TW201322732A - Method for adjusting moving depths of video - Google Patents

Abstract

A method for adjusting moving depths for a video is provided, which is adapted for 2D to 3D conversion. The method includes receiving a plurality of frames at a plurality time points and calculating a plurality of local motion vectors and a global motion vector in each of the frames. The method also includes determining a first difference degree between the local motion vectors and the global motion vector in the frames. The method further includes determining a second difference degree between a current frame and the other frames of the frames. The method also includes calculating a gain value according to the first difference degree and the second difference degree. The method further includes adjusting original depths of the current frame according to the gain value. Accordingly, a phenomenon of depth inversion could be avoided or reformed.

Description

Method of adjusting the moving depth of field of an image

The present invention relates to a method for adjusting the motion depth of field of an image, and more particularly to a method for adjusting the motion depth of field of an image that avoids or improves the depth of field inversion phenomenon.

As display technology advances, displays that provide three-dimensional (3D) imagery have sprung up. The image information required for such a stereoscopic display includes a two-dimensional (2D) image and its depth information. The stereoscopic display can reconstruct the corresponding 3D image by the 2D image and its depth information.

One of the methods for estimating the depth of field of a traditional image is to capture its depth by the degree of motion of the object. This is called the "depth-from-motion (DMP)" method. Among them, objects with a higher degree of motion are given a smaller (nearer) depth; conversely, objects with a lower degree of motion are given a larger (longer) depth.

For a general image, the depth of field obtained by the above DMP method does not cause the phenomenon of depth of field inversion. However, if the image has a windowed-moving object, the depth of field inversion occurs by the conventional DMP method. Please refer to FIG. 1. FIG. 1 is a schematic diagram of an image 100 including a window moving object 120. The image 100 is photographed in such a manner that the photographer himself is in a moving state, for example, the photographer is sitting in an ongoing vehicle such as a car or a train and photographing outside the window. The window moving object 120 presents the captured scene outside the window, while the background 110 in the image 100 presents the scene within the vehicle. In the conventional DMP method, since an object having a higher degree of motion is given a smaller (closer) depth of field, the depth of field of the window moving object 120 is smaller than the depth of field of the background 110, so that the image 100 seen by the viewer is made. The mid-view moving object 120 will be closer to the depth of field than the background 110.

The present invention provides a method of adjusting the motion depth of field of an image, which avoids or improves the depth of field inversion phenomenon.

One embodiment of the present invention provides a method for adjusting the motion depth of field of an image for two-dimensional to three-dimensional image processing. The method includes: (i) receiving a plurality of pictures at a plurality of time points, and calculating relative movement characteristic data of each of the images according to a plurality of narrow-area movement vectors and a wide-area movement vector of each picture; Ii) accumulating the relative movement characteristic data of the above picture to obtain the first cumulative relative movement characteristic data; (iii) accumulating the relative movement characteristic data of the remaining pictures except the latest picture in the above picture to obtain a first Two cumulative relative movement characteristic data; (iv) comparing the relative movement characteristic data of the latest picture with the second cumulative relative movement characteristic data to obtain a comparative relative movement characteristic data; (v) relative moving according to the first cumulative relative movement characteristic data The characteristic data is used to calculate a gain value; and (vi) the original motion depth of field of the recent picture is adjusted according to the gain value.

Another embodiment of the present invention provides a method for adjusting the motion depth of field of an image for two-dimensional to three-dimensional image processing. The method includes: receiving a plurality of pictures at a plurality of time points, and calculating a plurality of narrow-area motion vectors and a wide-area motion vector of each picture; determining the plurality of narrow-area moving vectors and the generalized movement in the plurality of pictures a first degree of difference between the vectors; determining a second degree of difference between a recent picture and the remaining previous pictures in the picture; calculating a gain value according to the first difference degree and the second difference degree; and adjusting the nearest according to the gain value The original motion depth of the picture.

In an embodiment of the present invention, the step (i) includes: (a) calculating a difference between each narrow-domain vector and a wide-area vector for each picture to obtain a plurality of relative motion vectors; and (b) The relative motion data of each picture is obtained according to the narrow field motion vector and the relative motion vector of each picture.

In an embodiment of the present invention, the step (b) includes: (b1) determining whether an absolute value of each narrow-field motion vector is greater than a first threshold; and (b2) determining whether an absolute value of each relative motion vector is greater than a second threshold value; and (b3) obtaining relative movement characteristic data of each picture according to the above judgment result.

In an embodiment of the present invention, the step (b3) includes: calculating, according to the determination result, a plurality of comparison result values corresponding to the plurality of narrow-area units in each screen; and placing the comparison result value along a line The /column map direction maps to produce a map motion vector, and the map motion vector represents relative motion profile data.

In an embodiment of the present invention, the step of generating a mapping motion vector includes: counting the comparison result values in a row/column mapping direction to generate a plurality of count values respectively corresponding to different rows/columns; and The values are respectively compared with a third threshold to generate a plurality of element values of the mapped motion vector based on the comparison of the count value with the third threshold.

In an embodiment of the present invention, the above steps (i) to (v) are respectively performed according to one to a plurality of directions of the above-mentioned screen.

In an embodiment of the invention, each of the relative motion characteristic data in the picture includes a plurality of element values corresponding to different rows/columns. The above step (ii) includes: ORing the element values corresponding to the same row/column in the above picture to obtain the first cumulative relative movement characteristic data.

In an embodiment of the present invention, the step (iii) includes: performing an OR operation on the element values corresponding to the same row/column in the relative movement characteristic data of the remaining pictures to obtain the second cumulative relative movement characteristic data.

In an embodiment of the present invention, the step (iv) includes: combining a plurality of element values of the relative movement characteristic data with opposite element values corresponding to the same row/column of the second cumulative relative movement characteristic data, To obtain comparative relative movement characteristics data.

In an embodiment of the present invention, the step (v) includes: obtaining a first gain value according to the first cumulative relative movement characteristic data; obtaining a second gain value according to comparing the relative movement characteristic data; A gain value and a second gain value calculate a gain value.

In an embodiment of the present invention, the step of obtaining the first gain value includes: obtaining a first gain value from a first gain curve according to a first sum value of one of a plurality of element values of the first cumulative relative motion characteristic data A gain value. The step of obtaining the second gain value includes: obtaining a second gain value from a second gain curve according to comparing the second summation value of one of the plurality of element values of the relative movement characteristic data.

In an embodiment of the invention, each of the first gain value and the second gain value is calculated according to the first and second directions, respectively.

In an embodiment of the invention, the calculating the gain value step comprises: obtaining a product of the first gain value and the second gain value in the first direction; obtaining the first gain value and the second gain value in the second direction; The product of the gain; and the gain value is determined according to one of the two products.

In an embodiment of the present invention, the step of determining the first degree of difference comprises: separately calculating a difference between the narrow field vector and the wide area vector for each picture to obtain a plurality of relative motion vectors; a domain motion vector and a relative motion vector to obtain a relative motion characteristic data of each picture; and cumulatively calculating a relative motion characteristic data of the picture to obtain a first cumulative relative motion characteristic data, and the first cumulative relative motion characteristic data represents the first The degree of difference.

In an embodiment of the present invention, the step of obtaining the relative movement characteristic data of each picture comprises: determining whether an absolute value of the narrow range motion vector is greater than a first critical value; and determining whether an absolute value of the relative motion vector is greater than a first value The second critical value; and the relative movement characteristic data of each picture is obtained according to the above judgment result.

In an embodiment of the present invention, the step of determining the second degree of difference comprises: accumulating the relative movement characteristic data of the remaining pictures of the picture other than the latest picture to obtain a second cumulative relative movement characteristic data; And comparing the relative movement characteristic data of the latest picture with the second cumulative relative movement characteristic data to obtain a comparative relative movement characteristic data, and comparing the relative movement characteristic data to represent the second difference degree.

In an embodiment of the invention, the gain value is set to be smaller as the first difference degree is larger, and the gain value is set to be smaller as the second difference degree is smaller.

Based on the above, the above embodiment obtains the gain value by calculating the first cumulative relative movement characteristic data and comparing the relative movement characteristic data. The first cumulative relative movement characteristic data is used to determine the degree of difference between the narrow field vector and the wide area vector, and the relative moving characteristic data is used to determine the degree of difference between the latest picture and the remaining previous pictures. Therefore, the resulting gain value is related to the degree of difference between the narrow-area vector and the wide-area vector, and is related to the degree of difference between the most recent picture and the rest of the previous picture. Thereby, the motion depth of the recent picture adjusted according to the gain value can truly reflect the shooting situation, and avoid or improve the depth of field inversion phenomenon.

The above described features and advantages of the present invention will be more apparent from the following description.

Please refer to FIG. 2A. FIG. 2A is a functional block diagram of an image processing circuit 200 according to an embodiment of the present invention. The image processing circuit 200 includes a receiving port 210, a logic circuit 220, and a buffer memory 230. The receiving port 210 is configured to receive a plurality of pictures of the image IMG1 at a plurality of time points. 3, having a plurality of picture image IMG1 M0 ~ M9, and the picture M0 ~ M9 correspond to the time point T ₀ ~ T _9. The picture M0 corresponds to the time point T ₀ , the picture M1 corresponds to the time point T ₁ , and so on. Here, the screen M0 is defined as the currently processed screen, and is referred to as the "recent screen", and the remaining screens M1 to M9 are referred to as the "previous screen". It should be understood that although the image IMG1 illustrated in FIG. 3 has 10 pictures, those skilled in the art should understand that the number of pictures of the image IMG1 may be other numbers.

The logic circuit 220 is coupled to the receiving port 210 for performing the method of adjusting the moving depth of field of the image of the present invention. After the logic circuit 220 adjusts the motion depth of the image in any of the images IMG1, a corresponding picture in the other image IMG2 can be generated and output. The motion depth of each picture of the image IMG2 is adjusted by the logic circuit 220, and the image IMG2 can be transmitted to a display device, and the display device displays the corresponding picture based on the image IMG2.

The buffer memory 230 is coupled to the logic circuit 220 for temporarily storing the data generated by the logic circuit 220 during operation. The method of adjusting the motion depth of the image by the image processing circuit 200 will be further described below.

2B is a schematic flow diagram of a method of adjusting the motion depth of field of an image, which may be implemented by image processing circuit 200 of FIG. 2A, in accordance with an embodiment. First, in step S201, the logic circuit 220 receives a plurality of pictures at a plurality of time points, and calculates a plurality of narrow-area motion vectors and a wide-area motion vector for each of the pictures.

Next, in step S202, the logic circuit 220 determines a first degree of difference between the plurality of narrow-range motion vectors and the generalized motion vector among the plurality of pictures. When the degree of difference between the narrow-domain vector and the wide-area vector is high overall in several pictures, it means that there may be a moving object of a certain size in the picture.

Next, in step S203, the logic circuit 220 determines a second degree of difference between a recent picture and the remaining previous pictures among the pictures. When the second difference degree is larger, it indicates that the moving object in the picture has a certain spatial displacement within a certain time.

Next, in step S204, the logic circuit 220 calculates a gain value according to the first difference degree and the second difference degree. Preferably, when the first difference degree is larger, the gain value is set to be smaller; conversely, when the second difference degree is smaller, the smaller the gain value is set. Finally, in step S205, the original motion depth of the recent picture can be adjusted according to the gain value.

As a result, when the method is applied, if the moving object in the picture is a windowed-moving object, the calculated first difference degree is too large, and the calculated second difference degree is biased. Small, and in turn, a small gain value G can be obtained. In this way, when the moving object in the picture is a window moving object, the adjusted depth of field according to the small gain value G is small, thereby avoiding or improving the phenomenon of depth of field inversion.

In contrast, if the moving object in the picture is not a window moving object, the calculated first difference degree and the second difference degree may be too large, thereby generating a large gain value. As a result, the adjusted motion depth of field will be larger, and the user will see the image with normal depth of field.

With continued reference to FIG. 2C, which is a detailed flow chart of a method of adjusting the motion depth of field of an image in accordance with an embodiment, for further details of the steps of FIG. 2B. As shown in FIG. 2C, it includes the following steps:

First, in step S211, the logic circuit 220 receives a plurality of pictures at a plurality of time points, and further calculates each of the pictures according to a plurality of narrow field motion vectors and a wide area motion vector of each of the pictures. One of the respective relative movement characteristics data. This relative mobility characteristic data represents the degree of difference between a plurality of narrow-area motion vectors and a wide-area motion vector in each picture.

Next, in step S212, the logic circuit 220 cumulatively calculates the relative movement characteristic data of the screens to obtain the first cumulative relative movement characteristic data. The first cumulative relative movement characteristic data is used to represent the first degree of difference described in FIG. 2B.

Next, in step S213, the logic circuit 220 cumulatively calculates the relative movement characteristic data of the remaining pictures except the most recent picture among the pictures to obtain a second cumulative relative movement characteristic data. Then, in step S214, the relative movement characteristic data of the latest picture and the second cumulative relative movement characteristic data can be compared to obtain a comparative relative movement characteristic data. Among them, the comparative relative movement characteristic data is used to represent the second degree of difference described in FIG. 2B.

Next, in step S215, a gain value may be calculated according to the first cumulative relative movement characteristic data and the compared relative movement characteristic data. Finally, in step S216, the original motion depth of the recent picture can be adjusted according to the gain value.

It should be noted that the above steps S211 to S215 can be implemented according to one to more directions of the screen, for example, the horizontal direction X or/and the vertical direction Y. When the steps S211 to S215 are performed in accordance with a plurality of directions of the screen, the plurality of directions may be the horizontal direction X and the vertical direction Y. The various steps of the method of adjusting the motion depth of field of the image of FIG. 2C will be further described below using various embodiments.

First, in step S211, the logic circuit 220 calculates a plurality of narrow-area motion vectors and a wide-area motion vector V _G corresponding to the plurality of narrow-area units 410 in each of the pictures M0 to M9. Please refer to FIG. 4A. FIG. 4A illustrates a narrow-area motion vector corresponding to each narrow-area unit in the picture M0 according to an embodiment of the present invention. The picture M0 is divided into a plurality of narrow-area units 410, and each of the narrow-area units 410 has one or more pixels of the picture M0, and the plurality of narrow-area units 410 are arranged in an array of M columns multiplied by N lines, where M and N is a positive integer. Similarly to the screen M0, each of the other screens (for example, the screens M1 to M9) of the image IMG1 has a plurality of narrow-area units 410 arranged in an array of M columns by N rows.

For convenience of description, as shown in FIG. 4A, a plurality of narrow-area motion vectors corresponding to the plurality of narrow-area units 410 in the picture M0 calculated by the logic circuit 220 are respectively V _{(0, 1, 1)} to V _{(0). , M, N)} . Similarly, the narrow-area motion vector corresponding to the narrow-area unit 410 of the j-th column and the k-th row in the picture Mi at the time point T _i is represented by V _{(i, j, k)} , where 0 ≦ i, 1 ≦ j ≦M,1≦k≦N. The manner of calculating the narrow range motion vector and the wide area motion vector V _G may be a calculation of a local motion vector and a global motion vector used in the motion acquisition depth (DMP) method. The manner in which the DMP method is known to those of ordinary skill in the art is not described here.

It is worth noting that, preferably, each narrow-field motion vector extracts components in two directions, such as a horizontal motion vector and a vertical motion vector, wherein the horizontal motion vector and the vertical motion vector are perpendicular to each other. Please refer to FIG. 4B. FIG. 4B illustrates a relative motion vector corresponding to each narrow-area unit 410 in the picture M0 according to an embodiment of the present invention. For convenience of explanation, the narrow-area motion vector corresponding to the narrow-area unit 410 of the j-th column and the k-th row in the picture Mi at the time point T _i is [V _{X(i, j, k)} , V _{Y(i, j, k)} ], where 0≦i,1≦j≦M,1≦k≦N, V _X(i,j,k) represent the horizontal component of the narrow-field motion vector in the horizontal direction X, and V _{Y( i, j, k)} represents the vertical component of the narrow range motion vector in the vertical direction Y. 4B, the plurality of narrow-area motion vectors corresponding to the plurality of narrow-area units 410 in the picture M0 calculated by the logic circuit 220 are respectively [V _X(0 , _{1, 1 )} , V _{Y(0, 1,1)} ]~[V _X(0,M,N) , V _Y(0,M,N) ]. Further, the wide-area motion vector V _G calculated by the logic circuit 220 is represented by [G _X , G _Y ]. Where G _X is the component of the wide-area motion vector V _G in the horizontal direction X, and G _Y is the component of the wide-area motion vector V _G in the vertical direction Y.

Thereafter, the logic circuit 220 compares the plurality of narrow-area vectors with the wide-area vector V _G for each picture to calculate a plurality of relative motion vectors. For convenience of explanation, the relative motion vector corresponding to the narrow-area unit 410 of the jth column and the k-th row in the picture Mi at the time point T _i is represented by Δ _{( i , j , k )} , where 0≦i, 1≦j ≦M,1≦k≦N. Taking FIG. 5A as an example, FIG. 5A illustrates a relative motion vector corresponding to each narrow-area unit 410 in the picture M0. As shown in FIG. 5A, the plurality of relative motion vectors corresponding to the plurality of narrow-area units 410 in the picture M0 calculated by the logic circuit 220 are respectively Δ _{(0, 1, 1 )} to Δ _{(0, M, N).} Said.

In an embodiment of the invention, the logic circuit 220 calculates a difference between the plurality of narrow-area vectors and the wide-area vector V _G for each picture to obtain a plurality of relative motion vectors. In other words, each relative motion vector is obtained according to the following equation (1):

Δ _(i,j,k) =V _(i,j,k) -V _G (1)

It is worth noting that, preferably, when calculating the relative motion vector Δ _{( i , j , k )} , the components of the two directions are respectively calculated. Referring to FIG. 5B, FIG. 5B illustrates a relative motion vector corresponding to each narrow-area unit 410 in the picture M0. For convenience of explanation, the relative motion vector corresponding to the narrow-area unit 410 of the jth column and the k-th row in the picture Mi at the time point T _i is [Δ _{X(i, j, k)} , Δ _{Y(i, j, k )]} represents sum, where 0 ≦ i, 1 ≦ j ≦ M, 1 ≦ k ≦ N, Δ X (i, j, k) represents the component of the vector in the horizontal direction X relative movement, and Δ _{Y (i, j , k)} represents the component of the relative motion vector in the vertical direction Y. Taking FIG. 5B as an example, the plurality of relative motion vectors corresponding to the plurality of narrow-area units 410 in the picture M0 calculated by the logic circuit 220 are respectively [Δ _X(i,1,1) , Δ _{Y(i,1 , 1)} ]~[Δ _X(i,M,N) , Δ _Y(i,M,N) ]. In an embodiment of the invention, the components of the relative motion vectors in the horizontal direction X and the vertical direction Y are obtained according to the following equations (1-1) and (1-2):

Δ _X(i,j,k) =V _X(i,j,k) -G _X (1-1)

Δ _Y(i,j,k) =V _Y(i,j,k) -G _Y (1-2)

Thereafter, the logic circuit 220 obtains a relative movement characteristic data of each picture according to the narrow range motion vector and the relative motion vector of each picture. Please refer to FIG. 6A. FIG. 6A illustrates the relative movement characteristics of each picture. Where H[0] is represented as the relative movement characteristic data of the picture M0; H[1] is represented as the relative movement characteristic data of the picture M1; H[2] is represented as the relative movement characteristic data of the picture M2, and so on.

In an embodiment of the invention, each of the relative movement characteristic data H[0] to H[P] is represented by a one-dimensional matrix or a vector, respectively. As shown in FIG. 6A, each of the relative movement characteristic data H[0] to H[P] includes a plurality of element values corresponding to different rows/columns. Taking the relative movement characteristic data H[1] as an example, the relative movement characteristic data H[1] includes a plurality of element values H[1, 1] to H[1, Q], which correspond to the first to the Mth of the screen M1, respectively. The column, or corresponding to the first to Nth rows of the picture M1; and the relative movement characteristic data H[P] as an example, the relative movement characteristic data H[P] includes a plurality of element values H[P, 1] to H [P, Q] corresponds to the first to Mth columns of the picture MP, respectively, or to the first to Nth lines of the picture MP. The element value H[s, t] represents the t-th element value of the relative movement characteristic data H[s] corresponding to the picture at the time point T _s .

In addition, it is worth noting that, as shown in FIG. 6B, when calculating the relative movement characteristic data H[0] to H[P] of each picture, it is preferable to calculate a horizontal component and a vertical component, respectively. For example, the relative movement characteristic data (H _X [0], H _Y [0]) of the picture M0 includes the horizontal component H _X [0] and the vertical component H _Y [0]; the relative movement characteristic data of the picture M1 (H _X [ 1], H _Y [1]) includes the horizontal component H _X [1] and the vertical component H _Y [1]; the relative movement characteristic data (H _X [2], H _Y [2]) of the picture M2 includes the horizontal component H _X [2] and the vertical component H _Y [2], and so on.

More specifically, the j-th element value of the horizontal component H _X [i] of the relative movement characteristic data (H _X [i], H _Y [i]) corresponding to the picture at the time point T _i may be H _X [ i, j] indicates that the kth element value of the vertical component H _Y [i] of the relative movement characteristic data (H _X [i], H _Y [i]) corresponding to the picture at the time point T _i may be H _Y [i,k] indicates that 1≦j≦M,1≦k≦N. Taking the relative movement characteristic data (H _X [1], H _Y [1]) as an example, the horizontal component H _X [1] includes a plurality of element values H _X [1, 1] to H _X [1, M], The vertical component H _Y [1] includes a plurality of element values H _Y [1,1] to H _Y [1,N]; and the relative movement characteristic data (H _X [P], H _Y [P]) is taken as an example. , its horizontal component H _X [P] includes a plurality of element values H _X [P, 1] ~ H _X [P, M], and its vertical component H _Y [P] includes a plurality of element values H _Y [P, 1] ~H _Y [P,N].

In the above embodiment, the respective element values of the relative movement characteristic data H[1] to H[P] shown in FIG. 6A are respectively used as the respective narrow-area movement vectors V _{(i, j)} on a corresponding row/column. _{whether k)} the absolute value of the relative motion vectors Δ _{(i, j, k)} sufficiently large absolute value symbol. The following is a detailed description of the detailed calculation of the relative movement characteristics data H[0] to H[P].

Regarding the process of obtaining the relative mobility characteristic data H[0] to H[P], in an embodiment of the present invention, the logic circuit 220 obtains the relative movement characteristic data H[0] to H[P] of each picture. The logic circuit 220 first determines whether the absolute value of each narrow-field motion vector V _{(i, j, k)} is greater than the first critical value, and determines whether the absolute value of each relative motion vector Δ _{(i, j, k)} is greater than The second critical value is then obtained based on the above two judgment results to obtain the relative movement characteristic data H[0] to H[P] of each picture.

With regard to the process of obtaining the relative movement characteristic data H[0] to H[P] of each picture based on the above two kinds of judgment results, in an embodiment of the present invention, the logic circuit 220 first calculates the narrow field in each picture. The unit 410 corresponds to a plurality of comparison result values A _{(i, j, k)} , and the comparison result value represents the above two determination results, and then the logic circuit 220 further follows the comparison result values A _{(i, j, k).} A row/column map direction map is generated to generate a map motion vector C _X [0] or C _Y [0], and this map is used to move the vector C _X [0] or C _Y [0] to represent the relative motion characteristic data H above [ 0]~H[P].

Please refer to FIG. 7A. FIG. 7A illustrates a comparison result value A _{(i, j, k)} corresponding to each narrow-area unit 410 in the picture M0 to describe the comparison result value A _{(i, j, k)} and the mapping motion vector. The process of generating C _X [0] or C _Y [0]. For convenience of explanation, the comparison result value corresponding to the narrow-area unit 410 of the jth column and the k-th row in the picture Mi of the time point T _i is represented by A _{(i, j, k)} , where 0 ≦ i, 1 ≦ j ≦M,1≦k≦N. Taking FIG. 7A as an example, the plurality of comparison result values corresponding to the plurality of narrow-area units 410 in the picture M0 calculated by the logic circuit 220 are respectively A _{(0, 1, 1 )} to A _{(0, M, N).} Said.

In an embodiment of the invention, the comparison result values are obtained according to the following equation (2):

Wherein Th1 is the first critical value described above, and Th2 is the second critical value described above. In other words, if |V _(i,j,k) | is greater than the first threshold Th1 and |Δ _(i,j,k) | is greater than the second threshold Th2, the comparison result value A _{(i,j,k) is set.} Equal to 1. In contrast, if |V _(i,j,k) | is not greater than the first threshold Th1 or |Δ _(i,j,k) | is not greater than the second threshold Th2, the comparison result value A _(i,j is set) _{, k)} is equal to 0. Therefore, only |V _(i,j,k) | and |Δ _(i,j,k) | are large enough to set the comparison result value A _(i,j,k) equal to 1, in the remaining cases. The comparison result value A _{(i, j, k) is} set equal to zero.

After obtaining the result value A _{(i, j, k)} , the logic circuit 220 then maps the above comparison result values along the row/column mapping direction to generate a mapping motion vector C _X [0] or C _Y [0], and This mapping movement vector C _X [0] or C _Y [0] represents the relative movement characteristics data. The so-called row/column mapping direction here is, for example, a horizontal direction X or a vertical direction Y in which the horizontal direction X and the vertical direction Y are perpendicular to each other. Taking FIG. 7A as an example, the logic circuit 220 may map the comparison result values A _{(0, 1, 1 )} to A _{(0, M, N)} in the picture M0 in the horizontal direction X to generate a mapping motion vector C _X [0]. ]. Alternatively, the logic circuit 220 may map the comparison result values A _{( 0, 1, 1 )} to A _{(0, M, N)} in the picture M0 in the vertical direction Y to generate a mapping motion vector C _Y [0]. As shown in FIG. 7A, the mapping motion vector C _X [0] may have a plurality of element values C[1] to C[M], and the mapping motion vector C _Y [0] may have a plurality of element values C[M+1 ]~C[M+N]. Each element value C[1]~C[M+N] corresponds to a list of narrow domain units 410 or a row of narrow domain units 410.

In an embodiment, in the process that the logic circuit 220 maps the comparison result value in the row/column mapping direction to generate the mapping motion vector, the logic circuit 220 counts the comparison result value in the row/column mapping direction. To generate a plurality of count values respectively corresponding to different rows/columns, and then compare the above-mentioned count values with the third threshold value Th3, respectively, to generate a map motion vector according to the comparison result of the count value and the third threshold value Th3. Multiple element values.

Taking the picture M0 shown in FIG. 7A as an example, if the above-mentioned row/column mapping direction is the vertical direction Y, that is, Q is equal to M, the logic circuit 220 compares the result value A _(0, 1 _{, 1 )} to A _{(0). , M, N)} are counted in the vertical direction Y to generate a plurality of count values S[1] to S[M] respectively corresponding to different columns, and the above-mentioned count values S[1] to S[M] are respectively The three threshold value Th3 is compared to generate a plurality of element values C[1]-C of the mapping motion vector C _X [0] according to the comparison result of the above-mentioned count values S[1] to S[M] and the third threshold value Th3. [M]. The count values S[1] to S[M] are obtained according to the following equation (3), and the element values C[1] to C[M] are obtained according to the following equation (4):

The mapping motion vector C _X [0] in FIG. 7A is the relative motion characteristic data H[0] in FIG. 6A, and the element values C[1] to C[M] are relative motion characteristic data H[0]. Element values H[0,1]~H[0,Q].

Similarly, if the row/column mapping direction is the horizontal direction X, that is, Q is equal to N, the logic circuit 220 will compare the comparison result values A _{(0, 1, 1 )} to A _{(0, M, N)} in the horizontal direction. X counts to generate a plurality of count values S[M+1] to S[M+N] respectively corresponding to different rows, and the above-mentioned count values S[M+1] to S[M+N] respectively The three threshold value Th3 is compared to generate a plurality of element values C[M+1] of the mapping motion vector according to the comparison result of the count values S[M+1] to S[M+N] and the third threshold value Th3. C[M+N]. The count values S[M+1] to S[M+N] are obtained according to the following equation (5), and the element values C[M+1] to C[M+N] are based on the following equations ( 6) Acquire:

The mapping movement vector C _Y [0] in FIG. 7A is the relative movement characteristic data H[0] in FIG. 6A, and the element values C[M+1] to C[M+N] are relative movement characteristic data H. The element value [0, 1] to H [0, Q] of [0].

It is worth noting that, as mentioned above, in calculating the relative motion characteristics data, it is better to calculate the horizontal component and the vertical component. Therefore, in an embodiment of the invention, the logic circuit 220 can move the horizontal component V _{X(i, of} each narrow domain motion vector [V _{X(i, j, k)} , V _{Y(i, j, k)} ] _. The absolute value of _{j, k) and} the absolute value of the vertical component V _{Y(i, j, k)} are respectively compared with the first critical value Th1, and each relative motion vector [Δ _{X(i, j, k)} , Δ _{Y ( i, j, k)]} of the horizontal component of _{Δ X (i, j, k} ) and the absolute value of the vertical component of _{Δ Y (i, j, k} ) , respectively, the absolute value is compared with a second threshold value Th2, then the logic circuit 220 According to the above comparison result, the relative movement characteristic data of each picture is obtained.

In addition, the logic circuit 220 can also map the plurality of comparison result values in the horizontal direction X and the vertical direction Y representing the result of the determination in the horizontal direction X and the vertical direction Y to generate a horizontal mapping motion vector and a vertical mapping motion vector, respectively. They represent the horizontal and vertical components of the relative motion characteristic data, respectively. Please refer to FIG. 7B. FIG. 7B illustrates comparison result values corresponding to the narrow-area units 410 in the picture M0 in FIG. 6B. Wherein, the comparison result value corresponding to the narrow-area unit 410 of the jth column and the k-th row in the picture Mi at the time point T _i is [A _{X(i, j, k)} , A _{Y(i, j, k)} ] Representation, 0≦i,1≦j≦M,1≦k≦N, and each comparison result value [A _X(i,j,k) , A _Y(i,j,k) ] contains a horizontal comparison result The value A _{X(i, j, k)} and the vertical comparison result value A _{Y(i, j, k)} .

Similar to equation (5), in one embodiment of the present invention, each horizontal comparison result value A _{X(i, j, k)} and the vertical comparison result value A _{Y(i, j, k)} are based on the following equation ( 2-1), (2-2) obtained:

Next, in a similar manner, the logic circuit 220 may map the horizontal comparison result values A _X(0,1,1) to A _X(0,M,N) in the above-described picture M0 in the vertical direction Y to generate a horizontal map. The vector C _X [0] is moved, and the logic circuit 220 can map the vertical comparison result values A _Y(0,1,1) to A _Y(0,M,N) in the above-mentioned picture M0 in the horizontal direction X to generate a vertical Map the motion vector C _Y [0].

In the process of generating the horizontal map vector C _X [0] and the vertical map vector C _Y [0], the logic circuit 220 also horizontally compares the result values A _X(0,1,1) to A _{X(0,M,N )} Y count in the vertical direction to generate a plurality of columns corresponding to different count values _{S X [1] ~ S X} [M], and the vertical comparison value _{a Y (0,1,1) ~ a Y} (0 _{, M, N)} are counted in the horizontal direction X to generate a plurality of count values S _Y [1] to S _Y [N] respectively corresponding to different rows. Next, the logic circuit 220 may apply the following equations to generate the horizontal motion vector map C _X [0] of the plurality of element values _{C X [1] ~ C X} [M] , and a vertical motion vector map C _Y [0] as much as The element values C _Y [1] ~ C _Y [N]. The count values S _X [1] to S _X [M] and S _Y [1] to S _Y [N] are obtained according to the following equations (3-1) and (5-1), and the element value C _X [1] to C _X [M] and C _Y [1] to C _Y [N] are obtained according to the following equations (4-1) and (6-1):

In addition, it is also worth noting that the relative movement characteristics of the data H[0]~H[Q] (whether Hx[0]~H[M] or Hx[0]~H[N]) are the values of the elements. It is a symbol respectively whether the absolute value of each narrow-field motion vector V _{(i, j, k)} on a corresponding row/column and the absolute value of the relative motion vector Δ _{(i, j, k)} are sufficiently large. Therefore, in other embodiments, the relative movement characteristic data H[1]~H[Q] can be calculated in various ways, and the manner of the above specific embodiment is not limited.

Please refer to FIG. 2C and FIG. 6A again. After the logic circuit 220 obtains the relative movement characteristic data H[0] to H[P] of each picture, the logic circuit 220 proceeds to step S212 to compare the relative movement characteristic data H[0] of the (P+1) pictures. H[P] is cumulatively calculated to obtain a first cumulative relative movement characteristic data OR1.

In an embodiment, the logic circuit 220 performs an OR operation on the element values corresponding to the same row/column in the (P+1) pictures to obtain the first cumulative relative movement characteristic data OR1 described above. Mathematically, the first cumulative relative movement characteristic data OR1 has a plurality of element values O[1] to O[Q], and each element value O[1] to O[Q] is a relative movement characteristic data H[0]~ The value of the element value corresponding to the same row/column in H[P] is ORed. In detail, the element values O[1] to O[Q] are obtained according to the following equation (7):

Among them, 1≦q≦Q.

Please refer to FIG. 6B again. After the logic circuit 220 obtains the horizontal components H _X [0] to H _X [P] and the vertical components H _Y [0] to H _Y [P] of the relative movement characteristic data of each picture, the logic circuit 220 will display each picture. The relative movement characteristic data is cumulatively calculated to obtain a first cumulative relative movement characteristic data (OR1 _X , OR1 _Y ), wherein OR1 _X represents a horizontal component of the first cumulative relative movement characteristic data, and OR1 _Y represents a first cumulative relative movement characteristic data. Vertical component. In other words, the logic circuit 220 cumulatively calculates the horizontal components H _X [0] to H _X [P] and the vertical components H _Y [0] to H _Y [P] of the relative movement characteristic data of the (P+1) pictures. Obtain the first cumulative relative movement characteristic data (OR1 _X , OR1 _Y ).

It should be noted that, as shown in FIG. 6B, the first cumulative relative movement characteristic data can also be calculated separately in the vertical direction and the horizontal direction. Furthermore, the first cumulative relative movement characteristic data (OR1 _X , OR1 _Y ) has a plurality of element values O _X [1] to O _X [M], O _Y [1] to O _Y [N], such as It can be obtained according to the following equations (7-1) and (7-2):

O _X [j]=H _X [0,j]∨H _X [1,j]∨H _X [2,j]∨...H _X [P,j] (7-1)

O _Y [k]=H _Y [0,k]∨H _Y [1,k]∨H _Y [2,k]∨...H _Y [P,k] (7-2)

Among them, 1≦j≦M,1≦k≦N.

In addition, it is also worth noting that the first cumulative relative movement characteristic data OR1 or (OR1 _X , OR1 _Y ) is used to determine the degree of overall difference between the narrow-domain vector and the wide-area vector in several consecutive pictures. When the degree of difference between the narrow-domain vector and the wide-area vector is high overall in several pictures, it means that there may be a certain size of moving object in the picture. Therefore, in other embodiments, the relative movement characteristic data OR1 may be calculated according to other manners to represent such a difference, and is not limited to the specific manner exemplified in this embodiment.

Next, step S213 in Fig. 2C will be explained. In addition to obtaining the first cumulative relative movement characteristic data OR1, the logic circuit 220 also excludes the most recent picture M0, and accumulates the relative movement characteristic data H[1] to H[P] of the remaining pictures among the plurality of pictures. A second cumulative relative movement characteristic data OR2 is obtained, as shown in FIG. 6A. In an embodiment of the invention, the second cumulative relative movement characteristic data OR2 includes a plurality of element values U[1] to U[Q] corresponding to different rows/columns. Similar to the first cumulative relative movement characteristic data OR1, each element value U[1] to U[Q] is a logic circuit 220 corresponding to the relative movement characteristic data H[1] to H[P] of the remaining pictures described above. The element values of the same row/column are obtained by OR operation. In detail, the element values U[1] to U[Q] are obtained according to the following equation (8):

U[q]=H[1,q]∨H[2,q]∨H[3,q]∨...H[P,q] (8)

Among them, 1≦q≦Q.

It should be noted that, as shown in FIG. 6B, the logic circuit 220 can also calculate the second cumulative relative movement characteristic data (OR2 _X , OR2 _Y ) in two directions, wherein OR2 _X represents the horizontal component of the second cumulative relative movement characteristic data. OR2 _Y represents the vertical component of the second cumulative relative motion characteristic data. In an embodiment of the invention, the second cumulative relative movement characteristic data (OR2 _X , OR2 _Y ) includes a plurality of element values U _X [1] to U _X [M], U _Y [1] corresponding to different rows/columns ]~U _Y [N], which can be obtained according to the following equations (8-1), (8-2):

U _X [j]=H _X [1,j]∨H _X [2,j]∨H _X [3,j]∨...H _X [P,j] (8-1)

U _Y [k]=H _Y [1,k]∨H _Y [2,k]∨H _Y [3,k]∨...H _Y [P,k] (8-2)

Among them, 1≦j≦M,1≦k≦N.

Next, step S214 in Fig. 2C will be explained. The logic circuit 220 compares the relative movement characteristic data H[0] of the most recent picture M0 with the second cumulative relative movement characteristic data OR2 to obtain a comparison relative movement characteristic data AND1 (as shown in FIG. 6A). In the course of performing this comparison, for example, the opposite data of the relative movement characteristic data H[0] of the latest picture M0 and the second cumulative relative movement characteristic data OR2 (the element value 1 becomes 0, and 0 becomes 1) Perform an AND operation to obtain a comparison relative movement characteristic data AND1.

More specifically, the element values H[0, 1] to H[0, Q] of the relative movement characteristic data H[0] of the recent picture correspond to the same row/column of the second cumulative relative movement characteristic data OR2. Opposite element value to Perform and operate to obtain the comparison relative movement characteristic data AND1. Therefore, the comparative relative movement characteristic data AND1 includes a plurality of element values A[1] to A[Q], and the element values A[1] to A[Q] are obtained according to the following equation (9):

Among them, 1≦q≦Q.

It should be noted that, as shown in FIG. 6B, the logic circuit 220 compares the relative movement characteristic data (H _X [0], H _Y [0]) and the second cumulative relative movement characteristic data (OR2 _X , OR2) of the latest picture M0. _Y ) to obtain comparative relative motion characteristic data (AND1 _X , AND1 _Y ), where AND1 _X is the horizontal component of the comparative relative motion characteristic data, and AND1 _Y is the vertical component of the comparative relative motion characteristic data. Similarly, the comparative relative movement characteristic data (AND1 _X , AND1 _Y ) contains a plurality of element values A _X [1] to A _X [M], A _Y [1] to A _Y [N], respectively The equations (9-1) and (9-2) are obtained:

Among them, 1≦j≦M,1≦k≦N.

In addition, it should be noted that the comparative relative movement characteristic data AND1 is used to determine the degree of difference between the most recent picture M0 and the remaining previous pictures. When the degree of difference is larger, it means that the moving object in the picture has a certain spatial displacement within a certain time. Therefore, in other embodiments, the relative movement characteristic data AND1 may be calculated according to other manners to represent such a difference, and is not limited to the specific manner exemplified in this embodiment.

Finally, steps S215 and S216 in FIG. 2C can be performed, wherein the logic circuit 220 can calculate according to the first cumulative relative movement characteristic data OR1 or (OR1X, OR1Y) and the comparative relative movement characteristic data AND1 or (AND1X, AND1Y). A gain value G is obtained, and the original motion depth of field corresponding to the narrow-area unit 410 in the most recent picture M0 is adjusted according to the calculated gain value G.

Please refer to FIG. 8. FIG. 8 illustrates the original motion depth of field corresponding to each of the narrow-area units 410 in the picture M0. For convenience of explanation, the original motion depth field corresponding to the narrow field unit 410 of the jth column and the kth line in the picture Mi at the time point T _i is represented by D _{(i, j, k)} , where 0 ≦ i, 1 ≦ j ≦M,1≦k≦N. As shown in FIG. 8 , the original motion depths corresponding to the narrow-area units 410 in the picture M0 calculated by the logic circuit 220 are represented by D _{(0, 1, 1 )} to D _{(0, M, N)} , respectively. It is assumed here that the gain value calculated by the logic circuit 220 is G, and the adjusted motion depth corresponding to the narrow-area unit 410 of the jth column of the j-th row in the picture Mi at the time point T _i is equal to (D _{(i) , j, k)} × G).

Regarding the process of calculating the gain value G (step S215), in a preferred embodiment, the logic circuit 220 can obtain a first gain value Gain1 according to the first cumulative relative movement characteristic data OR1, and according to the comparison relative movement characteristic data AND1 , obtaining a second gain value Gain2. Preferably, each of the first gain value Gain1 and the second gain value Gain2 can be calculated according to the first and second directions (for example, the row direction or the column direction). Next, the logic circuit can calculate the gain value according to the first gain value Gain1 and the second gain value Gain2. The above process will be described in detail below.

In an embodiment of the invention, the logic circuit 220 adds a plurality of element values O[1] to O[Q] of the first cumulative relative movement characteristic data OR1 to obtain a first total value Or_C. Wherein, the first total value Or_C is obtained according to the following equation (10):

Thereafter, the logic circuit 220 obtains the first gain value Gain1 according to the first total value Or_C. In an embodiment of the invention, the logic circuit 220 obtains the first gain value Gain1 from the first gain curve C1 according to the first summed value Or_C.

As shown in Figure 9, it shows a first gain curve C1 in accordance with an embodiment. As shown in FIG. 9, the first gain curve C1 is an increasing curve, so the larger the first gain value Ga_1 corresponding to the first total value Or_C is larger. It should be understood that the first total value Or_C (or the first cumulative relative movement characteristic data OR1) can be used to determine the degree of overall difference between the narrow-domain vector and the wide-area vector in several consecutive pictures. When the degree of difference between the narrow-domain vector and the wide-area vector is higher in several pictures, it means that there may be a certain size of the moving object in the picture, and the calculated first total value Or_C is relatively Big.

Similarly, in the process of obtaining the second gain value Gain1, the logic circuit 220 first adds a plurality of element values A[1] to A[Q] of the relative movement characteristic data AND1 to obtain the second total value And_C. . Wherein, the second total value And_C is obtained according to the following equation (11):

Next, the logic circuit 220 can obtain the second gain value Gain2 according to the second summed value And_C. In an embodiment of the invention, the logic circuit 220 obtains the second gain value Gain2 from the second gain curve C2 according to the second summed value And_C.

Figure 10 is a second gain curve C2 in accordance with an embodiment. As shown in the figure, the second gain curve C2 is a decreasing curve, so the second second gain value Gain2 corresponding to the larger second total value And_C is smaller. The second summed value And_C (or the comparison relative movement characteristic data AND1) can be used to determine the degree of difference between the most recent picture M0 and the remaining previous pictures. When the difference between the recent picture M0 and the rest of the previous pictures is greater, it indicates that the moving object in the picture may have a certain spatial displacement within a certain time, and the calculated second total value And_C will be larger.

Therefore, if the moving object in the picture is a windowed-moving object, the calculated first total value Or_C will be too large, and the second total value And_C will be too small, thereby obtaining A small gain value G. In this way, when the moving object in the picture is a window moving object, the adjusted depth of field according to the small gain value G is small, thereby avoiding or improving the phenomenon of depth of field inversion.

In contrast, if the moving object in the picture is not a window moving object, the calculated first total value Or_C and the second total value And_C are both larger, and a larger gain value G is obtained. In this way, when the moving object in the picture is not the moving object of the window, the adjusted depth of field will be larger due to the larger gain value G, and the user will see the image with normal depth of field.

Next, the logic circuit 220 may calculate the gain value G according to the first gain value Gain1 and the second gain value Gain2. In an embodiment of the invention, the gain value G is obtained according to the following equation (12):

G=1-Gain1×Gain2 (12)

Among them, 0≦Gain1≦1 and 0≦Gain2≦1, so 0≦G≦1.

It should be noted that, preferably, the first total value, the second total value, the first gain value and the second gain value are equally calculated in two directions. In an embodiment of the invention, the logic circuit 220 adds a plurality of element values O _X [1] to O _X [M] of the middle horizontal component of the first cumulative relative movement characteristic data (OR1 _X , OR1 _Y ) to A horizontal component Or_C _{X of the} first summed value is obtained, and a plurality of element values O _Y [1] to O _Y [N] of the vertical component are added to obtain a vertical component Or_C _{Y of the} first summed value. Wherein, the horizontal component Or_C _X and the vertical component Or_C _{Y of the} first total value are respectively obtained according to the following equations (10-1) and (10-2):

Afterwards, the logic circuit 220 may further obtain the horizontal component Gain1 _{X of the} first gain value according to the horizontal component Or_C _X of the first summed value, and obtain the vertical component Gain1 of the first gain value according to the vertical component Or_C _Y of the first summed value. _Y. Wherein the first gain, the greater the level of the first sum value of the corresponding component Or_C _X values of the horizontal component of the first gain Gain1 _X, the greater the vertical component sum value of the first Or_C _Y corresponding The vertical component Gain1 _{Y of} the value is larger. In a preferred embodiment, the manner in which the horizontal component Gain1 _X and the vertical component Gain1 _Y of the first gain value are obtained can be obtained by the first gain curve C1 of FIG. In the process of obtaining the horizontal component Gain1 _X of the first gain value, the logic circuit 220 regards the horizontal axis and the vertical axis of FIG. 9 as the horizontal component Or_C _X of the first summed value and the horizontal component Gain1 _{X of the} first gain value, respectively. And obtaining, according to the horizontal component Or_C _{X of} the first total value, the horizontal component Gain1 _{X of the} corresponding first gain value in the first gain curve C1. Similarly, in the process of obtaining the vertical component Gain1 _{Y of the} first gain value, the logic circuit 220 regards the horizontal axis and the vertical axis of FIG. 9 as the vertical component Or_C _{Y of the} first summed value and the vertical component of the first gain value, respectively. Gain1 _Y , and then obtain the vertical component Gain1 _{Y of the} corresponding first gain value from the first gain curve C1 according to the vertical component Or_C _Y of the first summed value.

Similarly, the logic circuit 220 adds a plurality of element values A _X [1] to A _X [M] of the horizontal component of the relative movement characteristic data (AND1 _X , AND1 _Y ) to obtain the horizontal component And_C of the second total value. _X , and add a plurality of element values A _Y [1] to A _Y [N] of the total vertical component to obtain the vertical component And_C _{Y of the} second total value. Wherein, the horizontal component And_C _X and the vertical component And_C _{Y of the} second total value can be obtained according to the following equations (11-1) and (11-2):

Next, the logic circuit 220 obtains the horizontal component Gain2 _{X of the} second gain value according to the horizontal component And_C _X of the second summed value, and obtains the vertical component Gain2 _{Y of the} second gain value according to the vertical component And_C _Y of the second summed value. . The larger the horizontal component Gain2 _X of the second gain value corresponding to the horizontal component And_C _X of the second total added value is preferably smaller, and the larger the second component of the second sum value, the second component corresponding to the vertical component And_C _Y The vertical component Gain2 _{Y of the} gain value is preferably smaller. In a preferred embodiment, the manner in which the horizontal component Gain2 _X and the vertical component Gain2 _Y of the second gain value are obtained can be obtained by the second gain curve C2 of FIG. In the process of obtaining the horizontal component Gain2 _{X of the} second gain value, the logic circuit 220 regards the horizontal axis and the vertical axis of FIG. 10 as the horizontal component And_C _X of the second total value and the horizontal component Gain2 _{X of the} second gain value, respectively. The horizontal component Gain2 _{X of the} corresponding second gain value is obtained in the second gain curve C2 according to the horizontal component And_C _{X of} the second total value. Similarly, in the process of obtaining the vertical component Gain2 _{Y of the} second gain value, the logic circuit 220 regards the horizontal axis and the vertical axis of FIG. 10 as the vertical component And_C _Y of the second total value and the vertical component of the second gain value, respectively. Gain2 _Y , according to the vertical component And_C _{Y of} the second sum total value, obtains the vertical component Gain2 _{Y of the} corresponding second gain value in the second gain curve C2.

Next, the logic circuit 220 may calculate the gain value G according to the horizontal and vertical components of the first gain value and the horizontal and vertical components Gain1 _X , Gain1 _Y , Gain2 _X , Gain2 _Y of the second gain value. In an embodiment of the invention, the gain value G can be obtained according to the following equation (12-1):

G=1-max(Gain1 _X ×Gain2 _X , Gain1 _Y ×Gain2 _Y ) (12-1)

According to the above formula, if (Gain1 _X × Gain2 _X ) is larger than (Gain1 _Y × Gain2 _Y ), the gain value G is equal to [1-(Gain1 _X × Gain2 _X )]; if (Gain1 _X × Gain2 _X ) is smaller than (Gain1 _Y ×Gain2 _Y ), the gain value G is equal to [1-(Gain1 _Y × Gain2 _Y )]. Among them, Gain1 _X , Gain1 _Y , Gain2 _X , and Gain2 _{Y are} all greater than or equal to 0 and less than or equal to 1, so 0≦G≦1.

In summary, the method described in the above embodiment obtains the gain value by calculating the first cumulative relative movement characteristic data and comparing the relative movement characteristic data. Wherein, the first cumulative relative movement characteristic data is obtained by judging the degree of overall difference between the narrow-area vector and the wide-area vector in each picture, and comparing the relative movement characteristic data by judging the latest picture and the rest The degree of difference between the previous pictures was obtained. Therefore, the obtained gain value is determined according to the degree of difference between the narrow field vector and the wide area vector and the degree of difference between the latest picture and the remaining previous pictures. Thereby, the motion depth of the recent picture adjusted according to the gain value can truly reflect the shooting situation, and avoid or improve the depth of field inversion phenomenon.

Although the present invention has been disclosed in the above embodiments, it is not intended to limit the invention, and any one of ordinary skill in the art can make some modifications and refinements without departing from the spirit and scope of the invention. The scope of the invention is defined by the scope of the appended claims.

100. . . image

110. . . background

120. . . Window moving object

200. . . Image processing circuit

210. . . Receiving 埠

220. . . Logic circuit

230. . . Buffer memory

410. . . Narrow domain unit

IMG1, IMG2. . . image

M0～M9. . . Picture

T ₀ to T ₉ . . . Time point

V _(0,1,1) ~V _(0,M,N) . . . Narrow domain motion vector

X. . . horizontal direction

Y. . . Vertical direction

V _G . . . Wide area motion vector

G _X . . . The component of the wide-area motion vector V _G in the horizontal direction X

G _Y . . . The component of the wide-area motion vector V _G in the vertical direction Y

Δ _(0,1,1) ~Δ _(0,M,N) , [Δ _X(0,1,1) ,Δ _Y(0,1,1) ]~[Δ _X(0,M,N) , Δ _Y(0,M,N)) ]. . . Relative motion vector

H[0]~H[P]. . . Relative mobility data

H _X [0] ~ H _X [P]. . . Horizontal component

H _Y [0] ~ H _Y [P]. . . Vertical component

H[0,1]~H[P,Q], H _X [0,1]～H _X [P,M], H _Y [0,1]～H _Y [P,N]. . . Element value

OR1. . . First cumulative relative movement characteristic data

OR1 _X . . . The horizontal component of the first cumulative relative motion characteristic data

OR1 _Y . . . The vertical component of the first cumulative relative motion characteristic data

O[1]~O[Q], O _X [1]～O _X [M], O _Y [1]～O _Y [N]. . . Element value

OR2. . . Second cumulative relative movement characteristic data

OR2 _X . . . The horizontal component of the second cumulative relative motion characteristic data

OR2 _Y . . . The vertical component of the second cumulative relative motion characteristic data

U[1]~U[Q], U _X [1]～U _X [M], U _Y [1]～U _Y [N]. . . Element value

AND1. . . Comparison of relative movement characteristics

AND1 _X . . . Comparing the horizontal component of relative mobility data

AND1 _Y . . . Compare the vertical component of relative motion characteristics data

A[1]~A[Q], A _X [1]～A _X [M], A _Y [1]～A _Y [N]. . . Element value

A _(0,1,1) ~A _(0,M,N) , [A _X(0,1,1) , A _Y(0,1,1) ]~[A _X(0,M,N) , A _Y(0,M,N) ]. . . Comparison result value

C _X [0]. . . Horizontal map motion vector

C _Y [0]. . . Vertical mapping motion vector

S[1]～S[M+N], S _X [1]～S _X [M], S _Y [1]～S _Y [N]. . . Count value

C[1]～C[M+N], C _X [1]～C _X [M], C _Y [1]～C _Y [N]. . . Element value

D _(0,1,1) ~D _(0,M,N) . . . Original movement depth of field

C1. . . First gain curve

C2. . . Second gain curve

Gain1. . . First gain value

Gain2. . . Second gain value

Or_C. . . First total value

And_C. . . Second total value

S201～S205, S211～S216. . . Process step

Figure 1 is a schematic illustration of an image containing a window moving object.

2A is a functional block diagram of an image processing circuit in accordance with an embodiment of the present invention.

2B is a schematic flow chart of a method of adjusting a moving depth of field of an image according to an embodiment.

2C is a detailed flow chart of a method of adjusting the motion depth of field of an image, in accordance with an embodiment.

FIG. 3 illustrates a plurality of pictures of the image of FIG. 2A in time series.

FIG. 4A illustrates a relative motion vector corresponding to each narrow-area unit in the picture M0 according to an embodiment of the present invention.

FIG. 4B illustrates a relative motion vector corresponding to each narrow-area unit in the picture M0 according to another embodiment of the present invention.

FIG. 5A illustrates a relative motion vector corresponding to each narrow-area unit of the picture M0 in an embodiment of the present invention.

FIG. 5B illustrates a relative motion vector corresponding to each narrow-area unit of the picture M0 in another embodiment of the present invention.

FIG. 6A illustrates relative motion characteristic data, first cumulative relative motion characteristic data, second cumulative relative motion characteristic data, and comparative relative motion characteristic data of each picture in an embodiment of the present invention.

6B is a diagram showing relative movement characteristic data, first cumulative relative movement characteristic data, second cumulative relative movement characteristic data, and comparative relative movement characteristic data of each screen in another embodiment of the present invention.

FIG. 7A illustrates a comparison result value corresponding to each narrow-area unit of the screen M0 according to an embodiment of the present invention.

FIG. 7B illustrates comparison result values corresponding to the narrow-area cells of the screen M0 in another embodiment of the present invention.

FIG. 8 illustrates an original motion depth of field corresponding to each narrow-area unit of the screen M0 according to an embodiment of the present invention.

Figure 9 is a schematic diagram of a first gain curve.

Figure 10 is a schematic diagram of a second gain curve.

S211～S216. . . Process step

Claims (18)

A method for adjusting a moving depth of field of an image for two-dimensional to three-dimensional image processing, the method comprising: (i) receiving a plurality of pictures at a plurality of time points, and according to each of the plurality of narrow-field moving vectors and one of the respective pictures a wide-area moving vector, calculating a relative movement characteristic data of each of the pictures; (ii) cumulatively calculating the relative movement characteristic data of the pictures to obtain a first cumulative relative movement characteristic data; (iii) Compulsing the relative movement characteristic data of the remaining pictures except the most recent picture among the pictures to obtain a second cumulative relative movement characteristic data; (iv) comparing the relative movement characteristic data of the recent picture with the second Accumulating the relative movement characteristic data to obtain a comparative relative movement characteristic data; (v) calculating a gain value according to the first cumulative relative movement characteristic data and the comparison relative movement characteristic data; and (vi) adjusting according to the gain value The original motion depth of the recent picture. The method for adjusting the depth of field of an image according to claim 1, wherein the step (i) comprises: (a) calculating a difference between the narrow-area vector and the wide-area vector for each of the pictures, Obtaining a plurality of relative motion vectors; and (b) obtaining the relative motion characteristic data of each of the pictures according to the narrow range motion vectors of the respective pictures and the relative motion vectors. The method for adjusting the depth of field of an image as described in claim 2, wherein the step (b) comprises: (b1) determining whether the absolute value of the narrow range motion vectors is greater than a first threshold; (b2) determining Whether the absolute values of the relative motion vectors are greater than a second threshold value; and (b3) obtaining the relative motion characteristic data of each of the pictures according to the foregoing determination result. The method for adjusting the depth of field of an image according to claim 3, wherein the step (b3) comprises: calculating a plurality of comparison results corresponding to the plurality of narrow-area units in each of the pictures according to the determination result; And mapping the comparison result values along a row/column mapping direction to generate a mapping motion vector, the mapping motion vector representing the relative motion characteristic data. The method for adjusting the depth of field of an image according to claim 4, wherein the step of generating the mapping motion vector comprises: counting the comparison result values along the row/column mapping direction to generate a plurality of respectively corresponding to different a row/column count value; and comparing the count values to a third threshold value to generate a plurality of element values of the map motion vector based on a comparison result of the count values and the third threshold value. The method for adjusting the depth of field of an image as described in claim 1, wherein the steps (i) to (v) are performed according to one to more directions of the screens, respectively. The method for adjusting the depth of field of an image as described in claim 1, wherein each of the relative movement characteristic data in the pictures includes a plurality of element values corresponding to different rows/columns, and the above steps (ii) comprising: performing an OR operation on the element values corresponding to the same row/column in the pictures to obtain the first cumulative relative movement characteristic data. The method for adjusting the depth of field of an image according to claim 7 , wherein the step (iii) comprises: selecting the element values corresponding to the same row/column of the relative movement characteristic data of the remaining screens. Performing an OR operation to obtain the second cumulative relative movement characteristic data. The method for adjusting the depth of field of an image according to claim 7 , wherein the step (iv) comprises: combining the plurality of element values of the relative movement characteristic data with the same line of the second cumulative relative movement characteristic data; The opposite element values of the /column are summed to obtain the relative movement characteristics of the comparison. The method for adjusting the depth of field of an image according to claim 1, wherein the step (v) comprises: obtaining a first gain value according to the first cumulative relative movement characteristic data; and comparing the relative movement characteristic data according to the comparison Obtaining a second gain value; and calculating the gain value according to the first gain value and the second gain value. The method for adjusting the depth of field of an image according to claim 10, wherein the step of obtaining the first gain value comprises: first summing a value of one of a plurality of element values according to the first cumulative relative movement characteristic data Obtaining the first gain value from a first gain curve; and obtaining the second gain value comprises: selecting a second total value of the plurality of element values according to the comparison relative movement characteristic data, from a first The second gain value is obtained in the second gain curve. The method for adjusting the depth of field of an image according to claim 10, wherein each of the first gain value and the second gain value is calculated according to the first and second directions, respectively. The method for adjusting the depth of field of an image according to claim 10, wherein the calculating the gain value comprises: obtaining a product of the first gain value and the second gain value along the first direction; obtaining along the first a product of the first gain value and the second gain value in the two directions; and determining the gain value based on one of the two products. A method for adjusting a moving depth of field of an image for two-dimensional to three-dimensional image processing, the method comprising: receiving a plurality of pictures at a plurality of time points, and calculating a plurality of narrow-area moving vectors and a wide-area movement of each of the pictures a vector; determining a first degree of difference between the plurality of narrow-area motion vectors and the generalized motion vector; determining a second degree of difference between a recent picture and the remaining previous pictures among the plurality of pictures; The first difference degree and the second difference degree are used to calculate a gain value; and the original motion depth of field of the recent picture is adjusted according to the gain value. The method for adjusting the depth of field of an image according to claim 14, wherein the step of determining the first degree of difference comprises: calculating, respectively, a difference between the narrow-area vectors and the wide-area vector for each of the pictures Obtaining a plurality of relative motion vectors; obtaining the relative motion characteristic data of each of the pictures according to the narrow range motion vectors of the respective pictures and the relative motion vectors; and moving the relative movements of the pictures The characteristic data is cumulatively calculated to obtain a first cumulative relative movement characteristic data, and the first cumulative relative movement characteristic data represents the first degree of difference. The method for adjusting the depth of field of an image according to claim 15 , wherein the step of obtaining the relative movement characteristic data of each of the pictures comprises: determining whether an absolute value of the narrow range motion vectors is greater than a first threshold Determining whether the absolute values of the relative motion vectors are greater than a second threshold value; and obtaining the relative motion characteristic data of each of the pictures according to the foregoing determination result. The method for adjusting the depth of field of an image according to claim 15 , wherein the step of determining the second degree of difference comprises: accumulating the relative movement characteristic data of the remaining pictures except the most recent picture among the pictures; Obtaining a second cumulative relative movement characteristic data; and comparing the relative movement characteristic data of the latest picture with the second cumulative relative movement characteristic data to obtain a comparative relative movement characteristic data, the comparative relative movement characteristic data representing the first The difference between the two. The method for adjusting the depth of field of an image as described in claim 14, wherein the gain value is set to be smaller when the first difference degree is larger, and the gain value is smaller when the second difference degree is smaller. The system is set to be smaller.