JPH0562876B2 - - Google Patents

Description

[Detailed description of the invention]

[Industrial Field of Application] The present invention relates to a motion vector smoothing circuit that smoothes a motion vector indicating the movement of a television camera when the entire shooting screen moves due to panning or the like of the television camera. [Prior Art] As one of the transmission methods to narrow the band of high-definition television signals, when the entire screen moves in the same direction, such as when a television camera pans, this image is regarded as a still image. Then, the amount of data is reduced to 1/4 by shifting the sub-sampling position in 4 field periods, and a motion vector indicating the movement of the television camera is transmitted, and the receiving side uses 4 field data to There is a method that performs motion correction by reproducing a still image and reading out a still image from memory by shifting the coordinate axes according to a motion vector. [Problem to be solved by the invention] However, when an incorrect motion vector is determined due to noise etc., if the above-mentioned motion correction is performed using this incorrect motion vector, discontinuity in the time axis direction becomes noticeable. It turns out. Therefore, an object of the present invention is to perform smoothing processing using the strong correlation of motion vectors in the time axis direction, thereby reducing discontinuities in the time axis direction of motion vectors, and to reduce discontinuities in the time axis direction of motion vectors. An object of the present invention is to provide a motion vector smoothing circuit that can improve the visual characteristics of a motion vector. SUMMARY OF THE INVENTION An object of the present invention is to provide a motion vector smoothing circuit that can quickly follow changes in a motion vector and can obtain a sufficient noise removal effect in a steady state. [Means for Solving the Problems] The present invention includes a weight generation section 14 that generates weight coefficients that are sequentially decreased from the time a scene change occurs and is held at a predetermined value, and a weight generation section 14 that generates weight coefficients that are held at predetermined values. According to the weighting coefficient from the weight generation unit 14, Y _K = W _K ×X _K + (1-W _K ) Y _{K -1} (where, The motion vector Y _K is the output motion vector of the k-th field, and W _K is the weighting coefficient of the k-th field.
This is a motion vector smoothing circuit characterized by comprising: [Operation] Using the strong correlation in the time axis direction, means 4, 5,
By performing smoothing processing in steps 6 and 8, discontinuities between detected motion vectors are eliminated. In addition, by gradually decreasing the weighting coefficient for this smoothing process from the reset state and holding it at a predetermined value, it is possible to follow changes in the motion vector well and obtain a sufficient noise removal effect in a steady state. . [Example] Hereinafter, an example of the present invention will be described with reference to the drawings. In FIG. 1, 1 indicates an input terminal to which data of a motion vector determined for each field is supplied. This motion vector is generated by the configuration shown in FIG. 2, for example. In FIG. 2, numeral 21 indicates an input terminal for a digital television signal taken by a television camera and digitized. This digital television signal is supplied to a representative point extraction circuit 22 and a block data extraction circuit 23, and the pixel data of the representative point obtained as an output of the representative point extraction circuit 22 is supplied to a representative point memory 23. The pixel data of the representative point read out from the representative point memory 23 is that of the previous frame. The block data extraction circuit 24 extracts pixel data for each block included in the current frame and supplies it to the subtraction circuit 25. With this subtraction circuit 25,
The difference between each pixel in the block and the representative point of the previous frame, ie, frame difference data, is determined. This frame difference data is supplied to the conversion circuit 26,
Converted to absolute value. The absolute value of the frame difference data from the conversion circuit 26 is supplied to an integration circuit 27, and the integrated value of the frame difference data is supplied to a switch circuit 28. As shown in FIG.
It is obtained for each 1B, 41C, and 41D. The integrated value of the frame difference data from the integrating circuit 27 is stored in the graph memories 29 and 3 for each area by the switch circuit 28.
It is divided into 0, 31, and 32. The graph memory 29 is a memory with a capacity of one block that stores the integrated value of frame difference data (frame difference integral data) of each block included in the area 41A. Each of the graph memories 30, 31, and 32 similarly stores frame difference integral data of areas 41B, 41C, and 41D. The frame difference integral data stored in each of these graph memories 29 to 32 is supplied to minimum value detection circuits 33, 34, 35, and 36, respectively. The minimum value detection circuits 33 to 36 detect the position of the minimum value in the frame difference integral data of the corresponding graph memories 29 to 32, ie, the motion vector of each of the regions 41A to 41D. Motion vectors from each of these minimum value detection circuits 33 to 36 are supplied to a selection circuit 37. The selection circuit 37 generates a motion vector for one field from the motion vectors as shown in FIG. 3B obtained for the four regions 41A to 41D. This selection circuit 37 has, for example, a majority logic circuit configuration or a configuration that calculates a weighted average. The motion vector in units of one field from the selection circuit 37 is supplied to a smoothing circuit 38, noise is removed, and the motion vector is outputted to an output terminal 39. As shown in Figure 3B, if the four motion vectors are the same,
This is an ideal case, but in reality, the motion vector in field units from the selection circuit 37 is affected by noise, image patterns (differences between low-frequency and high-frequency patterns), and moving objects. Some of them may include motion vectors with low accuracy. A smoothing circuit 38 is provided to reduce the discontinuity in the time axis direction that occurs due to this. FIG. 1 shows the configuration of an embodiment of the present invention applied as the smoothing circuit 38 described above. Since the motion vector is position data, it has an X component and a Y component on coordinates, and the configuration shown in FIG. 1 is applied to each component. In FIG. 1, a field-by-field motion vector from an input terminal indicated by 1 is supplied to a subtraction circuit 4, a subtraction circuit 7, and an AND gate 11, and the output of the subtraction circuit 4 is supplied to a multiplication circuit 5. This multiplication circuit 5 is supplied with weighting coefficients from a weight generation section 14 . The output of the multiplier circuit 5 is supplied to an adder circuit 6, and the output of the adder circuit 6 is supplied to a subtracter circuit 7, a register 8 that causes a delay of one field, and an AND gate 12. The subtraction circuit 4, the multiplication circuit 5, the addition circuit 6, and the register 8 constitute a smoothing filter. That is, if the k-th value of the input motion vector is _XK , and the weighting coefficient from the weight generator 14 at that time is _WK , the output data _YK output from the adder circuit 6 is expressed by the following equation. Become something. Y _K = (X _K − Y _K-1 ) W _K + Y _K-1 = W _K XX _K + (1− W _K ) Y _K-1 As is clear from the above equation, the motion vector processed by the smoothing filter Y _K is the previous field's motion vector Y _K-1 multiplied by a coefficient of (1-W _K ), and the current field's motion vector
It is the sum (ie, weighted average value) of Y _K multiplied by the weighting coefficient W _K. Although not shown, the weight generation unit 14 includes a counter and a ROM that stores a plurality of weighting coefficients.
The counter is supplied with a field clock from terminal 2. This field clock changes the value of the weighting coefficient using the field as the minimum cycle. At the same time, the scene change signal from the terminal 3 is supplied to the weight generator 14 as its reset signal. This scene change signal is used to reset the weight generator 14 and return the weight coefficients to their initial values when the television camera performs a scene change. If the count value of the counter of the weight generator 14 is set to 0 at the time of reset, the value of the weight coefficient is gradually decreased from 1 to 1/8 as shown in the table below every time the field clock advances. will be held.

【table】

Claims (1)

[Claims] 1. A weight generation unit that generates a weighting coefficient that gradually decreases from the time when a scene change occurs and is held at a predetermined value, and a motion vector obtained for each field is input, and the above-mentioned According to the weight coefficient from the weight generator, Y= _K = W× _K ×X _K + (1+W _K ) Y _K-1 (where, X _K is the input motion vector of the k-th field, and Y _K is 1. A motion vector smoothing circuit comprising means for performing a weighted average represented by: W _K , the output motion vector of the field, is a weighting coefficient of the k-th field.