I image a signal decoding unit; and an image signal encoding unit that re-encodes the decoded image signal into a P image, and decodes the encoded P image by entropy decoding to determine whether or not there is an action, and encodes the code The I image is re-encoded into a P image, so that it can be recompressed with a small amount of calculation, and the amount of data of the encoded image signal can be reduced.

101‧‧‧Image Type Discrimination Department

102‧‧‧ Entropy Decoding Department

103‧‧‧Action Judgment Department

104‧‧‧Image Removal Department

105‧‧•Image Recoding and Judging Department

106‧‧‧Image Signal Decoding Department

107‧‧‧Image Signal Coding Department

108‧‧‧Image buffer

Fig. 1 is a block diagram showing an example of an image coding apparatus according to the first embodiment of the present invention.

FIG. 2 is a view for explaining a method of determining an operation of the operation determining unit 103 of the image coding apparatus according to the first embodiment of the present invention.

Fig. 3 is a view showing the order of the images for performing the operation determination of the image coding apparatus according to the first embodiment of the present invention.

Fig. 4 is a flowchart showing the flow of processing of the GOP 1 by the image coding apparatus according to the first embodiment of the present invention.

Fig. 5 is a view showing an example of processing of a GOP by the image coding apparatus according to the first embodiment of the present invention.

Fig. 6 is a view showing an example of processing of a GOP by the image coding apparatus according to the first embodiment of the present invention.

Hereinafter, embodiments of the image coding mode conversion device and the image coding mode conversion method of the present invention will be described in detail with reference to the drawings. However, the present invention is not limited by the embodiment.

Embodiment 1.

In the first embodiment of the present invention, an image coding mode conversion apparatus is provided which includes an entropy decoding unit that performs entropy decoding processing on a coded P (Predictive) image, and is output by an entropy decoding unit. Coding information An operation determination unit that has no action; if the operation determination unit determines that there is no operation, the image deletion unit that deletes the coded P picture is deleted; if all the coded P pictures subsequent to the coded I (Intra) picture are deleted, the code is determined. An image re-encoding determination unit for re-encoding the I image; an image signal decoding unit that decodes the encoded I image; and an image signal encoding unit that re-encodes the decoded image signal into a P image, and encodes the P image only By entropy decoding, it is possible to determine whether or not there is an action, and re-encode the encoded I image into a P image, so that the amount of data of the encoded image signal can be reduced by recompressing with a small amount of calculation.

Fig. 1 is a block diagram showing an example of an image coding mode conversion device according to the first embodiment of the present invention. In the figure, the image type determination unit 101 as the coding type determination unit determines that the input coded image signal is an I picture or a P picture. The entropy decoding unit 102 performs entropy decoding on the encoded P video determined to be the P video by the video type determining unit 101, and outputs the encoded information. The motion determining unit 103 determines whether or not the encoded P video has an operation using the encoded information output from the entropy decoding unit. The image deletion unit 104 as the image deletion unit deletes the coded P image and outputs the image deletion information if the operation determination unit 103 determines that there is no operation.

The image re-encoding determination unit 105 as the image re-encoding determination unit determines whether or not to re-encode the encoded I-image based on the image deletion information output from the image deletion unit 104 when the image-type determination unit 101 determines that the P-picture is determined. . The image signal decoding unit 106 as the decoding unit decodes the encoded I video determined to be re-encoded by the video re-encoding determination unit 105, and outputs a decoded image signal. The image signal encoding unit 107 as the encoding unit refers to the past decoded image signal of the decoded image signal stored in the image buffer 108, which is output by the image signal decoding unit 106, and performs image encoding, and outputs re-encoded P. image. The video buffer 108 accumulates the decoded image signal output by the image signal decoding unit 106.

FIG. 2 is a view for explaining an operation determination method in the operation determination unit 103. The encoded image signal is as shown in the figure, and the I image is divided by a block unit called a macroblock. The macroblock is, for example, a block of 16 pixels × 16 lines, and has coded information for each macro block. The coding information has quantization parameters, macroblock type (16×16, 16×8, 8×16, 8×8 in Fig. 2 including the block division category), and if it is an in-frame macroblock The system has an in-frame prediction mode. If it is an inter-frame macroblock, it has reference image information and motion vectors (arrows in Fig. 2).

The motion vector indicates how many macroblocks are related to the reference image, for example, if the maximum value of the motion vector length (norm) in the image exceeds an arbitrary threshold, the motion determining unit 103 determines that there is action. In addition, due to the fine action in the captured image, if there is a block smaller than 16×16 in the block division type of the inter-frame macroblock (16×8, 8×16, 8×8 in Fig. 2) That is, it is determined that the maximum length of the motion vector is smaller than the above threshold.

The third figure shows the image sequence in which the operation determination unit 103 performs the operation determination. There are two GOPs (Group of Pictures) in the figure, and the processing sequence starts from P1-N which is the final image of GOP1, and is traced back to P1-(N-1), and is processed to 11- as the head of the GOP. After 1, move to GOP2. The GOP 2 is similarly processed by the P2-N as the final image.

Figure 4 is a view showing the GOP1 of the first embodiment of the present invention. The flow charter of the process. The image type determination unit 101 determines that the input image is an I image or a P image (ST11). In the case of the P video, the entropy decoding unit 102 performs entropy decoding and outputs encoded information (ST12). The motion determination unit 103 determines whether or not there is an operation by the coded information output from the entropy decoding unit 102 (ST13). If there is an action, the process ends. If there is no action, the matching image is deleted (ST14), and processing of a previous image is started. If all the P images are deleted, and the image type determination unit 101 determines that the I video is determined, it is determined whether or not the video other than the front video has been deleted in the preceding GOP (ST15), and if it has been deleted, the I video is reconverted into Processing of P image (ST16). If it is not deleted, the conversion is not implemented and the processing is terminated.

Fig. 5 shows an example in which the processing of GOP2 is performed by deleting the front image of the GOP1. In Fig. 5, an image surrounded by a solid line indicates an image that has not been deleted, and an image surrounded by a broken line indicates a deleted image. If it is determined that all of the P images are not operating in the GOP 2, the I video is reconverted into a P video of the leading image of the GOP 1 that is referred to before.

Fig. 6 shows an example in which the processing of GOP2 is performed except for deleting the front image of the GOP 1 and the second image. Similarly to the fifth drawing, the undeleted image and the deleted image are indicated by solid lines and broken lines. If it is determined that there is no motion in all the P pictures in the GOP2, since there are two videos remaining in the GOP1 before, the I picture retransformation is not performed.

In the first embodiment, the case where the operation is determined using the maximum value of the length of the motion vector and the block type of the inter-frame macroblock is described. However, other encodings obtained by entropy decoding may be used. Information to make a decision.

Further, in the first embodiment, the processing of the GOP composed of only the I video and the P video is described. However, since the B (Bidirectionally Predictive) video also includes the motion information in the encoded information, it may be configured as A GOP implementation that includes B images.

At this time, even if it is a B image, the same processing as the P image can be performed after considering the reference order between the images.

As described above, the image coding system according to the first embodiment includes: an entropy decoding unit that performs entropy decoding processing on the coded P video; and an encoding information that is output from the entropy decoding unit. If there is no action, the action determination unit determines that there is no action, that is, the image deletion unit that deletes the coded P picture; if all the coded P pictures subsequent to the coded I picture are deleted, it is determined that the coded I picture is re-encoded. An image re-encoding determination unit; an image signal decoding unit that decodes the encoded I video; and an image signal encoding unit that re-encodes the decoded image signal into a P video, and encodes the P video only by entropy decoding, that is, It is possible to determine whether or not there is an action, and re-encode the coded I picture into a P picture, so that recompression can be performed with a small amount of calculation, and the amount of data of the coded picture signal can be reduced.

Industrial availability

As described above, the image coding mode conversion device of the present invention is applicable as an image coding mode conversion device that re-encodes an encoded image signal.

101‧‧‧Image Type Discrimination Department

102‧‧‧ Entropy Decoding Department

103‧‧‧Action Judgment Department

104‧‧‧Image Removal Department

105‧‧•Image Recoding and Judging Department

106‧‧‧Image Signal Decoding Department

107‧‧‧Image Signal Coding Department

108‧‧‧Image buffer

Claims (6)

An image coding mode conversion device includes: an encoding type determining unit that inputs a plurality of encoded image signals by inputting an intra-coded image signal and an inter-coded image signal to determine the encoded image signal system; Encoded in-picture or inter-picture coded; the entropy decoding unit performs entropy decoding processing on the coded picture signal determined by the coding type determination unit as the inter-coded picture signal, and the output is separated by a predetermined unit The coded information of each region of the encoded image signal; the motion determining unit determines whether or not the inter-picture encoded image signal is operated based on the encoded information; and the image deleting unit determines that the motion determining unit determines The image re-encoding image signal is deleted, and the image re-encoding determination unit determines whether to re-encode the encoded image signal determined by the encoding type determining unit as the intra-coded image signal; Decoding the intra-coded signal determined to be re-encoded by the image re-encoding determination unit; and encoding Which will be inter-coded based on the decoded image signal decoded by the decoding unit. The image encoding method conversion device according to the first aspect of the invention, wherein the operation determining unit determines whether or not there is an operation based on information indicating an operation of each of the regions of the encoded information that are determined in a determined unit. An image coding mode conversion device according to claim 1 or 2, wherein The operation determination unit determines whether or not there is an operation based on a further division unit in an area of each of the regions in which the determined unit is divided among the encoded information. The image coding mode conversion device according to any one of claims 1 to 2, wherein the operation determination unit is based on an information amount of each of the pieces of the coded information that is divided by the determined unit. To determine the presence or absence of an action. The image coding method conversion device according to any one of the first aspect of the invention, wherein the image re-encoding determination unit determines that the coded signal is not operated when all the pictures are in the operation determination unit; For recoding. A method for converting an image coding method, comprising: a coding type determining step of inputting a plurality of coded image signals by inputting an intra-coded image signal and an inter-coded image signal, and determining the coded image signal system Encoded in-frame or inter-picture-encoded; entropy decoding step performs entropy decoding processing on the encoded image signal determined as the inter-coded image signal in the encoding type determining step, and the output is divided by a predetermined unit Encoding information of each region of the encoded image signal; an action determining step of determining whether the inter-picture encoded image signal has an action based on the encoding information; and an image deleting step, which is determined in the action determining step Deleting the inter-picture encoded image signal without motion; the image re-encoding determination step determines whether the encoding type is to be determined The encoding step is determined to be a re-encoding of the encoded image signal of the intra-coded image signal; the decoding step is to decode the intra-coded signal determined to be re-encoded in the image re-encoding determination step; and the encoding step, It performs inter-picture encoding on the decoded image signal decoded in the aforementioned decoding step.