WO1999038331A1

WO1999038331A1 - Device and method for detecting moving vector

Info

Publication number: WO1999038331A1
Application number: PCT/JP1999/000317
Authority: WO
Inventors: Takashi Urano
Original assignee: Sanyo Electric Co., Ltd.
Priority date: 1998-01-27
Filing date: 1999-01-26
Publication date: 1999-07-29
Also published as: KR20010034374A; JPH11215502A

Abstract

A device and method for detecting a moving vector with high precision by a small amount of calculation for motion compensative prediction coding for motion picture. A frame moving vector and a field moving vector of an interlace motion picture are detected to perform motion compensation by adaptively switching the vector from one to the other. A sub-sample filter (10) reduces the original picture, and a vector detector (14) detects a moving vector. Then, around the position, vector detectors (30 to 38) detect true moving vectors (frame and field). To detect the moving vectors in the original picture, the search range is so expanded as to extend into the reduced picture. By thus expanding the search range, the detection precision is improved.

Description

Description Motion vector detection device and method

The present invention relates to a motion vector detection apparatus and method, and more particularly to a motion vector detection apparatus and method for use in predictive coding of moving image data. Background art

2. Description of the Related Art Conventionally, there has been motion compensation prediction coding as a method for compressing and coding a moving image. This method uses a motion vector detector to detect motion vectors in blocks (16 pixels x 16 pixels) consisting of multiple pixels, and reduces the code amount of image data using correlation between images. How to

Generally, a frame motion vector between temporally adjacent frames is used as a motion vector. However, an in-line race signal in which one frame is composed of two fields, such as a general video signal, is used. If so, the motion vector (field motion vector) is detected for each field in addition to the frame motion vector, and both the frame motion vector and the field motion vector are adaptively switched to compensate for motion. It is carried out.

As described above, when detecting a frame motion vector or a field motion vector, a motion vector is defined as a unit (macroblock) of a total of 256 pixels of 16 pixels × 16 pixels. However, it is difficult to search for the motion vector over the entire range in the image due to hardware limitations, and the processing takes a long time. Therefore, a so-called hierarchical motion vector detection method that reduces the original image and detects a motion vector in the reduced image has been proposed. In this hierarchical motion vector detection method, an original image is reduced to, for example, 1Z2 to reduce image data, and a motion vector is detected in the reduced image. This motion vector is a guide when detecting a true motion vector, and is referred to as a temporary motion vector here. After that, the detected tentative motion vector is scaled (when the reduction ratio is 1/2, The x-coordinate value and y-coordinate value of the generated temporary motion vector are doubled), and the vicinity (about one pixel of soil) is searched around the position indicated by the scaled temporary motion vector, and This is to detect a true motion vector.

Note that such a hierarchical motion vector detection is disclosed in, for example, Japanese Patent Application Laid-Open Nos. 7-107486 and 8-98186.

However, a temporary motion vector is detected in the reduced image (hereinafter referred to as a first step for convenience), and the detected temporary motion vector is scaled to detect a true motion vector in the original image (hereinafter, referred to as a first step). In this case, the center position of the search range in the second step is obtained by enlarging the provisional motion vector detected in the first step by a reduction ratio (that is, a reduction ratio of 1/2). In this case, it is difficult to detect a true motion vector with high accuracy even if a search is made for about one pixel of soil around that position in the original image. That is, if the provisional motion vector detected in the first step is doubled, for example, the vector indicating the center position of the search in the second step is an even vector (a vector having even components). This is equivalent to simply detecting the motion vector by shifting the macroblock by two pixels in the original image, and the true motion vector cannot be detected with high accuracy. There was a problem.

In particular, when the temporary motion vector is detected in the first step, if the evaluation value of the two adjacent temporary motion vectors A and B, that is, the difference between the prediction errors is small, the evaluation value A is better ( Even if it is adopted as a temporary motion vector, the true motion vector actually exists in the vicinity of B in many cases. However, in the conventional technique, only the vicinity of the temporary motion vector A (± 1 pixel or several pixels centered on the position indicated by the temporary motion vector) is searched in the second step, so that it is close to B. There was a problem that a true motion vector could not be detected. Disclosure of the invention

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned problems of the related art, and has as its object to detect a motion vector with high accuracy without increasing the amount of computation. Motion vector detection device and method capable of performing high-precision motion compensation Is to provide a law.

In order to achieve the above object, the present invention provides a motion vector detection device that detects a motion vector from an original image, and detects a temporary motion vector from an image having less image data than the original image. A first detection unit, and a range including a range searched for when the temporary motion vector is detected, centered on a position determined by the temporary motion vector detected by the first detection unit; And a second detecting means for detecting the motion vector by searching within the device. By overlapping (overlapping) the search ranges, motion vectors can be reliably detected.

Here, the image having less image data than the original image is preferably a reduced image. When the reduction ratio of the reduced image is set to 1 / N, the range searched in the original image may be at least a soil N pixel centered on the position determined by the provisional motion vector. preferable. By using at least N pixels, it is possible to search multiple times. It should be noted that the present invention also includes a technique for setting a search range of at least N pixels or more.The present invention is a motion vector detecting apparatus that detects a frame motion vector and a field motion vector from an original image, It has a detecting means for detecting a frame motion vector from an original image, and a calculating means for calculating the field motion vector using the frame motion vector detected by the detecting means. By calculating the field motion vector from the detected frame motion vector instead of actually detecting it, the amount of calculation can be reduced compared to the case of actually detecting the field motion vector.

Here, the calculating means calculates an even parity vector in the field motion vector as being parallel to the frame motion vector, and calculates an odd parity vector in the above-mentioned intra-field motion vector. Is preferably calculated based on the frame motion vector and the distance between the fields.

Further, the present invention is a motion vector detecting device for detecting a frame motion vector and a field motion vector from an original image, and a detecting means for detecting a field motion vector from the original image; Computing means for calculating the frame motion vector using the field motion vector detected by the detecting means. Sign. Rather than actually detecting the frame motion vector, by calculating from the detected field motion vector by calculation, the amount of calculation can be reduced as compared with the case of actually detecting the frame motion vector.

Also, the present invention is a motion vector detecting device for detecting a frame motion vector and a field motion vector from an original image, wherein the image having less image data than the original image has a temporary frame as a temporary motion vector. First detecting means for detecting a motion vector, and searching in the original image for a range around a position determined by the temporary motion vector and including a range searched when detecting the temporary motion vector. Thus, a second detecting means for detecting the frame motion vector and a calculating means for calculating the field motion vector using the temporary motion vector are provided. By overlapping the search range, the frame motion vector and the field motion vector can be reliably detected, and the amount of computation can be calculated by calculating the field motion vector from the temporary frame motion vector. Can be reduced.

Here, it is preferable that the image having less image data than the original image is a reduced image, and when the reduction ratio is 1 / N, the range searched in the original image is the temporary motion vector. It is preferable that the number of pixels is at least N pixels centered on the determined position.

Further, the present invention is a method for detecting a motion vector from an original image, comprising: a first step of detecting a temporary motion vector from an image having less image data than the original image; And a second step of detecting a motion vector from the original image based on the provisional motion vector thus obtained.In the second step, a position determined by the temporary motion vector is set as a center, The motion vector is detected using a range including the range searched in the first step as a search range.

Here, the image having less image data than the original image in the first step is preferably a reduced image. When the reduction ratio is 1 / N, in the second step, the provisional motion vector is used. It is preferable to search a range of at least ± N pixels centering on the position determined by.

The present invention also relates to a motion vector detecting method for detecting a frame motion vector and a field motion vector from an original image, wherein the frame motion vector and the field motion vector are detected from the original image. A detecting step of detecting a vector; and a calculating step of calculating the field motion vector using the detected frame motion vector. Here, in the calculating step, an even parity vector in the field motion vector is calculated as being parallel to the frame motion vector, and an odd parity in the field motion vector is calculated. It is preferable to calculate the vector based on the frame motion vector and the distance between the fields.

Further, the present invention is a motion vector detection method for detecting a frame motion vector and a field motion vector from an original image, comprising: a detection step of detecting the field motion vector from the original image; A calculating step of calculating the frame motion vector by using the obtained field motion vector.

Further, the present invention is a motion vector detection method for detecting a frame motion vector and a field motion vector from an original image, wherein a temporary motion vector is detected from an image having less image data than the original image. A first step of detecting, and a second step of detecting the frame motion vector and the field motion vector based on the detected temporary motion vector, wherein the first step includes: A temporary frame motion vector is detected as a temporary motion vector, and in the second step, the frame is searched by searching a range including the range searched in the first step with the position determined by the temporary motion vector as a center. A motion vector is detected, and the field motion vector is calculated using the temporary motion vector.

Here, when an image having a smaller number of images than the original image is a reduced image having a ratio of 1 / N, the search range in the second step is centered on a position determined by the provisional motion vector. It is preferable to use at least N pixels. In the second step, an even parity vector in the field motion vector is calculated as being parallel to the frame motion vector, and an odd parity vector in the field motion vector is calculated. Preferably, the torque is calculated based on the frame motion vector and the distance between fields. BRIEF DESCRIPTION OF THE FIGURES FIG. 1 is a configuration diagram of a circuit according to an embodiment of the present invention.

FIG. 2 is an explanatory diagram showing a search range of the first step and the second step in the embodiment of the present invention.

FIG. 3 is an explanatory diagram showing processing for creating a field motion vector from a frame motion vector in the embodiment of the present invention. BEST MODE FOR CARRYING OUT THE INVENTION

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 shows a configuration block diagram of the present embodiment. Note that this configuration program diagram corresponds to conventional hierarchical motion vector detection, in which a frame motion vector and a field motion vector are detected using a hierarchical structure, and both are adaptively detected. This is a frame / field adaptive motion compensation detector that performs motion compensation by switching. The subsample filter 10 converts the input image into a 1/4 reduced image, and reduces the input image in each of the X and y directions to 1/4. The image data reduced in the sub-sample filter 10 is supplied to the frame memory 12 and also to the vector detector 14.

The frame memory 12 stores the reduced image as a reference image. The stored reduced image data is sequentially supplied to the vector detector 14 at an appropriate timing.

The vector detector 14 is a circuit for detecting a temporary motion vector in the first step, and detects a temporary frame motion vector in the present embodiment. More specifically, the reference macroblock readout circuit 14a reads out the reference image (reduced image) stored in the frame memory 12 in units of reduced macroblocks of 4 pixels × 4 pixels, and generates a macroblock. The circuit 14d extracts the reduced macroblock from the processed image output from the subsample filter 10 and supplies both macroblocks to the AE calculator 14c. In addition, the reason why the macroblock, which is the unit for detecting the temporary motion vector, is set to 4 pixels x 4 pixels instead of 16 pixels x 16 pixels is that the original image is reduced to 1/4, so the macroblock It is reduced to 1/4 to correspond to this. When reading a reference macroblock, the vector generator 14b generates a vector by setting a search range centered on (0, 0) and sequentially specifying the vector. The macro program to be read is read.

The AE calculator 14c calculates the error between the reference macroblock and the processing macroblock (the macroblock to be coded). Specifically, assuming that each pixel of the reference macroblock is Aij and each pixel of the processing macroblock is Xij, the evaluation value (prediction error) = ∑IXij-AijI. The calculated evaluation value (prediction error) is supplied to the minimum value comparator 14e.

The minimum value comparator 14e compares the evaluation values calculated for each vector generated by the vector generator 14b with each other, and selects the smallest evaluation value. In addition, the vector at this time is detected as a temporary motion vector (temporary frame motion vector MV). The detected temporary motion vector is supplied to a vector expander 28.

The vector expander 28 scales the input temporary motion vector, and specifically expands both the X component and the y component by the reduction rate. In the present embodiment, since the original image is reduced to 1/4 by the sub-sample filter 10, the provisional motion vector is enlarged four times here.

On the other hand, a frame memory 16, a field memory 18, and a field memory 20 are provided, and store an input image (original image) in frame units and field units, respectively. The field memory 18 stores the odd field data and the field memory 20 stores the even field data. These memories are used as a reference frame or a reference field for detecting a true frame motion vector and a true field motion vector in the second step. That is, the frame image data stored in the frame memory 16 is supplied to the vector detector 30 and used as a reference frame, and the odd field image data stored in the field memory 18 is used as the vector detector. The even field image data stored in the field memory 20 is supplied to the vector detector 32 and used as a reference field in the evening.

The vector detector 30 detects a true frame motion vector. The reference frame from the frame memory 16, the input image delayed by the delay unit 22 (the original image to be processed), And a temporary motion vector quadrupled by the vector magnifier 28

(Temporary frame motion vector) is input to detect a frame motion vector. In addition, The processing unit for detecting the frame motion vector is a normal macro-project unit (16 pixels x 16 pixels), and the delay by the delay unit 22 is to execute the second step after the first step That is to say, after detecting the provisional motion vector with the vector detector 14, the vector detector 30 detects the frame motion vector. The search range for detecting the frame motion vector will be described later.

The vector detector 32 detects the true field motion vector (even parity vector among the four field motion vectors), and refers to the reference odd field from the field memory 18, The input odd field (original image odd field) delayed by the delay unit 24 and the provisional motion vector expanded by the vector expander 28 are input to detect the field motion vector. The block to be processed is 16 pixels × 8 pixels because it is a field unit.

The vector detector 34 is for detecting a true field motion vector (an odd parity vector among the four field motion vectors), and is a reference even number from the field memory 20. Detect the field motion vector by inputting the field, the input odd field delayed by the delay unit 24, and the tentative motion vector enlarged and halved by the vector expander 28. I do. The tentative motion vector is multiplied by 1/2 in consideration of the inter-field distance between the reference even field and the input (processing target) odd field. The details will be described later.

The vector detector 36 detects the true field motion vector (the other odd parity vector among the four field motion vectors), and refers to the reference odd field from the field memory 20; The input even field delayed by the delay unit 26 and the tentative motion vector expanded by the vector expander 28 and multiplied by 3/2 are input to detect a field motion vector. The reason why the tentative motion vector is multiplied by 3/2 is that the distance between the reference odd field and the input (processed) even field is taken into consideration.

The vector detector 38 detects the true field motion vector (the other even parity vector among the four field motion vectors), and refers to the even field referenced from the field memory 20 and the delay unit. Input the input even field delayed in 26 and the tentative motion vector expanded in 28 The motion vector is detected.

Note that the circuit configuration denoted by reference numeral 50 in the figure is a circuit that executes the first step, and the circuit configuration denoted by reference numeral 60 is a circuit that executes the second step.

The motion vector detection device according to the present embodiment has such a configuration. Conventionally, an original image is reduced to, for example, 1/2, and a frame motion vector (temporary frame motion) is reduced in the reduced image. Vector motion vector) and the field motion vector (temporary field motion vector), and scale these temporary motion vectors to find a true frame motion vector within a range of about ± 1 pixel in the original image. In the present embodiment, the true frame motion vector and the true field motion vector are detected using the following two processes. That is,

(1) In the first step, the original image is reduced more than before (for example, reduced to 1/4), and in the second step, the search range is expanded more than before to obtain 16 pixels of soil and already in the first step. The motion vector is detected by searching the range including the searched range.

(2) Instead of detecting the temporary frame motion vector and the temporary field motion vector from the reduced image in the first step as in the past, in the first step, only the temporary frame motion vector is detected (temporary A field motion vector is not detected), but a true frame motion vector and a true field motion vector are detected from the temporary frame motion vector detected in the second step.

As described above, in the present embodiment, the search range of the second step is expanded, and the range when the temporary motion vector is detected in the first step is searched again (that is, the range of the first step and the second step). By overlapping the search ranges), the accuracy of true motion vector detection can be improved. In the first step, only the tentative frame motion vector is detected, and the true frame motion vector and the true field motion vector are detected from the tentative frame motion vector. Can reduce the amount of computation as compared with the case where both the temporary frame motion vector and the temporary field motion vector are detected, and are higher than when motion compensation is performed only from the frame motion vector. Accurate motion compensation can be performed. Hereinafter, these two processes, which are features of the present embodiment, will be described in detail.

FIG. 2 shows the relationship between the motion vector search range in the first step and the motion vector search range in the second step. In the figure, X marks indicate pixels, which are pixels in the original image. A, B, C,..., 0 are pixels searched in the first step when the original image is reduced to 1/4 (that is, every fourth pixel). As a result of detecting the temporary motion vector in the 1/4 reduced image, assume that A in the figure is the position indicated by the temporary motion vector. At this time, in the second step, a true motion vector is searched around the position A in the original image, but in the prior art, as described above, one pixel or at most ± 2, It is just searching for ± 3 pixels. The case where the search range is ± 3 pixels around the position of A is indicated by a broken line in the figure, and is indicated by reference numeral 100. This search range 100 is a range that does not overlap with the search range of the first step. Even if a true motion vector exists between the B position and the C position, such a search range is With this setting, it is impossible to detect the true motion vector.

Therefore, in the present embodiment, a range including the range searched in the first step should be searched in the second step, and at least ± 4 pixels centering on the position A are set as the search range. The minimum search range in the present embodiment is indicated by a chain line in the figure, and is indicated by reference numeral 102. By setting such a search range 102, it is possible to search again in the second step also near the point B or the point C excluded from the tentative motion vector because the prediction error is large in the first step. For example, even if a true motion vector exists between the point B and the point C, it can be reliably detected.

Note that the number of pixels required to overlap (overlap) the search ranges of the first step and the second step is ± 4 pixels as described above (of course, when the reduction ratio is 1 / N, N pixels have the minimum number of pixels), but a range larger than this can be used as the search range.

For example, when a temporary motion vector is detected in the first step, the evaluation value (prediction error) at points A and B in the figure does not differ greatly, and if A <B, In one step, A is adopted as the provisional motion vector, but the difference in the evaluation values In this case, it is quite possible that the true motion vector is not near the position of A but near the position of B. Therefore, in such a case, it is preferable to set the search range to ± 7 pixels centering on the position of A so that the vicinity of the position of B can be searched again in the second step. In the figure, the search range in the case of ± 7 pixels is indicated by a solid line, and is indicated by reference numeral 104. By setting such a search range 104, even if the true motion vector is at the position shown by b in the figure, for example, it is possible to reliably detect this. Note that the upper limit of the search range is determined from the hardware configuration. However, in the present embodiment, the search range is set to 16 pixels from the center of the position A as described above. In the second step, the position of b can be reliably detected.

On the other hand, FIG. 3 schematically shows a process of detecting (calculating) a true field motion vector from the temporary frame motion vector detected in the first step. In the figure, both the reference frame and the encoding target frame are composed of an odd field (odd) and an even field (even), and the pixels in each field are indicated by white circles for convenience. Further, a frame is virtually shown between odd and eVen in both the reference frame and the encoding target frame. Results The pixels on the virtual frame, which in configured _c such fin evening one race configured for adding pixels in the pixel and e V en in odd, was detected tentative frame motion base vector in the first step Assume that the y component (vertical component in the figure) of the temporary frame motion vector MV is y = 10. In the present embodiment, a true frame motion vector MV and a true field motion vector are detected in the second step from the temporary frame motion vector MV (y = 10). As described above, the true frame motion vector scales the temporary frame motion vector MV (y = 10) detected in the first step, and removes 16 pixels from the original image. It can be detected by searching as a search range.

On the other hand, the true field motion vector is created from the frame motion vector as follows. In other words, the true field motion vector includes 0 dd (odd field) of the reference frame, the field motion vector MV o 0 (even parity vector) detected from the odd of the encoding target frame, and the e V en and the mark The field motion vector MVoe (odd parity vector) detected from the odd of the encoding target frame, the field motion vector MVe 0 detected from the odd of the reference frame and the even of the encoding target frame. (Even parity vector), and a field motion vector MVe e (even parity vector) detected from the reference frame eVen and the encoding target frame eVen. It is created from the temporary frame motion vector MV as follows.

As shown in FIG. 3, the temporary frame motion vector MV can be created by translation.

Field motion vector (MVoo) y = Temporary frame motion vector (MV) y / 2… (1)

In addition, 1/2 considers that a shift of one pixel in the y direction of the field motion vector is equivalent to a shift of two pixels in the y direction of the frame motion vector. The X component (MVo o) X of the field motion vector MVo o is equal to (MV) X.

Similar to the field motion vector MVoo, it can be created by translating the temporary frame motion vector MV in parallel.

Field motion vector (MV e e) y = Temporary frame motion vector (MV) y / 2 = Field motion vector (MVo o) y · · · (2)

Note that the X component (MVe e) x of the field motion vector MVe e is equal to (MV) x.

This can be calculated by expanding and contracting the temporary frame motion vector MV according to the distance between the fields. More specifically, as shown in FIG. 3, the motion vector can be created as a motion vector ending at the intersection of the true field motion vector MVoo and the reference frame eVen.

Field motion vector (MVoe) y = 1/2 (MVoo) y + field noise offset = 1/2 (MVoo) y-0.5 (3) Note that the x component (MV oe) x of the field motion vector MV oe is equal to 1/2 (MV) x. In addition, the field parity offset is a quantity peculiar to MPEG2, and the odd parity field motion vector (M Voe) y parallel to (MVoo) y = 0 becomes -0.5 (downward in the figure is positive) This is taken into account.

This can be calculated by expanding and contracting the temporary frame motion vector MV according to the inter-field distance. Specifically, as shown in FIG. 3, the field motion vector M V e e can be extended to create a vector ending at the intersection with o d d of the reference frame.

Field motion vector (MVeo) y = 3/2 (MVoo) y + field parity offset = 3/2 (MVoo) y + 0.5 · · · (4) The X component of the field motion vector MVe o (MVeo) x is equal to 3/2 (MV) X. The field parity offset takes into account that the odd parity field motion vector (MVeo) y parallel to (MVoo) y = 0 is 0.5.

In this way, by creating four field motion vectors from the temporary frame motion vector, it is possible to detect a frame motion vector and a field motion vector by actually searching a predetermined search range as in the past. It is possible to reduce the amount of computation as compared with.

As described above, the embodiment of the present invention has been described in the case of the frame / field adaptive motion compensation in which both the frame motion vector and the field motion vector are used to adaptively switch between them to perform motion compensation. Can be applied to the case where only the frame motion vector is detected without detecting the field motion vector and motion compensation is performed. Specifically, only the temporary frame motion vector is detected in the first step, and only the true frame motion vector is detected based on the temporary frame motion vector detected in the second step. Encoding can be performed by performing motion compensation based on the vector. However, it should be noted that the prediction accuracy is lower when coding is performed based on only the frame motion vector than when coding is performed using both the frame motion vector and the field motion vector. Be It is.

In general, when encoding a moving image by detecting a motion vector using a hierarchical structure, the degree of accuracy is determined by the frame motion vector alone <the frame motion vector + the field calculated from the frame motion vector. The motion vector (in the case of the present embodiment) is a frame motion vector + a field motion vector (actually searched and detected), and (frame motion vector + field motion vector calculated from the frame motion vector (book In the case of the embodiment)) and (frame motion vector + field motion vector (actually searched and detected)), there is almost no difference in accuracy, and the amount of computation is small, so that the prediction encoding of this embodiment is small. Is better.

Although the present embodiment has described the hierarchical motion vector detection, the process of creating a field motion vector from a frame motion vector in the present embodiment can also be applied to non-hierarchical motion vector detection. It is possible. In other words, a predetermined search range is searched without reducing the original image, only the frame motion vector is detected, and a field motion vector is created from this frame motion vector to perform frame / field adaptive motion compensation. It is also possible to do.

Further, in the present embodiment, the provisional frame motion vector is detected in the first step, and the true frame motion vector and the true field motion are detected in the second step based on the provisional frame motion vector. Although the vector was detected, the tentative field motion vector was detected in the first step, and the true frame motion vector and the true field were detected in the second step based on this tentative field motion vector. It is also possible to detect motion vectors. For example, in the first step, only the tentative field motion vector MV oo is detected from the four field motion vectors, and the tentative field motion vector MV oo is scaled. It is possible to create a frame motion vector from, and to create four true field motion vectors MVoo, MVoe, MVeo, and MVee from the temporary field motion vector MVoo. The four field vectors are

Field motion vector (MVoo) y = provisional field motion vector (MVoo) y ... (5)

Field motion vector (MV eo) y = 3/2 Provisional field motion vector (MVo o) y + 0.5 (6)

Field motion vector (MVoe) y = 1/2 Temporary field motion vector (MVo o) y-0. 5

Field motion vector (MVee) y = temporary field motion vector (MVe e) y… (8)

And it is sufficient. However, since the field motion vector greatly depends on the direction of motion in the image (particularly when moving in the vertical direction), the accuracy decreases when the frame motion vector is created from the temporary field motion vector. There is a risk. Therefore, in general, it is desirable to detect the tentative frame motion vector in the first step, and then calculate the field motion vector from this tentative frame motion vector.

In the present embodiment, an image in which the original image is reduced to 1/4 as an image having less image data than the original image has been described. It is also possible to get.

As described above, according to the present invention, it is possible to detect a motion vector with high accuracy with a small amount of computation, and thereby to efficiently predict and encode a moving image.

Claims

The scope of the claims

1. A motion vector detection device that detects a motion vector from an original image, wherein the first detection means detects a temporary motion vector from an image having less image data than the original image;

Searching in the original image for a range including a range searched when detecting the temporary motion vector, centered on a position determined by the temporary motion vector detected by the first detection means. Second detection means for detecting the motion vector by

A motion vector detection device comprising:

2. The apparatus according to claim 1,

An image having a smaller number of images than the original image is a reduced image.

3. The apparatus according to claim 2,

When the reduction ratio is set to 1 / N, the motion range characterized in that the range searched in the original image is at least N pixels centered on the position determined by the provisional motion vector. Vector detector.

4. A motion vector detection device that detects a frame motion vector and a field motion vector from an original image,

Detecting means for detecting a frame motion vector from the original image,

Calculating means for calculating the field motion vector using the frame motion vector detected by the detection means;

A motion vector detection device comprising:

5. A motion vector detection device that detects a frame motion vector and a field motion vector from an original image,

Detecting means for detecting a field motion vector from the original image,

Calculating means for calculating the frame motion vector using the field motion vector detected by the detection means;

A motion vector detection device comprising:

6. The apparatus according to claim 4, wherein

The calculating means calculates an even parity vector in the field motion vector as being parallel to the frame motion vector, and calculates an odd parity vector in the field motion vector. A motion vector detecting device for calculating a torque based on the frame motion vector and a distance between fields.

7. A motion vector detecting device for detecting a frame motion vector and a field motion vector from an original image,

First detection means for detecting a temporary frame motion vector as a temporary motion vector from an image having less image data than the original image,

The frame motion vector is detected by searching the original image for a range around the position determined by the temporary motion vector and including the range searched when detecting the temporary motion vector. Second detecting means for

Calculating means for calculating the field motion vector using the temporary motion vector;

A motion vector detection device, comprising:

8. The apparatus according to claim 7, wherein:

An image having less image data than the original image is a reduced image.

9. The apparatus according to claim 8, wherein:

When the reduction ratio is 1 / N, the range to be searched in the original image is at least N pixels centered on the position determined by the temporary motion vector. Vector detector.

10. A method for detecting a motion vector from an original image,

A first step of detecting a temporary motion vector from an image having less image data than the original image;

A second step of detecting a motion vector from the original image based on the temporary motion vector detected in the first step;

Wherein, in the second step, the motion vector is detected with a position determined by the temporary motion vector as a center and a range including the range searched in the first step as a search range. Motion vector detection method.

11. The method according to claim 10, wherein:

An image having a smaller number of images than the original image in the first step is a reduced image.

1 2. The method according to claim 11, wherein:

In the case where the reduction ratio is 1 / N, the second step is a motion vector characterized by searching at least a range of at least N pixels around a position determined by the temporary motion vector. Detection method.

13. A motion vector detection method for detecting a frame motion vector and a field motion vector from an original image,

A detecting step of detecting the frame motion vector from the original image; a calculating step of calculating the field motion vector using the detected frame motion vector;

A motion vector detection method, comprising:

14. A motion vector detection method for detecting a frame motion vector and a field motion vector from an original image,

A detecting step of detecting the field motion vector from the original image; a calculating step of calculating the frame motion vector using the detected field motion vector;

A motion vector detection method, comprising:

15. The method of claim 13 wherein:

In the calculating step, an even parity vector in the field motion vector is calculated as being parallel to the frame motion vector, and an odd parity vector in the field motion vector is calculated as the odd parity vector in the field motion vector. A motion vector detection method, wherein the motion vector is calculated based on a frame motion vector and a distance between fields.

1 6. A motion vector detection method for detecting a frame motion vector and a field motion vector from an original image,

A second step of detecting the frame motion vector and the field motion vector based on the detected provisional motion vector;

Has,

In the first step, a temporary frame motion vector is detected as the temporary motion vector,

In the second step, the frame motion vector is detected by searching a range around the position determined by the temporary motion vector and including the range searched in the first step. A motion vector detection method, wherein the field motion vector is calculated using a motion vector.

17. The method of claim 16, wherein:

When an image having a smaller number of images than the original image is set as a reduced image having a ratio of 1 / N, the search range in the second step is at least N around a position determined by the temporary motion vector. A motion vector detection method characterized by using pixels.

18. The method of claim 16, wherein:

In the second step, an even parity vector in the field motion vector is calculated as being parallel to the frame motion vector, and an odd parity vector in the field motion vector is calculated. Is calculated based on the frame motion vector and the distance between the fields.