JP2004187023A - Moving image encoding device and moving image decoding device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To allow an area designated by a user to be automatically traced in accordance with movement of a subject, shake, or the like in a moving image processor which improves a subjective image quality by compressing the area designated by the user, with a low compression rate. <P>SOLUTION: Images from an imaging device 10 are turned to encoded data 17 through a movement compensation type moving picture encoding circuit 13. At the time of encoding, an area to be encoded with a high image quality is reported as important area information 19 from an important area selection circuit 14 to the movement compensation type moving picture encoding circuit 13, and the area designated as an important area is compressed at a low compression rate, and the other areas are compressed at a high compression rate. The important area is determined in the important area selection circuit 14 by a pointing device 12 like a mouse and motion vector information from the circuit 13. The important area is inputted to not only the circuit but also a display image composition part 15 and is displayed over an input image on a display device 11. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a video encoding device and a video decoding device, and more particularly to a video processing device capable of recording and transmitting a specific area of an image with high quality.
[0002]
[Prior art]
2. Description of the Related Art In video recording by a videophone or a portable terminal, information compression using a moving image compression method such as MPEG-4 is widely performed. However, when the capacity of the transmission path or file is small, there is a problem that sufficient image quality cannot be obtained. To alleviate this problem, the image frame is divided into areas that are important to the user and areas that are not important to the user, and the important areas have a low compression ratio and the non-important areas have a high compression rate. There are methods for improving image quality. Patent Literature 1 discloses a method for automatically detecting an important area, and Patent Literature 2 discloses a method for specifying an important area by a user.
[0003]
[Patent Document 1]
JP-A-8-130739 (pages 3 to 4, FIGS. 1 to 3)
[Patent Document 2]
JP-A-9-130767 (pages 3 to 6, FIGS. 1 to 8)
[0004]
[Problems to be solved by the invention]
However, the method of automatically detecting an important area as described in Patent Document 1 has a problem of erroneous detection such as detection of an area not intended by the user, and the detection method is complicated to increase the detection accuracy. In this case, there is a problem that the realization cost increases. Further, in the method specified by the user as described in Patent Document 2, there is a problem that an intended object protrudes from a set area when used in a mobile terminal with severe camera shake.
[0005]
SUMMARY OF THE INVENTION The present invention has been made in order to solve the above-mentioned conventional problems, and an object of the present invention is to provide a moving image processing that improves subjective image quality by compressing a region specified by a user at a low compression ratio. An object of the present invention is to provide a moving image encoding device and a moving image decoding device capable of automatically tracking an area designated by a user in accordance with movement of a subject or camera shake with a small amount of computation.
[0006]
[Means for Solving the Problems]
According to the present invention, there is provided a motion-compensated video encoding circuit that divides a video frame into important regions and non-important regions specified in advance and compresses the important regions and the non-important regions using different compression parameters. A moving image encoding apparatus, wherein the important region is corrected based on motion vector information output from the motion compensated moving image encoding circuit, and the corrected important region information is used as the motion compensated moving image. A moving picture coding apparatus characterized by including an important area selecting means for outputting to a picture coding circuit is obtained.
[0007]
And an input unit for designating the important region, wherein the important region selection unit corrects the important region based on input information from the input unit, in addition to the motion vector information. It is characterized by the following. Further, the important area selecting means detects a change amount of the input information from the input means and sets it as an important area position user correction vector, and selects a motion vector in the important area from the motion vector information. A averaging circuit for averaging the motion vector selected by the selection circuit to be an important region position automatic correction vector, adding the important region position user correction vector and the important region position automatic correction vector, Means for correcting the important area information using the addition result.
[0008]
Further, according to the present invention, there is provided a moving picture decoding apparatus including a motion compensation type moving picture decoding circuit for a transmitted moving picture, the pointing device comprising: An input unit, an important region selecting unit that corrects the important region based on the motion vector information output from the motion compensation type video decoding circuit, and outputs the corrected important region information; There is provided a moving picture decoding apparatus characterized by including means for notifying the moving picture coding apparatus of the corrected important area information.
[0009]
Then, the important area selecting means corrects the important area based on input information from the input means in addition to the motion vector information. Further, the important area selecting means detects a change amount of the input information from the input means and sets it as an important area position user correction vector, and selects a motion vector in the important area from the motion vector information. A averaging circuit for averaging the motion vector selected by the selection circuit to be an important region position automatic correction vector, adding the important region position user correction vector and the important region position automatic correction vector, Means for correcting the important area information using the addition result.
[0010]
The operation of the present invention will be described. By automatically correcting the important area specified by the user using the motion vector information output from the motion compensation type video coding circuit or the motion compensation type video decoding circuit, the important area specified by the user with a small amount of calculation can be obtained. The area is automatically tracked in accordance with the movement of the subject or camera shake.
[0011]
BEST MODE FOR CARRYING OUT THE INVENTION
Next, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram showing one embodiment of the present invention. An input image 16 output from an imaging device 10 such as a camera is encoded by a motion compensation type video encoding circuit 13 and converted into encoded data 17. At the time of this encoding, the region to be encoded with high image quality is notified as important region information 19 from the important region selection circuit 14 to the motion compensation type moving image encoding circuit 13.
[0012]
In the motion compensation type moving picture coding circuit 13, the important area and the non-important area are compressed with different compression parameters. That is, the compression operation is performed with the compression ratio in the region specified as the important region set to be small and the compression ratio in other regions set to be large. As a means for changing the compression ratio, adjustment of the quantization step size, adjustment of the strength of high frequency component removal by a noise removal filter, and the like are performed.
[0013]
The important region is basically determined by the important region selection circuit 14 by a user's operation using the pointing device 12 such as a mouse. However, the correction of the important region due to the movement of the subject or camera shake is also performed by the pointing device 12. The important region selection circuit 14 uses the motion vector information 18 output from the compensation type video encoding circuit 13 and the motion vector information 18. The important area information 19 determined and corrected in this way is input to the motion-compensated video coding circuit 13 and is also input to the display image synthesizing unit 15 and is superimposed on the input image 16 to form a display image 20, which is displayed. It is displayed on the device 11.
[0014]
FIG. 3 shows an example of an input image, and FIG. 4 shows an example of a display image in which a frame indicating an important area is overlapped with the input image.
[0015]
The details of the important area selection circuit 14 will be described with reference to FIG. From the motion vector information 18 output from the motion compensation type video encoding circuit 13, the selection circuit 40 selects only the motion vector in the important area. The averaging circuit 41 averages the selected motion vectors to obtain an important region position automatic correction vector 33. The input of the pointing device 12 operated by the user is input to the change vector detection circuit 42, where the change vector is detected, and is set as the important area position user correction vector 46. It is assumed that the initial information of the important area position stored in the memory 45 is information specified in advance by the user using the pointing device 12.
[0016]
Then, the important region position automatic correction vector 33 and the important region position user correction vector 46 are added by the adder 43, and the current important region position stored in the important region position storage memory 45 is further added by the adder 44. The important area information 19 is updated by the addition.
[0017]
FIGS. 5 and 6 are diagrams illustrating the operation of the selection circuit 40 and the averaging circuit 41. FIG. The motion compensation video coding circuit 13 divides the image frame into blocks of 8 × 8 pixels or 16 × 16 pixels, calculates a motion vector for each block, and outputs the motion vector information 18 to the selection circuit 40. I do. The selection circuit 40 divides the block into a block including an important region (important block 30) and a block not including an important region (insignificant block 31), and outputs only the motion vector corresponding to the important block 30 to the averaging circuit 41. Selectively output. The averaging circuit 41 averages the input motion vector of the important block 30 to obtain an important area position automatic correction vector 33.
[0018]
In the above embodiment, the correction of the important region is automatically corrected by the motion vector information 18 and also corrected by the important region position user correction vector 46 using the pointing device 12 of the user. Only the automatic correction may be performed. When automatic correction is performed only based on the motion vector information 18, sufficient correction may not be performed in some cases. Therefore, by performing manual correction by a user, correction by hand shake or the like in an important area can be performed more reliably. .
[0019]
Next, a second embodiment of the present invention will be described. In the second embodiment, in an opposite communication such as a TV (video) telephone, an important area is specified from a receiving side to a transmitting side, and a compression ratio in an important area specified by an encoding circuit on the transmitting side is specified. Is reduced. FIG. 7 is a block diagram showing the second embodiment, and the same parts as those in FIG. 1 are indicated by the same reference numerals.
[0020]
The motion-compensated video decoding circuit 51 decodes the encoded data 50 transmitted by the transmitting terminal (not shown), outputs the decoded image 52 to the display image synthesizing unit 15, and simultaneously sets the motion vector information 18 in the important area selection. Output to the circuit 14. The important area selection circuit 14 determines important area information 19 from the motion vector information 18 and the information from the pointing device 12, outputs the important area information 19 to the display image synthesizing unit 15, and outputs the important area information 19 to the transmitting terminal. The display image synthesizing unit 15 synthesizes the decoded image 52 with the important area information 19, creates a display image 20, and displays the display image 20 on the display device 11.
[0021]
The transmitting terminal performs the same encoding control as in the first embodiment based on the important area information 19 notified from the receiving terminal.
[0022]
In this embodiment, as in the previous embodiment, the correction of the important region is corrected by using the motion vector information 18 and the user input information by the pointing device 12, but only the motion vector information 18 may be used. It is.
[0023]
Next, a third embodiment of the present invention will be described. In the third embodiment, the averaging circuit 41 in FIG. 2 calculates a weighted average in which the motion vector at the center of the important area is weighted among the motion vectors output from the selection circuit 40. I do. As a result, the effect of improving the stability of automatic tracking of the important area can be obtained.
[0024]
Next, a fourth embodiment of the present invention will be described. In the fourth embodiment, the averaging circuit 41 shown in FIG. 2 further averages or averages the weighted or averaged motion vectors in the important area between frames at different times. I do. As a result, the effect of improving the stability of automatic tracking of the important area can be obtained.
[0025]
【The invention's effect】
As described above, according to the present invention, in a moving image processing apparatus that improves subjective image quality by compressing a region specified by a user at a low compression rate, a motion-compensated moving image coding circuit, Automatically corrects important areas by using motion vector information output by the video decoding circuit, and automatically tracks important areas specified by the user according to the movement of the subject or camera shake with a small amount of computation. There is an effect that it becomes possible.
[Brief description of the drawings]
FIG. 1 is a block diagram illustrating a moving picture encoding device according to an embodiment of the present invention.
FIG. 2 is a block diagram of a specific example of an important area selection circuit 14 in FIGS. 1 and 7;
FIG. 3 is an example of an input image 16 of FIG. 1;
FIG. 4 is an example of the display image 20 of FIGS. 1 and 7;
FIG. 5 is a diagram for explaining the operation of the selection circuit 40 of FIGS. 1 and 7;
FIG. 6 is a diagram for explaining the operation of the averaging circuit 41 of FIGS. 1 and 7;
FIG. 7 is a block diagram illustrating a video decoding device according to a second embodiment of the present invention.
[Explanation of symbols]
REFERENCE SIGNS LIST 10 imaging device 11 display device 12 pointing device 13 motion-compensated video coding circuit 14 important area selection circuit 15 display image synthesis unit 20 display image 21 important area 30 important block 31 non-important block 40 selection circuit 41 averaging circuit 42 change Vector detection circuits 43, 44 Adder 45 Important area position storage memory 51 Motion-compensated video decoding circuit

Claims (10)

A moving image encoding apparatus including a motion-compensated moving image encoding circuit that divides a moving image frame into a predetermined important region and a non-important region and compresses the important region and the non-important region with different compression parameters. So,
An important area for correcting the important area based on the motion vector information output from the motion compensated video coding circuit and outputting the corrected important area information to the motion compensated video coding circuit A moving picture coding apparatus comprising a selection unit. Input means for designating the important area, wherein the important area selecting means corrects the important area based on input information from the input means, in addition to the motion vector information. The moving picture encoding device according to claim 1, wherein The important area selecting means,
A change vector detection circuit that detects a change amount of input information from the input unit and sets it as an important region position user correction vector;
A selection circuit for selecting a motion vector in the important region from the motion vector information,
An averaging circuit for averaging the motion vector selected by the selection circuit to obtain an important region position automatic correction vector,
3. A moving picture coding apparatus according to claim 2, further comprising means for adding the important area position user correction vector and the important area position automatic correction vector, and using the addition result to correct the important area information. apparatus. The moving image code according to claim 3, wherein the averaging circuit calculates a weighted average of the motion vectors in the important region, which is weighted among the motion vectors in the important region. Device. The moving picture encoding apparatus according to claim 3, wherein the averaging circuit averages the motion vector in the important area over frames at different times. A video decoding device including a motion-compensated video decoding circuit for a transmitted video,
Input means for indicating an important area in the transmitted moving image frame,
Important area selecting means for correcting the important area based on the motion vector information output from the motion compensation type video decoding circuit, and outputting the corrected important area information;
A moving picture decoding apparatus comprising: means for notifying the corrected important area information to a moving picture coding apparatus on the transmission side. 7. The moving picture decoding apparatus according to claim 6, wherein said important area selecting means corrects said important area based on input information from said input means in addition to said motion vector information. The important area selecting means,
A change vector detection circuit that detects a change amount of input information from the input unit and sets it as an important region position user correction vector;
A selection circuit for selecting a motion vector in the important region from the motion vector information,
An averaging circuit for averaging the motion vector selected by the selection circuit to obtain an important region position automatic correction vector,
8. A moving picture decoding apparatus according to claim 7, further comprising means for adding the important area position user correction vector and the important area position automatic correction vector, and using the addition result to correct the important area information. apparatus. 9. The moving image decoding according to claim 8, wherein the averaging circuit calculates a weighted average of the motion vectors in the important region, which is weighted at the center of the important region. Device. 9. The moving picture decoding apparatus according to claim 8, wherein the averaging circuit averages the motion vector in the important area over frames at different times.

Images