JP4092935B2 - Motion vector correction apparatus and method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an apparatus and method for correcting a motion vector, and more particularly to an apparatus and method for correcting a motion vector that can be corrected to a motion vector adapted to an original motion in a block matching method.
[0002]
[Prior art]
Conventionally, as a scanning system for television broadcasting, an interlace scanning system that scans every other horizontal scanning line is most widely adopted. In this interlace scanning method, an interpolating process is performed on an input image signal to convert a field frequency from 50 GHz to 100 GHz, or a scan conversion from interlaced scanning to progressive scanning. Interference can be suppressed and a converted image with smooth movement can be obtained.
[0003]
In order to perform the motion compensation type conversion described above, in the input image signal, a difference is calculated between the current field and a one-frame delay field obtained by delaying the current field by one frame, and the calculated difference is calculated. Based on this, a motion vector is detected. Then, in the motion correction field located between the current field and the one frame delay field, the pixels are shifted based on the obtained motion vector. Incidentally, the block matching method is generally used for detecting the motion vector as a method which is generally put into practical use.
[0004]
In this block matching method, a reference field (hereinafter referred to as a reference field 80) and a reference field (hereinafter referred to as a reference field 90) are cut out as shown in FIG. Incidentally, in the example of FIG. 7, the one-frame delay field described above corresponds to the reference field 80, and the current field corresponds to the reference field. This reference field 80 is divided into a plurality of reference blocks 101, and a block showing the highest degree of correlation with the reference block 101 is detected from the search block 103 that moves within the search range 104 in the reference field 90. Then, a position shift (direction and magnitude of movement) between the detected search block 103 and the reference block 101 is used as a motion vector.
[0005]
In the above-described correlation degree determination, first, for each pixel value in the search block 103, a difference from the corresponding pixel value in the reference block 101 is obtained, and an evaluation value indicated by the difference, for example, a sum of absolute differences is obtained. Next, the above-described determination operation is performed for all the search blocks 103, and the minimum value is obtained from the obtained evaluation value sum, that is, the sum of the absolute differences. The search block 103 that gives the minimum sum of absolute differences is a block that has the highest degree of correlation with the reference block 101, and is specified between the pixel position 113 at the origin of the block and the origin pixel of the reference block 101. A vector that can be used is a motion vector.
[0006]
Next, based on the vector amount and vector direction of the obtained motion vector, an amount and a shift direction for shifting the pixel are determined. Then, in the motion correction field 100 located between the base field 80 and the reference field 90, the pixel is shifted according to the shift amount determined from the pixel position 113 at the origin of the base block (or the shift amount from the pixel position 113). Only shift the pixel). Here, if the shift amount is ½ of the vector amount of the motion vector, the pixel position of the pixel to be shifted in the motion correction field 100 is at the center of the pixel position in the reference field 80 and the pixel position in the reference field 90. Become. In other words, since a motion correction field can be inserted so that the motion of pixels following the base field 80 to the reference field 90 becomes smooth, it is possible to create an interlaced image that prevents deterioration of the screen quality.
[0007]
[Problems to be solved by the invention]
However, the above-described conventional block matching method is obtained by giving the highest priority to the sum of errors between the blocks 101 and 103 regardless of the movement of the original image displayed on the entire screen. For this reason, when a motion vector irrelevant to the actual motion is obtained, the motion vector varies, which causes deterioration in image quality and failure of processing in units of regions.
[0008]
For this reason, it has been necessary to correct the motion vector variation by correcting the vector direction of the motion vector unrelated to the actual image motion.
[0009]
Therefore, the present invention has been devised in view of the above-described problems, and by correcting the vector direction of a motion vector irrelevant to the actual image motion so as to match the actual image motion, It is an object of the present invention to provide a motion vector correction apparatus and method capable of preventing deterioration of a converted image.
[0010]
[Means for Solving the Problems]
In order to solve the above-described problem, the present invention provides a motion vector correction method for correcting a motion vector at a reference pixel position of an input image signal, and the polarity of a horizontal component of a motion vector at an adjacent pixel position adjacent to the reference pixel position. And the number of adjacent pixel positions in which the polarity of the horizontal component identified in the identification step is positive is compared with the number of adjacent pixel positions in which the polarity of the identified horizontal component is negative. Based on the identification result in the comparison step and the identification step, the average value of the horizontal and vertical components of the motion vector at the adjacent pixel position having a positive polarity is obtained, and the motion vector at the adjacent pixel position having a negative polarity Calculating the average value of the horizontal and vertical components of the horizontal component and the vertical component of the horizontal and vertical components Average value, and based on the comparison result of the comparing step, and a vector correction step of the horizontal and vertical components of the motion vectors of the reference pixel position, In the vector correction step, when the number of adjacent pixel positions where the polarity of the horizontal component identified in the identification step is positive is large based on the comparison result in the comparison step, the polarity obtained in the calculation step The average value of the horizontal and vertical components of the motion vector at the adjacent pixel position where is positive is used as the horizontal and vertical components of the motion vector at the reference pixel position, and the polarity of the horizontal component identified in the identification step is negative. When there are a large number of adjacent pixel positions, the average value of the horizontal and vertical components of the motion vector at the adjacent pixel position having the negative polarity obtained in the calculation step is calculated as the horizontal value of the motion vector at the reference pixel position. Component and vertical component It is characterized by that.
[0011]
The motion vector correction device identifies the polarity of the horizontal component of the motion vector at the adjacent pixel position adjacent to the reference pixel position, and the number of adjacent pixel positions where the polarity of the horizontal component is positive and the polarity of the horizontal component are negative. The number of adjacent pixel positions is compared, and the average of the motion vectors at the adjacent pixel positions is obtained based on the comparison result, thereby correcting the motion vector at the reference pixel position to match the actual image motion.
[0012]
In order to solve the above-described problem, the present invention provides a motion vector correction device for correcting a motion vector at a reference pixel position of an input image signal, and a horizontal component of a motion vector at an adjacent pixel position adjacent to the reference pixel position. The vector identification means for identifying the polarity of the identified pixel is compared with the number of adjacent pixel positions where the polarity of the identified horizontal component is positive and the number of adjacent pixel positions where the polarity of the identified horizontal component is negative. Based on the identification results of the comparison means and the identification means, the horizontal and vertical components of the motion vector having the maximum vector amount are obtained from the motion vectors at adjacent pixel positions having a positive polarity, and the polarity is negative. Calculating means for obtaining a horizontal component and a vertical component of a motion vector having a maximum vector amount among motion vectors at adjacent pixel positions, The horizontal and vertical components of the motion vector determined by the serial arithmetic unit, based on the comparison result by the comparing means, includes a vector correction means for the horizontal and vertical components of the motion vectors of the reference pixel position, The vector correction means, when the number of adjacent pixel positions where the polarity of the horizontal component identified by the vector identification means is positive based on the comparison result by the comparison means is large, is calculated by the calculation means. Among the adjacent pixel positions having a positive polarity, the horizontal and vertical components of the motion vector having the maximum vector amount are set as the horizontal and vertical components in the motion vector of the reference pixel position, and are identified by the vector identifying means. When the number of adjacent pixel positions having a negative polarity of the horizontal component is large, the horizontal component of the motion vector having the maximum vector amount among the adjacent pixel positions having the negative polarity obtained by the calculation means and The vertical component is the horizontal component and the vertical component in the motion vector at the reference pixel position. It is characterized by that.
[0013]
The motion vector correction device identifies the polarity of the horizontal component of the motion vector at the adjacent pixel position adjacent to the reference pixel position, and the number of adjacent pixel positions where the polarity of the horizontal component is positive and the polarity of the horizontal component are negative. The number of adjacent pixel positions is compared, and based on the comparison result, the motion vector at the reference pixel position is obtained by obtaining the horizontal and vertical components of the motion vector that maximizes the vector amount among the motion vectors at the adjacent pixel position. To match the actual motion of the image.
[0014]
In order to solve the above-described problem, the present invention provides a motion vector correction method for correcting a motion vector at a reference pixel position of an input image signal, and a horizontal component of a motion vector at an adjacent pixel position adjacent to the reference pixel position. An identification step for identifying the polarity of the pixel, a number of adjacent pixel positions in which the polarity of the horizontal component identified in the identification step is positive, and a number of adjacent pixel positions in which the polarity of the identified horizontal component is negative. Based on the comparison step to be compared and the identification result in the identification step, the average value of the horizontal and vertical components of the motion vector at the adjacent pixel position having a positive polarity is obtained, and at the adjacent pixel position having a negative polarity. A calculation step for obtaining an average value of horizontal and vertical components of the motion vector, and a horizontal component and a vertical value obtained in the above calculation step. The minute mean value based on the comparison result of the comparing step, a vector correction step of the horizontal and vertical components of the motion vectors of the reference pixel position In the vector correction step, when the number of adjacent pixel positions where the polarity of the horizontal component identified in the identification step is positive is large based on the comparison result in the comparison step, The average value of the horizontal and vertical components of the motion vector at the adjacent pixel position where the polarity is positive is used as the horizontal and vertical components of the motion vector at the reference pixel position, and the horizontal component identified in the identification step When the number of adjacent pixel positions having a negative polarity is large, the average value of the horizontal and vertical components of the motion vector at the adjacent pixel position having the negative polarity obtained in the calculation step is calculated as the reference pixel position. Use horizontal and vertical components in motion vectors It is characterized by that.
[0015]
This motion vector correction method identifies the polarity of the horizontal component of the motion vector at the adjacent pixel position adjacent to the reference pixel position, and the number of adjacent pixel positions where the polarity of the horizontal component is positive and the polarity of the horizontal component are negative. The number of adjacent pixel positions is compared, and the average of the motion vectors at the adjacent pixel positions is obtained based on the comparison result, thereby correcting the motion vector at the reference pixel position to match the actual image motion.
[0016]
Furthermore, in order to solve the above-described problem, the present invention provides a motion vector correction method for correcting a motion vector at a reference pixel position of an input image signal, and a horizontal component of a motion vector at an adjacent pixel position adjacent to the reference pixel position. An identification step for identifying the polarity of the pixel, a number of adjacent pixel positions in which the polarity of the horizontal component identified in the identification step is positive, and a number of adjacent pixel positions in which the polarity of the identified horizontal component is negative. Based on the comparison step to be compared and the identification result in the identification step, the horizontal component and the vertical component of the motion vector having the maximum vector amount among the motion vectors at adjacent pixel positions having a positive polarity are obtained. Among the motion vectors at adjacent pixel positions where is negative, the horizontal and vertical components of the motion vector with the largest vector amount are used. And a vector that uses the horizontal and vertical components of the motion vector obtained in the computation step as the horizontal and vertical components in the motion vector at the reference pixel position based on the comparison result in the comparison step. Correction steps and In the vector correction step, when the number of adjacent pixel positions where the polarity of the horizontal component identified in the identification step is positive is large based on the comparison result in the comparison step, Among the adjacent pixel positions having a positive polarity, the horizontal and vertical components of the motion vector having the maximum vector amount are set as the horizontal and vertical components in the motion vector of the reference pixel position. When the number of adjacent pixel positions where the polarity of the identified horizontal component is negative is large, the horizontal of the motion vector having the maximum vector amount among the adjacent pixel positions determined as negative in the calculation step. The horizontal and vertical components in the motion vector at the reference pixel position are used as the component and the vertical component. It is characterized by that.
[0017]
This motion vector correction method identifies the polarity of the horizontal component of the motion vector at the adjacent pixel position adjacent to the reference pixel position, and the number of adjacent pixel positions where the polarity of the horizontal component is positive and the polarity of the horizontal component are negative. The number of adjacent pixel positions is compared, and based on the comparison result, the motion vector at the reference pixel position is obtained by obtaining the horizontal and vertical components of the motion vector that maximizes the vector amount among the motion vectors at the adjacent pixel position. To match the actual motion of the image.
[0018]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[0019]
FIG. 1 shows a configuration example of a motion vector correction system 10 to which a motion vector correction device according to the present invention is applied. The moving vector correction system 10 is provided to correct a motion vector used for motion compensation of a field to be interpolated. The moving vector correction system 10 includes a first storage memory 21 connected to the motion vector correction device 1 and a second storage memory 21. Storage memory 22 and a control device 23.
[0020]
The first storage memory 21 is a memory for storing a motion vector obtained for each pixel or each block (hereinafter described for each pixel). Based on the request from 1, a motion vector at a predetermined pixel position is passed.
[0021]
The second storage memory 22 stores the corrected motion vector obtained by the motion vector correcting device for each pixel position.
[0022]
The control device 23 examines the variation degree of the motion vector from the motion vectors obtained for each pixel position, and obtains a pixel position for which a motion vector irrelevant to the actual image motion is obtained (hereinafter referred to as this pixel). Is referred to as a reference pixel position).
[0023]
Next, the configuration of the motion vector correction apparatus 1 to which the present invention is applied will be described.
[0024]
The motion vector correction device 1 transmits the reference pixel position notified from the control device 23 to the first storage memory 21. Further, the motion vector correction apparatus 1 obtains, from the first storage memory, motion vectors of a reference pixel position and a pixel position adjacent to the reference pixel position (hereinafter, these pixel positions are referred to as adjacent pixel positions). Receiving and identifying the polarity of the horizontal component of the motion vector at the adjacent pixel location. Incidentally, as shown in FIG. 2, the motion vector supplied from the first storage memory 21 to the motion vector correction apparatus 1 includes the reference pixel position, the reference pixel position, up and down, left and right, upper left, upper right, lower left, and right. Adjacent pixel positions A to H adjacent below. The motion vector correction apparatus 1 compares the number of adjacent pixel positions where the polarity of the horizontal component is positive with the number of adjacent pixel positions where the polarity of the horizontal component is negative, and further calculates the average of the motion vectors at the adjacent pixel positions. To obtain a corrected motion vector in which the motion vector at the reference pixel position is corrected to match the actual motion of the image.
[0025]
In addition, this motion vector correction apparatus 1 may identify the polarity in the horizontal component of the above-mentioned motion vector via the vector identification unit 11 shown in FIG. In addition, the motion vector correction apparatus 1 may perform the comparison of the number of adjacent pixel positions where the polarity of the horizontal component described above is positive or negative via the comparison unit 12. In addition, when the motion vector correction device 1 calculates the average of the motion vectors at the adjacent pixel positions, the motion vector correction device 1 calculates the average value of the horizontal and vertical components of the motion vectors at the adjacent pixel positions according to the comparison result by the comparison unit 12. It may be performed via the calculation unit 13 that can. Furthermore, the motion vector correction device 1 may specify the corrected motion vector at the reference pixel position via the vector correction unit 14.
[0026]
Incidentally, the entire operation flow of the motion vector correction device 1 is controlled by the control device 23 or controlled by a control unit (not shown) provided in the motion vector correction device 1.
[0027]
Next, the motion vector stored in the first storage memory will be described.
[0028]
The television receiver in which the motion vector correction apparatus 1 is built is supplied with an interlaced image signal composed of a field image composed of odd-numbered scanning lines and a field image composed of even-numbered scanning lines. As shown in FIG. 4, among the sequentially provided interlaced image signals, this image signal has a current field as a reference field 40 and a 1-frame delay field delayed by 1 frame from the current field as a reference field 30. . The motion correction field 50 is located between the reference field 30 and the reference field 40, is generated from the reference field 30 and the reference field 40 based on the motion vector, and combined with the reference field 30 at 0/100, One interlaced image is generated.
[0029]
By the way, in the television receiver, by applying a predetermined motion compensation process in the motion compensation field described above, it is possible to suppress visual interference such as a moving character appearing double or a horizontal pan motion being irregular, Prevent screen quality degradation. In order to perform this motion compensation processing, a difference is calculated between the standard field 30 and the reference field 40 in the input image signal, and a motion vector is detected based on the calculated difference (by the way, For the detection of the motion vector, a block matching method is generally used as a method that is generally put into practical use). Usually, based on the vector amount and vector direction of the obtained motion vector, the shift amount and the shift direction are determined, and in the motion correction field 50, the shift amount is shifted from the pixel position of the origin of the reference block by the shift amount. Shift pixels. The control device 23 extracts a reference pixel position unrelated to the actual image motion from the motion vector obtained for each pixel.
[0030]
Next, an actual processing procedure of the motion vector correction apparatus 1 to which the present invention is applied will be described using the flowchart of FIG.
[0031]
First, the motion vector correction device 1 initializes each variable in step S11, and then in step S12, in the first adjacent pixel position (for example, adjacent pixel position A) among the adjacent pixel positions A to H shown in FIG. Extract motion vectors.
[0032]
Next, in step S13, the motion vector correction apparatus 1 identifies the polarity Vx_i of the horizontal component of the motion vector at the adjacent pixel position. If the polarity is positive, the process proceeds to step S14, and if the polarity is negative. The process proceeds to step S15.
[0033]
In step S14, the motion vector correction apparatus 1 adds 1 to a counter Cnt_p (not shown), and extracts the horizontal component of the extracted motion vector to the sum Vx_ave_p of the horizontal components of adjacent pixel positions and the vertical component of the motion vector. The vertical component is added to the sum Vy_ave_p, and the process proceeds to step S16.
[0034]
In step S15, the motion vector correction apparatus 1 adds 1 to a counter Cnt_m (not shown), and the extracted horizontal component of the motion vector is vertical to the sum Vx_ave_m of the horizontal components of adjacent pixel positions. It adds to the sum Vy_ave_m of the components, and proceeds to step S16.
[0035]
In step S16, the motion vector correction apparatus 1 identifies whether or not the adjacent pixel position identified in step S13 is the last adjacent pixel position among the adjacent pixel positions A to H to be extracted. If it is the last adjacent pixel position, the process proceeds to step S18. If it is not the last adjacent pixel position, the process proceeds to step S17.
[0036]
In step S17, the motion vector correction apparatus 1 extracts a motion vector (for example, the adjacent pixel position B) at the next adjacent pixel position, and proceeds to step S13 again. By constructing this routine, it becomes possible to identify motion vectors for all adjacent pixel positions A to H shown in FIG. Further, when the motion vectors are identified for all the adjacent pixel positions A to H and the process proceeds to step S18, the counter Cnt_p indicates the number of adjacent pixel positions where the polarity of the horizontal component of the motion vector is positive, and the counter Cnt_m. The number of adjacent pixel positions where the polarity of the horizontal component of the motion vector is negative is shown. Further, the horizontal component sum Vx_ave_p and the vertical component sum Vy_ave_p at the adjacent pixel position where the polarity of the horizontal component of the motion vector is positive are obtained, and at the adjacent pixel position where the polarity of the horizontal component of the motion vector is negative. The sum Vx_ave_m of the horizontal components and the sum Vy_ave_m of the vertical components are obtained.
[0037]
Incidentally, in step S17, the order in which adjacent pixel positions A to H are extracted may be clockwise, counterclockwise, or random. The adjacent pixel positions to be extracted do not have to be extracted from all the adjacent pixel positions A to H. For example, the adjacent pixel positions A to C, the adjacent pixel positions B, D, and E, or the adjacent pixel positions A and B in the upper stage. , C, D, E.
[0038]
In step S18, the motion vector correction apparatus 1 compares the numerical values of the counter Cnt_p and the counter Cnt_p. As a result, when the counter Cnt_p is larger, it is indicated that there are more adjacent pixel positions where the polarity of the motion vector horizontal component is positive among the adjacent pixel positions, and the process proceeds to step S19. On the other hand, when the counter Cnt_m is larger, it is indicated that there are more adjacent pixel positions where the polarity of the horizontal component of the motion vector is negative among the adjacent pixel positions, and the process proceeds to step S20.
[0039]
In step S19, the motion vector correction apparatus 1 obtains a corrected motion vector at the reference pixel position based on the adjacent pixel position where the polarity of the horizontal component of the motion vector is positive. Each component (Vx_n, Vy_n) of the corrected motion vector is an average value of each component at the adjacent pixel position where the polarity of the horizontal component is positive. Specifically, it is calculated based on the following formula.
Vx_n = Vx_ave_p / Cnt_p
Vy_n = Vy_ave_p / Cnt_p
In step S20, the motion vector correction device 1 obtains a corrected motion vector at the reference pixel position based on the adjacent pixel position where the polarity of the horizontal component of the motion vector is negative. Each component (Vx_n, Vy_n) of the corrected motion vector is an average value of each component at an adjacent pixel position where the polarity of the horizontal component is negative. Specifically, it is calculated based on the following formula.
Vx_n = Vx_ave_m / Cnt_m
Vy_n = Vy_ave_m / Cnt_m
That is, the motion vector correction apparatus 1 to which the present invention is applied performs the processing procedure shown in FIG. 5 described above, thereby counting the motion vector at the reference pixel position and the vector direction of the motion vector at the adjacent pixel position. Correction can be made so as to match the vector direction indicated by the adjacent pixel position. In other words, the majority of motion vector directions of adjacent pixel positions located around the reference pixel position are determined, the direction of the larger motion vector is estimated to be the actual image motion, and the motion of the reference pixel position is determined. The vector direction of the vector can be modified. As a result, it is possible to eliminate variations in motion vectors, and consequently, it is possible to prevent deterioration of the converted image.
[0040]
The present invention is not limited to the embodiment described above. In the flow shown in FIG. 5, the majority is determined according to the horizontal component of the motion vector at the adjacent pixel position, but the majority may never be determined according to the vertical component of the motion vector at the adjacent pixel position. In such a case, a majority decision is made in step S13 by identifying the polarity of the vertical component of the motion vector at the adjacent pixel position and determining whether the polarity is positive or negative.
[0041]
In the flow shown in FIG. 5, each component (Vx_n, Vy_n) of the corrected motion vector is an average value of each component at the adjacent pixel position, but may be replaced with the largest motion vector. For example, when there are many adjacent pixel positions where the polarity of the horizontal component of the motion vector is positive, a motion vector (Vx_max_p, Vy_max_p) having the maximum motion vector amount is searched from the adjacent pixel positions. For each component of the corrected motion vector, Vx_n is set to Vx_max_p, and Vy_n is set to Vy_max_p.
[0042]
The present invention is not limited to the above-described embodiment, and can be applied to the moving vector correction system 19 shown in FIG. In this motion vector correction system 19, a vector storage memory 25 and a control device 23 are connected to the motion vector correction device 1.
[0043]
The vector storage memory 25 includes the functions of the first storage memory 21 and the second storage memory 22 described above, stores a motion vector obtained for each pixel, The corrected motion vector obtained by the vector correction device is stored for each pixel position.
[0044]
The present invention can also be applied to a television receiver adopting a field frequency double speed system that converts a field frequency from 50 Hz to 100 Hz image signal in, for example, a PAL (Phase Alternation by Line) system. It is. By incorporating the present invention in a television that adopts this field frequency double speed system, a motion vector can be obtained while suppressing visual interference such as double-looking moving characters and irregular horizontal pan movement. Therefore, the image quality can be improved synergistically.
[0045]
Further, the present invention is not limited to pixel unit processing, but can be applied to block unit processing. In this case, the reference pixel position may be replaced with a reference block having a predetermined number of pixels, and the adjacent pixel position may be replaced with an adjacent block having a predetermined number of pixels and adjacent to the reference block.
[0046]
Furthermore, the present invention is not limited to the case where it is applied to the interpolation processing of the input image signal in the above-described interlace scanning method, and is also applied to, for example, inter-frame coding based on motion compensation prediction in the MPEG (Moving Picture Expert Group) method. Is possible.
[0047]
【The invention's effect】
As described above in detail, in the motion vector correction apparatus and method according to the present invention, the polarity of the horizontal component of the motion vector at the adjacent pixel position adjacent to the reference pixel position is identified, and the polarity of the identified horizontal component is correct. Is compared with the number of adjacent pixel positions for which the polarity of the identified horizontal component is negative, and based on the identification result, the motion vector at the adjacent pixel position for which the polarity is positive is compared. Obtain an average value of the horizontal component and the vertical component, and obtain an average value of the horizontal component and the vertical component of the motion vector at the adjacent pixel position where the polarity is negative, Based on the comparison result, when the number of adjacent pixel positions where the polarity of the identified horizontal component is positive is large, the horizontal and vertical components of the motion vector at the obtained adjacent pixel position where the polarity is positive are obtained. When the average value is a horizontal component and a vertical component in the motion vector of the reference pixel position, and the number of adjacent pixel positions where the polarity of the identified horizontal component is negative is large, the adjacent pixel position where the polarity is negative The average value of the horizontal and vertical components of the motion vector at is used as the horizontal and vertical components of the motion vector at the reference pixel position. As a result, the motion vector at the reference pixel position is corrected to match the actual image motion. Also, the polarity of the horizontal component of the motion vector at the adjacent pixel position adjacent to the reference pixel position is identified, and the number of adjacent pixel positions where the polarity of the identified horizontal component is positive and the polarity of the identified horizontal component are Compare the number of negative adjacent pixel positions and, based on the identification result, find the horizontal and vertical components of the motion vector that maximizes the vector amount among the motion vectors at the adjacent pixel positions that are positive in polarity. Further, among the motion vectors at adjacent pixel positions having a negative polarity, a horizontal component and a vertical component of the motion vector having the maximum vector amount are obtained, Based on the comparison result, when the number of adjacent pixel positions where the polarity of the identified horizontal component is positive is large, among the adjacent pixel positions where the polarity is positive, the horizontal of the motion vector having the maximum vector amount is obtained. If the component and vertical component are the horizontal and vertical components in the motion vector of the reference pixel position and the number of adjacent pixel positions where the polarity of the identified horizontal component is negative is large, the adjacent polarity is negative. Among the pixel positions, the horizontal and vertical components of the motion vector having the maximum vector amount are set as the horizontal and vertical components in the motion vector of the reference pixel position. As a result, the motion vector at the reference pixel position is corrected to match the actual image motion.
[0048]
As a result, it is possible to eliminate the variation of the motion vector, and thus it is possible to prevent the conversion image from being deteriorated.
[Brief description of the drawings]
FIG. 1 is a diagram showing a configuration example of a motion vector correction system to which a motion vector correction device according to the present invention is applied.
FIG. 2 is a diagram illustrating a relationship between an arrangement of reference pixel positions and adjacent pixel positions.
FIG. 3 is a diagram showing an internal configuration example of a motion vector correction apparatus to which the present invention is applied.
FIG. 4 is a diagram showing interlaced image signals sequentially provided to a motion vector correction apparatus to which the present invention is applied.
FIG. 5 is a flowchart showing an actual processing procedure of the motion vector correction apparatus to which the present invention is applied.
FIG. 6 is a diagram showing another configuration example of the motion vector correction system.
FIG. 7 is a diagram for explaining a conventional block matching method;
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 Motion vector correction apparatus, 11 Vector identification part, 12 Comparison part, 13 Operation part, 14 Vector correction part, 21 1st memory for storage, 22 2nd memory for storage, 23 Control apparatus

Claims (8)

In a motion vector correction device for correcting a motion vector at a reference pixel position of an input image signal,
Vector identifying means for identifying the polarity of the horizontal component of the motion vector at an adjacent pixel position adjacent to the reference pixel position;
Comparing means for comparing the number of adjacent pixel positions where the polarity of the identified horizontal component is positive and the number of adjacent pixel positions where the polarity of the identified horizontal component is negative;
Based on the identification result by the vector identification means, the average value of the horizontal and vertical components of the motion vector at the adjacent pixel position having a positive polarity is obtained, and the horizontal component of the motion vector at the adjacent pixel position having a negative polarity. And an arithmetic means for obtaining an average value of vertical components;
Vector correction means for setting the average value of the horizontal component and the vertical component obtained by the arithmetic means based on the comparison result by the comparison means to be the horizontal component and the vertical component in the motion vector of the reference pixel position ,
The vector correction means, when the number of adjacent pixel positions where the polarity of the horizontal component identified by the vector identification means is positive based on the comparison result by the comparison means is large, is calculated by the calculation means. The average value of the horizontal and vertical components of the motion vector at adjacent pixel positions having a positive polarity is used as the horizontal and vertical components of the motion vector at the reference pixel position, and the polarities of the horizontal components identified by the vector identifying means When the number of adjacent pixel positions having a negative value is large, the average value of the horizontal and vertical components of the motion vector at the adjacent pixel position having the negative polarity obtained by the calculation means is calculated as the motion vector at the reference pixel position. A motion vector correction apparatus characterized by having a horizontal component and a vertical component in   The adjacent pixel position is a pixel position adjacent to the left, right, upper, lower, upper left, upper right, lower left, lower right of the reference pixel position, or a pixel position adjacent to the upper left, upper, upper right of the reference pixel position, 2. The movement according to claim 1, wherein the pixel position is adjacent to the left, upper, right of the reference pixel position, or is adjacent to the left, upper left, upper, upper right, right of the reference pixel position. Vector correction device. The calculation means calculates the sum for each horizontal component and vertical component of the motion vector at the adjacent pixel position, and calculates the total number of adjacent pixel positions where the polarity of the identified horizontal component is positive, or by dividing by the number of neighboring pixel positions polarity of the identified horizontal component is negative, according to claim 1, characterized in that an average value of the horizontal and vertical components of the motion vector in the adjacent pixel position Motion vector correction device. In a motion vector correction device for correcting a motion vector at a reference pixel position of an input image signal,
Vector identifying means for identifying the polarity of the horizontal component of the motion vector at an adjacent pixel position adjacent to the reference pixel position;
Comparing means for comparing the number of adjacent pixel positions where the polarity of the identified horizontal component is positive and the number of adjacent pixel positions where the polarity of the identified horizontal component is negative;
Based on the identification result by the identification means, among the motion vectors at adjacent pixel positions having a positive polarity, the horizontal and vertical components of the motion vector having the maximum vector amount are obtained, and the adjacent pixels having a negative polarity An arithmetic means for obtaining a horizontal component and a vertical component of a motion vector having a maximum vector amount among the motion vectors at the position;
Vector correction means for setting the horizontal and vertical components of the motion vector obtained by the calculation means to be the horizontal and vertical components in the motion vector at the reference pixel position based on the comparison result by the comparison means,
The vector correction means, when the number of adjacent pixel positions where the polarity of the horizontal component identified by the vector identification means is positive based on the comparison result by the comparison means is large, is calculated by the calculation means. Among the adjacent pixel positions having a positive polarity, the horizontal and vertical components of the motion vector having the maximum vector amount are set as the horizontal and vertical components in the motion vector of the reference pixel position, and are identified by the vector identifying means. When the number of adjacent pixel positions having a negative polarity of the horizontal component is large, the horizontal component of the motion vector having the maximum vector amount among the adjacent pixel positions having the negative polarity obtained by the calculation means and A motion vector correction apparatus characterized in that a vertical component is a horizontal component and a vertical component in a motion vector at the reference pixel position . In a motion vector correction method for correcting a motion vector at a reference pixel position of an input image signal,
An identification step for identifying a polarity of a horizontal component of a motion vector at an adjacent pixel position adjacent to the reference pixel position;
A comparison step for comparing the number of adjacent pixel positions in which the polarity of the horizontal component identified in the identification step is positive with the number of adjacent pixel positions in which the polarity of the identified horizontal component is negative;
Based on the identification result in the identification step, an average value of the horizontal component and the vertical component of the motion vector in the adjacent pixel position having a positive polarity is obtained, and the horizontal component of the motion vector in the adjacent pixel position having a negative polarity and A calculation step for obtaining an average value of vertical components;
A vector correction step in which the average value of the horizontal component and the vertical component obtained in the calculation step is set as a horizontal component and a vertical component in the motion vector of the reference pixel position based on the comparison result in the comparison step;
Have
In the vector correction step, when the number of adjacent pixel positions where the polarity of the horizontal component identified in the identification step is positive is large based on the comparison result in the comparison step, the polarity obtained in the calculation step The average value of the horizontal and vertical components of the motion vector at the adjacent pixel position where is positive is used as the horizontal and vertical components of the motion vector at the reference pixel position, and the polarity of the horizontal component identified in the identification step is negative. When there are a large number of adjacent pixel positions, the average value of the horizontal and vertical components of the motion vector at the adjacent pixel position having the negative polarity obtained in the calculation step is calculated as the horizontal value of the motion vector at the reference pixel position. A motion vector correction method characterized by comprising a component and a vertical component . The adjacent pixel position is a pixel position adjacent to the left, right, upper, lower, upper left, upper right, lower left, lower right of the reference pixel position, or a pixel position adjacent to the upper left, upper, upper right of the reference pixel position, 6. The movement according to claim 5, wherein the position is a pixel position adjacent to the left, upper, right of the reference pixel position, or a pixel position adjacent to the left, upper left, upper, upper right, right of the reference pixel position. Vector correction method. In the calculation step, a sum is calculated for each horizontal component and vertical component of the motion vector at the adjacent pixel position, and the calculated sum is the number of adjacent pixel positions where the polarity of the identified horizontal component is positive, or by dividing by the number of neighboring pixel positions polarity of the identified horizontal component is negative, according to claim 5, characterized in that an average value of the horizontal and vertical components of the motion vector in the adjacent pixel position Motion vector correction method. In a motion vector correction method for correcting a motion vector at a reference pixel position of an input image signal,
An identification step for identifying a polarity of a horizontal component of a motion vector at an adjacent pixel position adjacent to the reference pixel position;
A comparison step for comparing the number of adjacent pixel positions in which the polarity of the horizontal component identified in the identification step is positive with the number of adjacent pixel positions in which the polarity of the identified horizontal component is negative;
Based on the identification result in the identification step, the horizontal and vertical components of the motion vector having the maximum vector amount among the motion vectors at the adjacent pixel positions having a positive polarity are obtained, and the adjacent pixels having a negative polarity. A calculation step for obtaining a horizontal component and a vertical component of a motion vector having a maximum vector amount among motion vectors at a position;
A vector correction step in which the horizontal component and the vertical component of the motion vector obtained in the calculation step are set as the horizontal component and the vertical component in the motion vector of the reference pixel position based on the comparison result in the comparison step;
Have
In the vector correction step, when the number of adjacent pixel positions where the polarity of the horizontal component identified in the identification step is positive is large based on the comparison result in the comparison step, the polarity obtained in the calculation step Among the adjacent pixel positions where is positive, the horizontal and vertical components of the motion vector having the maximum vector amount are set as the horizontal and vertical components in the motion vector of the reference pixel position, and the horizontal component identified in the identification step is used. When the number of adjacent pixel positions having a negative component polarity is large, the horizontal and vertical components of the motion vector having the maximum vector amount among the adjacent pixel positions having the negative polarity obtained in the calculation step. Is a horizontal component and a vertical component in the motion vector at the reference pixel position .

Images